ICML 2025 Papers

Layout:

mini compact topic detail

Secant Line Search for Frank-Wolfe Algorithms

AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML

Cradle: Empowering Foundation Agents towards General Computer Control

EGPlace: An Efficient Macro Placement Method via Evolutionary Search with Greedy Repositioning Guided Mutation

EnIGMA: Interactive Tools Substantially Assist LM Agents in Finding Security Vulnerabilities

Federated Oriented Learning: A Practical One-Shot Personalized Federated Learning Framework

Invariant Deep Uplift Modeling for Incentive Assignment in Online Marketing via Probability of Necessity and Sufficiency

Making Hard Problems Easier with Custom Data Distributions and Loss Regularization: A Case Study in Modular Arithmetic

OrcaLoca: An LLM Agent Framework for Software Issue Localization

POQD: Performance-Oriented Query Decomposer for Multi-vector retrieval

RepoAudit: An Autonomous LLM-Agent for Repository-Level Code Auditing

B-score: Detecting biases in large language models using response history

Towards Practical Defect-Focused Automated Code Review

Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functional

An All-Atom Generative Model for Designing Protein Complexes

Capturing Temporal Dynamics in Large-Scale Canopy Tree Height Estimation

CoastalBench: A Decade-Long High-Resolution Dataset to Emulate Complex Coastal Processes

Objective drives the consistency of representational similarity across datasets

Code-Generated Graph Representations Using Multiple LLM Agents for Material Properties Prediction

DiffMS: Diffusion Generation of Molecules Conditioned on Mass Spectra

Sparsing Law: Towards Large Language Models with Greater Activation Sparsity

Diffusion on Language Model Encodings for Protein Sequence Generation

ELoRA: Low-Rank Adaptation for Equivariant GNNs

From Uncertain to Safe: Conformal Adaptation of Diffusion Models for Safe PDE Control

GenMol: A Drug Discovery Generalist with Discrete Diffusion

KinDEL: DNA-Encoded Library Dataset for Kinase Inhibitors

Adversarial Perturbations Are Formed by Iteratively Learning Linear Combinations of the Right Singular Vectors of the Adversarial Jacobian

LDMol: A Text-to-Molecule Diffusion Model with Structurally Informative Latent Space Surpasses AR Models

Learning Condensed Graph via Differentiable Atom Mapping for Reaction Yield Prediction

Identifying Neural Dynamics Using Interventional State Space Models

LLM-Augmented Chemical Synthesis and Design Decision Programs

Maximum Update Parametrization and Zero-Shot Hyperparameter Transfer for Fourier Neural Operators

Hierarchical Graph Tokenization for Molecule-Language Alignment

Multi-Timescale Dynamics Model Bayesian Optimization for Plasma Stabilization in Tokamaks

Reducing Variance of Stochastic Optimization for Approximating Nash Equilibria in Normal-Form Games

Neural Graph Matching Improves Retrieval Augmented Generation in Molecular Machine Learning

Neural Interpretable PDEs: Harmonizing Fourier Insights with Attention for Scalable and Interpretable Physics Discovery

Piloting Structure-Based Drug Design via Modality-Specific Optimal Schedule

PPDiff: Diffusing in Hybrid Sequence-Structure Space for Protein-Protein Complex Design

Pretraining Generative Flow Networks with Inexpensive Rewards for Molecular Graph Generation

ReQFlow: Rectified Quaternion Flow for Efficient and High-Quality Protein Backbone Generation

Rethink GraphODE Generalization within Coupled Dynamical System

Rhomboid Tiling for Geometric Graph Deep Learning

Riemann Tensor Neural Networks: Learning Conservative Systems with Physics-Constrained Networks

RISE: Radius of Influence based Subgraph Extraction for 3D Molecular Graph Explanation

SCENT: Robust Spatiotemporal Learning for Continuous Scientific Data via Scalable Conditioned Neural Fields

Sidechain conditioning and modeling for full-atom protein sequence design with FAMPNN

Average Sensitivity of Hierarchical $k$-Median Clustering

Skip the Equations: Learning Behavior of Personalized Dynamical Systems Directly From Data

Steering Protein Language Models

Tensor-Var: Efficient Four-Dimensional Variational Data Assimilation

Topology-aware Neural Flux Prediction Guided by Physics

Unifying Knowledge from Diverse Datasets to Enhance Spatial-Temporal Modeling: A Granularity-Adaptive Geographical Embedding Approach

Zebra: In-Context Generative Pretraining for Solving Parametric PDEs

Zero-Shot Cyclic Peptide Design via Composable Geometric Constraints

Persistent Topological Features in Large Language Models

AGAV-Rater: Adapting Large Multimodal Model for AI-Generated Audio-Visual Quality Assessment

Asymmetric Decision-Making in Online Knowledge Distillation: Unifying Consensus and Divergence

Better to Teach than to Give: Domain Generalized Semantic Segmentation via Agent Queries with Diffusion Model Guidance

Beyond Entropy: Region Confidence Proxy for Wild Test-Time Adaptation

Test-Time Adaptation with Binary Feedback

Ca2-VDM: Efficient Autoregressive Video Diffusion Model with Causal Generation and Cache Sharing

Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attention

Complex Wavelet Mutual Information Loss: A Multi-Scale Loss Function for Semantic Segmentation

ConText: Driving In-context Learning for Text Removal and Segmentation

Contrastive Visual Data Augmentation

Control and Realism: Best of Both Worlds in Layout-to-Image without Training

DocVXQA: Context-Aware Visual Explanations for Document Question Answering

DyPolySeg: Taylor Series-Inspired Dynamic Polynomial Fitting Network for Few-shot Point Cloud Semantic Segmentation

Elucidating the design space of language models for image generation

Ex-VAD: Explainable Fine-grained Video Anomaly Detection Based on Visual-Language Models

The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models Via Visual Information Steering

Exploring Vision Semantic Prompt for Efficient Point Cloud Understanding

Wyckoff Transformer: Generation of Symmetric Crystals

Few-Shot Learner Generalizes Across AI-Generated Image Detection

Flex3D: Feed-Forward 3D Generation with Flexible Reconstruction Model and Input View Curation

An Effective and Secure Federated Multi-View Clustering Method with Information-Theoretic Perspective

FlowDrag: 3D-aware Drag-based Image Editing with Mesh-guided Deformation Vector Flow Fields

FourierMamba: Fourier Learning Integration with State Space Models for Image Deraining

From Thousands to Billions: 3D Visual Language Grounding via Render-Supervised Distillation from 2D VLMs

Gamma Distribution PCA-Enhanced Feature Learning for Angle-Robust SAR Target Recognition

Redundancy Undermines the Trustworthiness of Self-Interpretable GNNs

Can Compressed LLMs Truly Act? An Empirical Evaluation of Agentic Capabilities in LLM Compression

Generalizable Multi-Camera 3D Object Detection from a Single Source via Fourier Cross-View Learning

Generative Point Cloud Registration

Geometric Feature Embedding for Effective 3D Few-Shot Class Incremental Learning

GeoPixel: Pixel Grounding Large Multimodal Model in Remote Sensing

Learning Soft Sparse Shapes for Efficient Time-Series Classification

Learning Imbalanced Data with Beneficial Label Noise

GoIRL: Graph-Oriented Inverse Reinforcement Learning for Multimodal Trajectory Prediction

HetSSNet: Spatial-Spectral Heterogeneous Graph Learning Network for Panchromatic and Multispectral Images Fusion

Hierarchical Masked Autoregressive Models with Low-Resolution Token Pivots

History-Guided Video Diffusion

IntLoRA: Integral Low-rank Adaptation of Quantized Diffusion Models

L-Diffusion: Laplace Diffusion for Efficient Pathology Image Segmentation

Learning Adaptive Lighting via Channel-Aware Guidance

LieRE: Lie Rotational Positional Encodings

LightningDrag: Lightning Fast and Accurate Drag-based Image Editing Emerging from Videos

LLaVA-ReID: Selective Multi-image Questioner for Interactive Person Re-Identification

LOCATE 3D: Real-World Object Localization via Self-Supervised Learning in 3D

MiraGe: Editable 2D Images using Gaussian Splatting

MoMa: Modulating Mamba for Adapting Image Foundation Models to Video Recognition

Multi-Modal Object Re-identification via Sparse Mixture-of-Experts

An Efficient Search-and-Score Algorithm for Ancestral Graphs using Multivariate Information Scores for Complex Non-linear and Categorical Data

One Diffusion Step to Real-World Super-Resolution via Flow Trajectory Distillation

PiD: Generalized AI-Generated Images Detection with Pixelwise Decomposition Residuals

Privacy-Shielded Image Compression: Defending Against Exploitation from Vision-Language Pretrained Models

Probabilistic Interactive 3D Segmentation with Hierarchical Neural Processes

ReferSplat: Referring Segmentation in 3D Gaussian Splatting

ReFocus: Visual Editing as a Chain of Thought for Structured Image Understanding

ReFrame: Layer Caching for Accelerated Inference in Real-Time Rendering

Scaling Large Motion Models with Million-Level Human Motions

Scaling Laws in Patchification: An Image Is Worth 50,176 Tokens And More

Separating Knowledge and Perception with Procedural Data

Steerable Transformers for Volumetric Data

Tackling View-Dependent Semantics in 3D Language Gaussian Splatting

Taming Rectified Flow for Inversion and Editing

Temporal Misalignment in ANN-SNN Conversion and its Mitigation via Probabilistic Spiking Neurons

Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion

BaxBench: Can LLMs Generate Correct and Secure Backends?

Rethinking Point Cloud Data Augmentation: Topologically Consistent Deformation

UniMC: Taming Diffusion Transformer for Unified Keypoint-Guided Multi-Class Image Generation

Unlocking the Capabilities of Large Vision-Language Models for Generalizable and Explainable Deepfake Detection

Visual Graph Arena: Evaluating Visual Conceptualization of Vision and Multimodal Large Language Models

When Every Millisecond Counts: Real-Time Anomaly Detection via the Multimodal Asynchronous Hybrid Network

When Model Knowledge meets Diffusion Model: Diffusion-assisted Data-free Image Synthesis with Alignment of Domain and Class

Think Twice, Act Once: A Co-Evolution Framework of LLM and RL for Large-Scale Decision Making

Analytical Lyapunov Function Discovery: An RL-based Generative Approach

AUTOCIRCUIT-RL: Reinforcement Learning-Driven LLM for Automated Circuit Topology Generation

Concurrent Reinforcement Learning with Aggregated States via Randomized Least Squares Value Iteration

SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding

Nearly Optimal Sample Complexity for Learning with Label Proportions

Bridging Layout and RTL: Knowledge Distillation based Timing Prediction

Bridging Protein Sequences and Microscopy Images with Unified Diffusion Models

CodeSync: Synchronizing Large Language Models with Dynamic Code Evolution at Scale

Dataflow-Guided Neuro-Symbolic Language Models for Type Inference

DEFAME: Dynamic Evidence-based FAct-checking with Multimodal Experts

EditLord: Learning Code Transformation Rules for Code Editing

Guided Structural Inference: Leveraging Priors with Soft Gating Mechanisms

HyperIV: Real-time Implied Volatility Smoothing

Rethinking Latent Redundancy in Behavior Cloning: An Information Bottleneck Approach for Robot Manipulation

In-Context Adaptation to Concept Drift for Learned Database Operations

Learning Cascade Ranking as One Network

Machines and Mathematical Mutations: Using GNNs to Characterize Quiver Mutation Classes

MapEval: A Map-Based Evaluation of Geo-Spatial Reasoning in Foundation Models

NEAR: Neural Electromagnetic Array Response

Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?

PatchPilot: A Cost-Efficient Software Engineering Agent with Early Attempts on Formal Verification

SK-VQA: Synthetic Knowledge Generation at Scale for Training Context-Augmented Multimodal LLMs

Supervised Contrastive Learning from Weakly-Labeled Audio Segments for Musical Version Matching

Unbiased Recommender Learning from Implicit Feedback via Weakly Supervised Learning

A Variational Perspective on Generative Protein Fitness Optimization

Aligning Protein Conformation Ensemble Generation with Physical Feedback

All-atom inverse protein folding through discrete flow matching

AutoElicit: Using Large Language Models for Expert Prior Elicitation in Predictive Modelling

Boosting Masked ECG-Text Auto-Encoders as Discriminative Learners

Playmate: Flexible Control of Portrait Animation via 3D-Implicit Space Guided Diffusion

BoxLM: Unifying Structures and Semantics of Medical Concepts for Diagnosis Prediction in Healthcare

SPMC: Self-Purifying Federated Backdoor Defense via Margin Contribution

Causal Invariance-aware Augmentation for Brain Graph Contrastive Learning

Do Multiple Instance Learning Models Transfer?

EmoGrowth: Incremental Multi-label Emotion Decoding with Augmented Emotional Relation Graph

From Mechanistic Interpretability to Mechanistic Biology: Training, Evaluating, and Interpreting Sparse Autoencoders on Protein Language Models

From Token to Rhythm: A Multi-Scale Approach for ECG-Language Pretraining

FlashTP: Fused, Sparsity-Aware Tensor Product for Machine Learning Interatomic Potentials

H-Tuning: Toward Low-Cost and Efficient ECG-based Cardiovascular Disease Detection with Pre-Trained Models

InfoSEM: A Deep Generative Model with Informative Priors for Gene Regulatory Network Inference

MedRAX: Medical Reasoning Agent for Chest X-ray

MMedPO: Aligning Medical Vision-Language Models with Clinical-Aware Multimodal Preference Optimization

P(all-atom) Is Unlocking New Path For Protein Design

PyTDC: A multimodal machine learning training, evaluation, and inference platform for biomedical foundation models

SAFER: A Calibrated Risk-Aware Multimodal Recommendation Model for Dynamic Treatment Regimes

SPACE: Your Genomic Profile Predictor is a Powerful DNA Foundation Model

Staged and Physics-Grounded Learning Framework with Hyperintensity Prior for Pre-Contrast MRI Synthesis

SToFM: a Multi-scale Foundation Model for Spatial Transcriptomics

ViTally Consistent: Scaling Biological Representation Learning for Cell Microscopy

A Variational Framework for Improving Naturalness in Generative Spoken Language Models

Agent Reviewers: Domain-specific Multimodal Agents with Shared Memory for Paper Review

Delta Decompression for MoE-based LLMs Compression

Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent

Fleet of Agents: Coordinated Problem Solving with Large Language Models

Freeze-Omni: A Smart and Low Latency Speech-to-speech Dialogue Model with Frozen LLM

MindAligner: Explicit Brain Functional Alignment for Cross-Subject Visual Decoding from Limited fMRI Data

Grammar-Forced Translation of Natural Language to Temporal Logic using LLMs

Learning from others' mistakes: Finetuning machine translation models with span-level error annotations

Long-Form Speech Generation with Spoken Language Models

OWLS: Scaling Laws for Multilingual Speech Recognition and Translation Models

SING: Spatial Context in Large Language Model for Next-Gen Wearables

Explainable Concept Generation through Vision-Language Preference Learning for Understanding Neural Networks' Internal Representations

Sortformer: A Novel Approach for Permutation-Resolved Speaker Supervision in Speech-to-Text Systems

Synthesizing Privacy-Preserving Text Data via Finetuning *without* Finetuning Billion-Scale LLMs

Can Biologically Plausible Temporal Credit Assignment Rules Match BPTT for Neural Similarity? E-prop as an Example

Differential Coding for Training-Free ANN-to-SNN Conversion

EEG-Language Pretraining for Highly Label-Efficient Clinical Phenotyping

Efficient ANN-SNN Conversion with Error Compensation Learning

Feature Learning beyond the Lazy-Rich Dichotomy: Insights from Representational Geometry

Hgformer: Hyperbolic Graph Transformer for Collaborative Filtering

Inverse Reinforcement Learning with Switching Rewards and History Dependency for Characterizing Animal Behaviors

Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse

Neural Encoding and Decoding at Scale

SpikeVideoFormer: An Efficient Spike-Driven Video Transformer with Hamming Attention and $\mathcal{O}(T)$ Complexity

Testing the Limits of Fine-Tuning for Improving Visual Cognition in Vision Language Models

TTFSFormer: A TTFS-based Lossless Conversion of Spiking Transformer

Unraveling the Interplay between Carryover Effects and Reward Autocorrelations in Switchback Experiments

Learning-Augmented Algorithms for MTS with Bandit Access to Multiple Predictors

A Physics-Informed Machine Learning Framework for Safe and Optimal Control of Autonomous Systems

ABNet: Adaptive explicit-Barrier Net for Safe and Scalable Robot Learning

DexScale: Automating Data Scaling for Sim2Real Generalizable Robot Control

DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning

Efficient Robotic Policy Learning via Latent Space Backward Planning

Empowering World Models with Reflection for Embodied Video Prediction

Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models

Hierarchical Planning for Complex Tasks with Knowledge Graph-RAG and Symbolic Verification

Latent Diffusion Planning for Imitation Learning

Learning Efficient Robotic Garment Manipulation with Standardization

OTTER: A Vision-Language-Action Model with Text-Aware Visual Feature Extraction

EraseAnything: Enabling Concept Erasure in Rectified Flow Transformers

SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation

Three-Dimensional Trajectory Prediction with 3DMoTraj Dataset

Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling

Towards Learning to Complete Anything in Lidar

WOMD-Reasoning: A Large-Scale Dataset for Interaction Reasoning in Driving

Cross-City Latent Space Alignment for Consistency Region Embedding

Flow Matching for Denoised Social Recommendation

BRIDGE: Bootstrapping Text to Control Time-Series Generation via Multi-Agent Iterative Optimization and Diffusion Modeling

CFPT: Empowering Time Series Forecasting through Cross-Frequency Interaction and Periodic-Aware Timestamp Modeling

Exploring Representations and Interventions in Time Series Foundation Models

FSTLLM: Spatio-Temporal LLM for Few Shot Time Series Forecasting

Latent Variable Estimation in Bayesian Black-Litterman Models

Compositional Risk Minimization

LOB-Bench: Benchmarking Generative AI for Finance - an Application to Limit Order Book Data

ProDiff: Prototype-Guided Diffusion for Minimal Information Trajectory Imputation

TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting

Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems

GS-Bias: Global-Spatial Bias Learner for Single-Image Test-Time Adaptation of Vision-Language Models

How Compositional Generalization and Creativity Improve as Diffusion Models are Trained

The Devil Is in the Details: Tackling Unimodal Spurious Correlations for Generalizable Multimodal Reward Models

How Effective Can Dropout Be in Multiple Instance Learning ?

How to Synthesize Text Data without Model Collapse?

Noise-Guided Predicate Representation Extraction and Diffusion-Enhanced Discretization for Scene Graph Generation

SBGD: Improving Graph Diffusion Generative Model via Stochastic Block Diffusion

Semantic Shift Estimation via Dual-Projection and Classifier Reconstruction for Exemplar-Free Class-Incremental Learning

FairICP: Encouraging Equalized Odds via Inverse Conditional Permutation

SLiM: One-shot Quantization and Sparsity with Low-rank Approximation for LLM Weight Compression

The Importance of Being Lazy: Scaling Limits of Continual Learning

Towards Understanding Fine-Tuning Mechanisms of LLMs via Circuit Analysis

Understanding Mode Connectivity via Parameter Space Symmetry

ZeroFlow: Overcoming Catastrophic Forgetting is Easier than You Think

Accelerating PDE-Constrained Optimization by the Derivative of Neural Operators

Beyond Low-rank Decomposition: A Shortcut Approach for Efficient On-Device Learning

Eigenspectrum Analysis of Neural Networks without Aspect Ratio Bias

From Weight-Based to State-Based Fine-Tuning: Further Memory Reduction on LoRA with Parallel Control

Fundamental Limits of Visual Autoregressive Transformers: Universal Approximation Abilities

High Dynamic Range Novel View Synthesis with Single Exposure

Improved Algorithm for Deep Active Learning under Imbalance via Optimal Separation

LapSum - One Method to Differentiate Them All: Ranking, Sorting and Top-k Selection

Learning from Loss Landscape: Generalizable Mixed-Precision Quantization via Adaptive Sharpness-Aware Gradient Aligning

Learning to Stop: Deep Learning for Mean Field Optimal Stopping

Matryoshka Quantization

Merge-Friendly Post-Training Quantization for Multi-Target Domain Adaptation

MERGE$^3$: Efficient Evolutionary Merging on Consumer-grade GPUs

On Exact Bit-level Reversible Transformers Without Changing Architecture

PEAKS: Selecting Key Training Examples Incrementally via Prediction Error Anchored by Kernel Similarity

Physics-Informed DeepONets for drift-diffusion on metric graphs: simulation and parameter identification

Prompt-based Depth Pruning of Large Language Models

RZ-NAS: Enhancing LLM-guided Neural Architecture Search via Reflective Zero-Cost Strategy

Sassha: Sharpness-aware Adaptive Second-order Optimization with Stable Hessian Approximation

Stream-level Flow Matching with Gaussian Processes

The Sharpness Disparity Principle in Transformers for Accelerating Language Model Pre-Training

Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometries

$\infty$-Video: A Training-Free Approach to Long Video Understanding via Continuous-Time Memory Consolidation

Beyond the Permutation Symmetry of Transformers: The Role of Rotation for Model Fusion

Curvature-aware Graph Attention for PDEs on Manifolds

ESPFormer: Doubly-Stochastic Attention with Expected Sliced Transport Plans

HashAttention: Semantic Sparsity for Faster Inference

In-Context Linear Regression Demystified: Training Dynamics and Mechanistic Interpretability of Multi-Head Softmax Attention

Scalable Approximation Algorithms for $p$-Wasserstein Distance and Its Variants

MVA: Linear Attention with High-order Query-Keys Integration and Multi-level Vocabulary Decomposition

On Expressive Power of Looped Transformers: Theoretical Analysis and Enhancement via Timestep Encoding

RePaViT: Scalable Vision Transformer Acceleration via Structural Reparameterization on Feedforward Network Layers

The underlying structures of self-attention: symmetry, directionality, and emergent dynamics in Transformer training

Concept-Centric Token Interpretation for Vector-Quantized Generative Models

Function-Space Learning Rates

Improving Soft Unification with Knowledge Graph Embedding Methods

An Augmentation-Aware Theory for Self-Supervised Contrastive Learning

Inductive Gradient Adjustment for Spectral Bias in Implicit Neural Representations

Investigating the Overlooked Hessian Structure: From CNNs to LLMs

Learning without Isolation: Pathway Protection for Continual Learning

Lego Sketch: A Scalable Memory-augmented Neural Network for Sketching Data Streams

On Temperature Scaling and Conformal Prediction of Deep Classifiers

One-Shot Heterogeneous Federated Learning with Local Model-Guided Diffusion Models

Outlier Gradient Analysis: Efficiently Identifying Detrimental Training Samples for Deep Learning Models

SAN: Hypothesizing Long-Term Synaptic Development and Neural Engram Mechanism in Scalable Model's Parameter-Efficient Fine-Tuning

Tackling Dimensional Collapse toward Comprehensive Universal Domain Adaptation

ITBench: Evaluating AI Agents across Diverse Real-World IT Automation Tasks

Tool Unlearning for Tool-Augmented LLMs

Data-Juicer Sandbox: A Feedback-Driven Suite for Multimodal Data-Model Co-development

Expert Race: A Flexible Routing Strategy for Scaling Diffusion Transformer with Mixture of Experts

Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning

FG-CLIP: Fine-Grained Visual and Textual Alignment

M+: Extending MemoryLLM with Scalable Long-Term Memory

Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment

MUDDFormer: Breaking Residual Bottlenecks in Transformers via Multiway Dynamic Dense Connections

Orthus: Autoregressive Interleaved Image-Text Generation with Modality-Specific Heads

Primitive Vision: Improving Diagram Understanding in MLLMs

Rethinking the Bias of Foundation Model under Long-tailed Distribution

SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training

Towards Graph Foundation Models: Learning Generalities Across Graphs via Task-Trees

WildChat-50M: A Deep Dive Into the Role of Synthetic Data in Post-Training

MAPLE: Many-Shot Adaptive Pseudo-Labeling for In-Context Learning

Accelerated Diffusion Models via Speculative Sampling

Beyond Task-Specific Reasoning: A Unified Conditional Generative Framework for Abstract Visual Reasoning

CHATS: Combining Human-Aligned Optimization and Test-Time Sampling for Text-to-Image Generation

Compositional Condition Question Answering in Tabular Understanding

Compute Optimal Inference and Provable Amortisation Gap in Sparse Autoencoders

Conditional Diffusion Model with Nonlinear Data Transformation for Time Series Forecasting

Continuous Semi-Implicit Models

Provably Near-Optimal Federated Ensemble Distillation with Negligible Overhead

Diffuse Everything: Multimodal Diffusion Models on Arbitrary State Spaces

Diffusion Adversarial Post-Training for One-Step Video Generation

DiTAR: Diffusion Transformer Autoregressive Modeling for Speech Generation

Does Generation Require Memorization? Creative Diffusion Models using Ambient Diffusion

EasyRef: Omni-Generalized Group Image Reference for Diffusion Models via Multimodal LLM

Efficient Molecular Conformer Generation with SO(3)-Averaged Flow Matching and Reflow

EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling

Fast Video Generation with Sliding Tile Attention

Feynman-Kac Correctors in Diffusion: Annealing, Guidance, and Product of Experts

FlexiClip: Locality-Preserving Free-Form Character Animation

FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching

Benchmarking Quantum Reinforcement Learning

Parrot: Multilingual Visual Instruction Tuning

FrameBridge: Improving Image-to-Video Generation with Bridge Models

Generalized Interpolating Discrete Diffusion

Generative Data Mining with Longtail-Guided Diffusion

Geometry Informed Tokenization of Molecules for Language Model Generation

Highly Compressed Tokenizer Can Generate Without Training

I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models

Improved Discretization Complexity Analysis of Consistency Models: Variance Exploding Forward Process and Decay Discretization Scheme

Inductive Moment Matching

Compelling ReLU Networks to Exhibit Exponentially Many Linear Regions at Initialization and During Training

Inverse Problem Sampling in Latent Space Using Sequential Monte Carlo

Kinetic Langevin Diffusion for Crystalline Materials Generation

Learning Extrapolative Sequence Transformations from Markov Chains

MDDM: Practical Message-Driven Generative Image Steganography Based on Diffusion Models

MissScore: High-Order Score Estimation in the Presence of Missing Data

Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion

On the Guidance of Flow Matching

Overcoming Spurious Solutions in Semi-Dual Neural Optimal Transport: A Smoothing Approach for Learning the Optimal Transport Plan

Privacy Attacks on Image AutoRegressive Models

Rényi Neural Processes

Protein Structure Tokenization: Benchmarking and New Recipe

RIFLEx: A Free Lunch for Length Extrapolation in Video Diffusion Transformers

Decoding Rewards in Competitive Games: Inverse Game Theory with Entropy Regularization

SADA: Stability-guided Adaptive Diffusion Acceleration

TimeDART: A Diffusion Autoregressive Transformer for Self-Supervised Time Series Representation

Scaling Laws for Pre-training Agents and World Models

Score-of-Mixture Training: One-Step Generative Model Training Made Simple via Score Estimation of Mixture Distributions

SketchDNN: Joint Continuous-Discrete Diffusion for CAD Sketch Generation

SongGen: A Single Stage Auto-regressive Transformer for Text-to-Song Generation

Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion

The Diffusion Duality

Towards a Mechanistic Explanation of Diffusion Model Generalization

Towards a Unified Framework of Clustering-based Anomaly Detection

Tractable Transformers for Flexible Conditional Generation

Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions

Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes

Upcycling Text-to-Image Diffusion Models for Multi-Task Capabilities

A General Graph Spectral Wavelet Convolution via Chebyshev Order Decomposition

Analytical Construction on Geometric Architectures: Transitioning from Static to Temporal Link Prediction

Reflection-Bench: Evaluating Epistemic Agency in Large Language Models

AutoGFM: Automated Graph Foundation Model with Adaptive Architecture Customization

Covered Forest: Fine-grained generalization analysis of graph neural networks

Do We Really Need Message Passing in Brain Network Modeling?

ENAHPool: The Edge-Node Attention-based Hierarchical Pooling for Graph Neural Networks

Expressive Power of Graph Neural Networks for (Mixed-Integer) Quadratic Programs

FicGCN: Unveiling the Homomorphic Encryption Efficiency from Irregular Graph Convolutional Networks

G-Adaptivity: optimised graph-based mesh relocation for finite element methods

GCAL: Adapting Graph Models to Evolving Domain Shifts

PoisonedEye: Knowledge Poisoning Attack on Retrieval-Augmented Generation based Large Vision-Language Models

GPEN: Global Position Encoding Network for Enhanced Subgraph Representation Learning

Graph Adaptive Autoregressive Moving Average Models

Graph World Model

How Much Can Transfer? BRIDGE: Bounded Multi-Domain Graph Foundation Model with Generalization Guarantees

Reflection-Window Decoding: Text Generation with Selective Refinement

HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks

Implicit degree bias in the link prediction task

Learn Beneficial Noise as Graph Augmentation

Less is More: Federated Graph Learning with Alleviating Topology Heterogeneity from A Causal Perspective

Open Your Eyes: Vision Enhances Message Passing Neural Networks in Link Prediction

SCENIR: Visual Semantic Clarity through Unsupervised Scene Graph Retrieval

Simple Path Structural Encoding for Graph Transformers

SPHINX: Structural Prediction using Hypergraph Inference Network

Stable Fair Graph Representation Learning with Lipschitz Constraint

Structure Is All You Need: Structural Representation Learning on Hyper-Relational Knowledge Graphs

Test-Time Graph Neural Dataset Search With Generative Projection

Topology-Aware Dynamic Reweighting for Distribution Shifts on Graph

Toward Data-centric Directed Graph Learning: An Entropy-driven Approach

What Makes a Good Feedforward Computational Graph?

WILTing Trees: Interpreting the Distance Between MPNN Embeddings

A Stronger Mixture of Low-Rank Experts for Fine-Tuning Foundation Models

Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies

Adapting While Learning: Grounding LLMs for Scientific Problems with Tool Usage Adaptation

Adaptive Localization of Knowledge Negation for Continual LLM Unlearning

Aligning LLMs by Predicting Preferences from User Writing Samples

AlphaDPO: Adaptive Reward Margin for Direct Preference Optimization

AlphaPO: Reward Shape Matters for LLM Alignment

Differentially Private Boxplots

AlphaVerus: Bootstrapping Formally Verified Code Generation through Self-Improving Translation and Treefinement

Masked Autoencoders Are Effective Tokenizers for Diffusion Models

any4: Learned 4-bit Numeric Representation for LLMs

AnyEdit: Edit Any Knowledge Encoded in Language Models

Audio Flamingo 2: An Audio-Language Model with Long-Audio Understanding and Expert Reasoning Abilities

Automated Benchmark Generation for Repository-Level Coding Tasks

Automated Hypothesis Validation with Agentic Sequential Falsifications

Beyond Zero Initialization: Investigating the Impact of Non-Zero Initialization on LoRA Fine-Tuning Dynamics

Boosting Multi-Domain Fine-Tuning of Large Language Models through Evolving Interactions between Samples

Bounded Rationality for LLMs: Satisficing Alignment at Inference-Time

Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging

Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models

Can Large Language Models Understand Intermediate Representations in Compilers?

Can We Predict Performance of Large Models across Vision-Language Tasks?

CodeIO: Condensing Reasoning Patterns via Code Input-Output Prediction

CodeSteer: Symbolic-Augmented Language Models via Code/Text Guidance

Collapse or Thrive: Perils and Promises of Synthetic Data in a Self-Generating World

Compositional Causal Reasoning Evaluation in Language Models

Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead

Cost-efficient Collaboration between On-device and Cloud Language Models

CPCF: A Cross-Prompt Contrastive Framework for Referring Multimodal Large Language Models

CRANE: Reasoning with constrained LLM generation

DA-KD: Difficulty-Aware Knowledge Distillation for Efficient Large Language Models

Data Mixing Optimization for Supervised Fine-Tuning of Large Language Models

Inverse problems with experiment-guided AlphaFold

Closed-form Solutions: A New Perspective on Solving Differential Equations

PertEval-scFM: Benchmarking Single-Cell Foundation Models for Perturbation Effect Prediction

De-mark: Watermark Removal in Large Language Models

Defending LVLMs Against Vision Attacks Through Partial-Perception Supervision

Demonstration Selection for In-Context Learning via Reinforcement Learning

Discriminative Finetuning of Generative Large Language Models without Reward Models and Human Preference Data

DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMs

Diverging Preferences: When do Annotators Disagree and do Models Know?

Domain2Vec: Vectorizing Datasets to Find the Optimal Data Mixture without Training

DS-VLM: Diffusion Supervision Vision Language Model

Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuning

Emergence and Effectiveness of Task Vectors in In-Context Learning: An Encoder Decoder Perspective

Energy-Based Preference Model Offers Better Offline Alignment than the Bradley-Terry Preference Model

Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition

Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective

EvoPress: Accurate Dynamic Model Compression via Evolutionary Search

Explicit Preference Optimization: No Need for an Implicit Reward Model

FactTest: Factuality Testing in Large Language Models with Finite-Sample and Distribution-Free Guarantees

Federated In-Context Learning: Iterative Refinement for Improved Answer Quality

FedPHA: Federated Prompt Learning for Heterogeneous Client Adaptation

Ferret: Federated Full-Parameter Tuning at Scale for Large Language Models

FlipAttack: Jailbreak LLMs via Flipping

FloE: On-the-Fly MoE Inference on Memory-constrained GPU

WorldSimBench: Towards Video Generation Models as World Simulators

Scalable Generation of Spatial Transcriptomics from Histology Images via Whole-Slide Flow Matching

From Debate to Equilibrium: Belief‑Driven Multi‑Agent LLM Reasoning via Bayesian Nash Equilibrium

From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?

From RAG to Memory: Non-Parametric Continual Learning for Large Language Models

GANQ: GPU-Adaptive Non-Uniform Quantization for Large Language Models

Generalists vs. Specialists: Evaluating LLMs on Highly-Constrained Biophysical Sequence Optimization Tasks

Gravity-Bench-v1: A Benchmark on Gravitational Physics Discovery for Agents

HyperTree Planning: Enhancing LLM Reasoning via Hierarchical Thinking

IBCircuit: Towards Holistic Circuit Discovery with Information Bottleneck

Heterogeneous Sufficient Dimension Reduction and Subspace Clustering

Improving Rationality in the Reasoning Process of Language Models through Self-playing Game

A Machine Learning Approach to Duality in Statistical Physics

Independence Tests for Language Models

Inference-Time Decomposition of Activations (ITDA): A Scalable Approach to Interpreting Large Language Models

Interpreting the Repeated Token Phenomenon in Large Language Models

Invariance Makes LLM Unlearning Resilient Even to Unanticipated Downstream Fine-Tuning

Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient

Large Language Models are Demonstration Pre-Selectors for Themselves

Larger or Smaller Reward Margins to Select Preferences for LLM Alignment?

Learning Dynamics in Continual Pre-Training for Large Language Models

LensLLM: Unveiling Fine-Tuning Dynamics for LLM Selection

Lexico: Extreme KV Cache Compression via Sparse Coding over Universal Dictionaries

LLMs Can Reason Faster Only If We Let Them

Locate-then-edit for Multi-hop Factual Recall under Knowledge Editing

LongRoPE2: Near-Lossless LLM Context Window Scaling

LoRA-Gen: Specializing Large Language Model via Online LoRA Generation

MAGELLAN: Metacognitive predictions of learning progress guide autotelic LLM agents in large goal spaces

MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systems

Mask-Enhanced Autoregressive Prediction: Pay Less Attention to Learn More

Mechanistic Unlearning: Robust Knowledge Unlearning and Editing via Mechanistic Localization

Memorization Sinks: Isolating Memorization during LLM Training

Memory Layers at Scale

MetaAgent: Automatically Constructing Multi-Agent Systems Based on Finite State Machines

Metadata Conditioning Accelerates Language Model Pre-training

Efficient Logit-based Knowledge Distillation of Deep Spiking Neural Networks for Full-Range Timestep Deployment

Mitigating Heterogeneous Token Overfitting in LLM Knowledge Editing

An Online Adaptive Sampling Algorithm for Stochastic Difference-of-convex Optimization with Time-varying Distributions

Mitigating Object Hallucination in Large Vision-Language Models via Image-Grounded Guidance

Mixture of Lookup Experts

MoE-SVD: Structured Mixture-of-Experts LLMs Compression via Singular Value Decomposition

Multi-agent Architecture Search via Agentic Supernet

Multi-Turn Code Generation Through Single-Step Rewards

MxMoE: Mixed-precision Quantization for MoE with Accuracy and Performance Co-Design

Nemotron-CORTEXA: Enhancing LLM Agents for Software Engineering Tasks via Improved Localization and Solution Diversity

Simple and Critical Iterative Denoising: A Recasting of Discrete Diffusion in Graph Generation

NextCoder: Robust Adaptation of Code LMs to Diverse Code Edits

NExtLong: Toward Effective Long-Context Training without Long Documents

Olica: Efficient Structured Pruning of Large Language Models without Retraining

SOLD: Slot Object-Centric Latent Dynamics Models for Relational Manipulation Learning from Pixels

OmniBal: Towards Fast Instruction-Tuning for Vision-Language Models via Omniverse Computation Balance

On the Duality between Gradient Transformations and Adapters

GL-LowPopArt: A Nearly Instance-Wise Minimax-Optimal Estimator for Generalized Low-Rank Trace Regression

On the Robustness of Reward Models for Language Model Alignment

On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention for Long-Context LLM Serving

Pre-training Auto-regressive Robotic Models with 4D Representations

ELITE: Enhanced Language-Image Toxicity Evaluation for Safety

Optimizing Test-Time Compute via Meta Reinforcement Finetuning

Oracle-MoE: Locality-preserving Routing in the Oracle Space for Memory-constrained Large Language Model Inference

Overtrained Language Models Are Harder to Fine-Tune

PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling

Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models

Selective Prompt Anchoring for Code Generation

Efficient Personalized Adaptation for Physiological Signal Foundation Model

Computing Optimal Transport Maps and Wasserstein Barycenters Using Conditional Normalizing Flows

sciLaMA: A Single-Cell Representation Learning Framework to Leverage Prior Knowledge from Large Language Models

PDE-Controller: LLMs for Autoformalization and Reasoning of PDEs

Pivoting Factorization: A Compact Meta Low-Rank Representation of Sparsity for Efficient Inference in Large Language Models

Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks

Predictive Data Selection: The Data That Predicts Is the Data That Teaches

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference

Principled Data Selection for Alignment: The Hidden Risks of Difficult Examples

A Cognac Shot To Forget Bad Memories: Corrective Unlearning for Graph Neural Networks

Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective

QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search

R.I.P.: Better Models by Survival of the Fittest Prompts

Reinforced Lifelong Editing for Language Models

Reliable and Efficient Amortized Model-based Evaluation

ResQ: Mixed-Precision Quantization of Large Language Models with Low-Rank Residuals

Rethinking Chain-of-Thought from the Perspective of Self-Training

Revisiting Chain-of-Thought in Code Generation: Do Language Models Need to Learn Reasoning before Coding?

Reward-Augmented Data Enhances Direct Preference Alignment of LLMs

Reward-Guided Speculative Decoding for Efficient LLM Reasoning

Splitting with Importance-aware Updating for Heterogeneous Federated Learning with Large Language Models

RLTHF: Targeted Human Feedback for LLM Alignment

RuleAdapter: Dynamic Rules for training Safety Reward Models in RLHF

RULEBREAKERS: Challenging LLMs at the Crossroads between Formal Logic and Human-like Reasoning

RWKVQuant: Quantizing the RWKV Family with Proxy Guided Hybrid of Scalar and Vector Quantization

Safe Delta: Consistently Preserving Safety when Fine-Tuning LLMs on Diverse Datasets

SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models

Scaling Inference-Efficient Language Models

Scaling Laws for Floating–Point Quantization Training

Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI

SIMPLEMIX: Frustratingly Simple Mixing of Off- and On-policy Data in Language Model Preference Learning

Scaling Laws for Upcycling Mixture-of-Experts Language Models

Soft Reasoning: Navigating Solution Spaces in Large Language Models through Controlled Embedding Exploration

SpeCache: Speculative Key-Value Caching for Efficient Generation of LLMs

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation

Steer LLM Latents for Hallucination Detection

Streamline Without Sacrifice - Squeeze out Computation Redundancy in LMM

Structure-Guided Large Language Models for Text-to-SQL Generation

T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling

Targeted Low-rank Refinement: Enhancing Sparse Language Models with Precision

Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback

The Berkeley Function Calling Leaderboard (BFCL): From Tool Use to Agentic Evaluation of Large Language Models

The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence

To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models

Enhancing Statistical Validity and Power in Hybrid Controlled Trials: A Randomization Inference Approach with Conformal Selective Borrowing

Towards Cost-Effective Reward Guided Text Generation

Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond

Training Software Engineering Agents and Verifiers with SWE-Gym

TuCo: Measuring the Contribution of Fine-Tuning to Individual Responses of LLMs

UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning

Unbiased Evaluation of Large Language Models from a Causal Perspective

Understanding Bias Reinforcement in LLM Agents Debate

Understanding Multimodal LLMs Under Distribution Shifts: An Information-Theoretic Approach

Understanding the Skill Gap in Recurrent Models: The Role of the Gather-and-Aggregate Mechanism

Towards Attributions of Input Variables in a Coalition

What Makes In-context Learning Effective for Mathematical Reasoning

Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?

Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas

Contradiction Retrieval via Contrastive Learning with Sparsity

CoPINN: Cognitive Physics-Informed Neural Networks

Double-Filter: Efficient Fine-tuning of Pre-trained Vision-Language Models via Patch&Layer Filtering

Integrating Intermediate Layer Optimization and Projected Gradient Descent for Solving Inverse Problems with Diffusion Models

Multi-band Frequency Reconstruction for Neural Psychoacoustic Coding

Hi-Patch: Hierarchical Patch GNN for Irregular Multivariate Time Series

HyperIMTS: Hypergraph Neural Network for Irregular Multivariate Time Series Forecasting

Janus: Dual-Server Multi-Round Secure Aggregation with Verifiability for Federated Learning

Advancing Personalized Learning with Neural Collapse for Long-Tail Challenge

Adversarial Inputs for Linear Algebra Backends

An Adaptive Orthogonal Convolution Scheme for Efficient and Flexible CNN Architectures

Beyond One-Hot Labels: Semantic Mixing for Model Calibration

Boosting Adversarial Robustness with CLAT: Criticality Leveraged Adversarial Training

DiffAdvMAP: Flexible Diffusion-Based Framework for Generating Natural Unrestricted Adversarial Examples

Efficient Robust Conformal Prediction via Lipschitz-Bounded Networks

ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks

Learning State-Based Node Representations from a Class Hierarchy for Fine-Grained Open-Set Detection

LEVIS: Large Exact Verifiable Input Spaces for Neural Networks

Phase and Amplitude-aware Prompting for Enhancing Adversarial Robustness

A Tale of Two Structures: Do LLMs Capture the Fractal Complexity of Language?

Pixel2Feature Attack (P2FA): Rethinking the Perturbed Space to Enhance Adversarial Transferability

Self-supervised Adversarial Purification for Graph Neural Networks

SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds

InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective

GraphCL: Graph-based Clustering for Semi-Supervised Medical Image Segmentation

Long-Short Alignment for Effective Long-Context Modeling in LLMs

ML$^2$-GCL: Manifold Learning Inspired Lightweight Graph Contrastive Learning

Revisiting Diffusion Models: From Generative Pre-training to One-Step Generation

Multi-View Graph Clustering via Node-Guided Contrastive Encoding

A Closer Look at Transformers for Time Series Forecasting: Understanding Why They Work and Where They Struggle

AEQA-NAT : Adaptive End-to-end Quantization Alignment Training Framework for Non-autoregressive Machine Translation

Channel Normalization for Time Series Channel Identification

Continuously Updating Digital Twins using Large Language Models

LightGTS: A Lightweight General Time Series Forecasting Model

LSCD: Lomb--Scargle Conditioned Diffusion for Time series Imputation

SDE Matching: Scalable and Simulation-Free Training of Latent Stochastic Differential Equations

Towards a General Time Series Forecasting Model with Unified Representation and Adaptive Transfer

TransPL: VQ-Code Transition Matrices for Pseudo-Labeling of Time Series Unsupervised Domain Adaptation

Understanding and Improving Length Generalization in Recurrent Models

An analytic theory of creativity in convolutional diffusion models

Constrained Belief Updates Explain Geometric Structures in Transformer Representations

IT$^3$: Idempotent Test-Time Training

Features are fate: a theory of transfer learning in high-dimensional regression

Fishers for Free? Approximating the Fisher Information Matrix by Recycling the Squared Gradient Accumulator

Global Convergence and Rich Feature Learning in $L$-Layer Infinite-Width Neural Networks under $\mu$ Parametrization

Implicit Bias of Gradient Descent for Non-Homogeneous Deep Networks

Minimum Width for Universal Approximation using Squashable Activation Functions

On Path to Multimodal Generalist: General-Level and General-Bench

On the Query Complexity of Verifier-Assisted Language Generation

Optimization for Neural Operators can Benefit from Width

CaDA: Cross-Problem Routing Solver with Constraint-Aware Dual-Attention

Rethinking Benign Overfitting in Two-Layer Neural Networks

Fine-Grained Captioning of Long Videos through Scene Graph Consolidation

Risk and cross validation in ridge regression with correlated samples

Simplicity Bias and Optimization Threshold in Two-Layer ReLU Networks

Sparse Training from Random Initialization: Aligning Lottery Ticket Masks using Weight Symmetry

Spurious Correlations in High Dimensional Regression: The Roles of Regularization, Simplicity Bias and Over-Parameterization

Robust Secure Swap: Responsible Face Swap With Persons of Interest Redaction and Provenance Traceability

Understanding Input Selectivity in Mamba: Impact on Approximation Power, Memorization, and Associative Recall Capacity

A Closer Look at Generalized BH Algorithm for Out-of-Distribution Detection

ABKD: Pursuing a Proper Allocation of the Probability Mass in Knowledge Distillation via $\alpha$-$\beta$-Divergence

Adaptive Estimation and Learning under Temporal Distribution Shift

Bifurcate then Alienate: Incomplete Multi-view Clustering via Coupled Distribution Learning with Linear Overhead

Branches: Efficiently Seeking Optimal Sparse Decision Trees via AO*

Conformal Prediction with Cellwise Outliers: A Detect-then-Impute Approach

Deep Sturm–Liouville: From Sample-Based to 1D Regularization with Learnable Orthogonal Basis Functions

Rethinking Confidence Scores and Thresholds in Pseudolabeling-based SSL

Direct Prediction Set Minimization via Bilevel Conformal Classifier Training

Does One-shot Give the Best Shot? Mitigating Model Inconsistency in One-shot Federated Learning

Efficient Heterogeneity-Aware Federated Active Data Selection

Enhancing Adversarial Robustness with Conformal Prediction: A Framework for Guaranteed Model Reliability

False Coverage Proportion Control for Conformal Prediction

Fast Min-$\epsilon$ Segmented Regression using Constant-Time Segment Merging

Federated Learning for Feature Generalization with Convex Constraints

FedSSI: Rehearsal-Free Continual Federated Learning with Synergistic Synaptic Intelligence

Foundation Model Insights and a Multi-Model Approach for Superior Fine-Grained One-shot Subset Selection

Importance Sampling for Nonlinear Models

Joint Metric Space Embedding by Unbalanced Optimal Transport with Gromov–Wasserstein Marginal Penalization

Learning with Selectively Labeled Data from Multiple Decision-makers

SpargeAttention: Accurate and Training-free Sparse Attention Accelerating Any Model Inference

Locality Preserving Markovian Transition for Instance Retrieval

Multi-Session Budget Optimization for Forward Auction-based Federated Learning

One-Step Generalization Ratio Guided Optimization for Domain Generalization

Parametric Scaling Law of Tuning Bias in Conformal Prediction

Provable Maximum Entropy Manifold Exploration via Diffusion Models

Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges

Solving Satisfiability Modulo Counting Exactly with Probabilistic Circuits

Testing Conditional Mean Independence Using Generative Neural Networks

UltraTWD: Optimizing Ultrametric Trees for Tree-Wasserstein Distance

A Causal World Model Underlying Next Token Prediction: Exploring GPT in a Controlled Environment

A Recipe for Causal Graph Regression: Confounding Effects Revisited

A Sample Efficient Conditional Independence Test in the Presence of Discretization

Non-Asymptotic Length Generalization

Adjustment for Confounding using Pre-Trained Representations

An Improved Clique-Picking Algorithm for Counting Markov Equivalent DAGs via Super Cliques Transfer

Action Dubber: Timing Audible Actions via Inflectional Flow

Vision-Language Models Create Cross-Modal Task Representations

Gradient Flow Provably Learns Robust Classifiers for Orthonormal GMMs

Arrow: Accelerator for Time Series Causal Discovery with Time Weaving

Balancing Interference and Correlation in Spatial Experimental Designs: A Causal Graph Cut Approach

Gradient Aligned Regression via Pairwise Losses

Causal Effect Identification in lvLiNGAM from Higher-Order Cumulants

What Limits Bidirectional Model's Generative Capabilities? A Uni-Bi-Directional Mixture-of-Expert Method For Bidirectional Fine-tuning

Counterfactual Contrastive Learning with Normalizing Flows for Robust Treatment Effect Estimation

Counterfactual Effect Decomposition in Multi-Agent Sequential Decision Making

Data-Driven Selection of Instrumental Variables for Additive Nonlinear, Constant Effects Models

COMRECGC: Global Graph Counterfactual Explainer through Common Recourse

Discovering Latent Causal Graphs from Spatiotemporal Data

Distinguishing Cause from Effect with Causal Velocity Models

Distributionally Robust Policy Learning under Concept Drifts

Doubly Protected Estimation for Survival Outcomes Utilizing External Controls for Randomized Clinical Trials

Enhancing Treatment Effect Estimation via Active Learning: A Counterfactual Covering Perspective

FairPFN: A Tabular Foundation Model for Causal Fairness

Falsification of Unconfoundedness by Testing Independence of Causal Mechanisms

Meta-Black-Box-Optimization through Offline Q-function Learning

Improving the Variance of Differentially Private Randomized Experiments through Clustering

Integer Programming for Generalized Causal Bootstrap Designs

Rethinking Causal Ranking: A Balanced Perspective on Uplift Model Evaluation

Strong and Weak Identifiability of Optimization-based Causal Discovery in Non-linear Additive Noise Models

Variational Counterfactual Intervention Planning to Achieve Target Outcomes

Almost Optimal Fully Dynamic $k$-Center Clustering with Recourse

CTBench: A Library and Benchmark for Certified Training

Automatically Identify and Rectify: Robust Deep Contrastive Multi-view Clustering in Noisy Scenarios

Dynamic Similarity Graph Construction with Kernel Density Estimation

Determining Layer-wise Sparsity for Large Language Models Through a Theoretical Perspective

Efficient Federated Incomplete Multi-View Clustering

Federated Incomplete Multi-view Clustering with Globally Fused Graph Guidance

Not All Tokens Matter All The Time: Dynamic Token Aggregation Towards Efficient Detection Transformers

Federated Node-Level Clustering Network with Cross-Subgraph Link Mending

From Logits to Hierarchies: Hierarchical Clustering made Simple

From Spectrum-free towards Baseline-view-free: Double-track Proximity Driven Multi-view Clustering

Modified K-means Algorithm with Local Optimality Guarantees

PROTOCOL: Partial Optimal Transport-enhanced Contrastive Learning for Imbalanced Multi-view Clustering

Revisiting Instance-Optimal Cluster Recovery in the Labeled Stochastic Block Model

Scalable Attribute-Missing Graph Clustering via Neighborhood Differentiation

AKRMap: Adaptive Kernel Regression for Trustworthy Visualization of Cross-Modal Embeddings

CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective

Guardians of Image Quality: Benchmarking Defenses Against Adversarial Attacks on Image Quality Metrics

How Do Large Language Monkeys Get Their Power (Laws)?

KernelBench: Can LLMs Write Efficient GPU Kernels?

Efficiently Access Diffusion Fisher: Within the Outer Product Span Space

MathConstruct: Challenging LLM Reasoning with Constructive Proofs

MCU: An Evaluation Framework for Open-Ended Game Agents

Measuring Diversity in Synthetic Datasets

Minerva: A Programmable Memory Test Benchmark for Language Models

Overestimation in LLM Evaluation: A Controlled Large-Scale Study on Data Contamination’s Impact on Machine Translation

Prompt-to-Leaderboard: Prompt-Adaptive LLM Evaluations

Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing

Regression for the Mean: Auto-Evaluation and Inference with Few Labels through Post-hoc Regression

Robust ML Auditing using Prior Knowledge

SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?

SyncMind: Measuring Agent Out-of-Sync Recovery in Collaborative Software Engineering

Efficient Bisection Projection to Ensure Neural-Network Solution Feasibility for Optimization over General Set

Flexible, Efficient, and Stable Adversarial Attacks on Machine Unlearning

Fully Dynamic Euclidean Bi-Chromatic Matching in Sublinear Update Time

G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibration

LBI-FL: Low-Bit Integerized Federated Learning with Temporally Dynamic Bit-Width Allocation

Learning Along the Arrow of Time: Hyperbolic Geometry for Backward-Compatible Representation Learning

Learning Dynamics under Environmental Constraints via Measurement-Induced Bundle Structures

Online Differentially Private Conformal Prediction for Uncertainty Quantification

Statistical Hypothesis Testing for Auditing Robustness in Language Models

HiRemate: Hierarchical Approach for Efficient Re-materialization of Neural Networks

QUTE: Quantifying Uncertainty in TinyML models with Early-exit-assisted ensembles for model-monitoring

Joker: Joint Optimization Framework for Lightweight Kernel Machines

Kernel Quantile Embeddings and Associated Probability Metrics

Toward Efficient Kernel-Based Solvers for Nonlinear PDEs

Active feature acquisition via explainability-driven ranking

Active Learning with Selective Time-Step Acquisition for PDEs

Adaptive Data Collection for Robust Learning Across Multiple Distributions

An Optimistic Algorithm for online CMDPS with Anytime Adversarial Constraints

Causal Logistic Bandits with Counterfactual Fairness Constraints

Competing Bandits in Matching Markets via Super Stability

FisherSFT: Data-Efficient Supervised Fine-Tuning of Language Models Using Information Gain

Fusing Reward and Dueling Feedback in Stochastic Bandits

One-Pass Feature Evolvable Learning with Theoretical Guarantees

Sample Efficient Demonstration Selection for In-Context Learning

SEAD: Unsupervised Ensemble of Streaming Anomaly Detectors

Balancing Model Efficiency and Performance: Adaptive Pruner for Long-tailed Data

BDC-CLIP: Brownian Distance Covariance for Adapting CLIP to Action Recognition

Beyond Matryoshka: Revisiting Sparse Coding for Adaptive Representation

Contrastive Learning with Simplicial Convolutional Networks for Short-Text Classification

Deep Unsupervised Hashing via External Guidance

Promoting Ensemble Diversity with Interactive Bayesian Distributional Robustness for Fine-tuning Foundation Models

Discovering Global False Negatives On the Fly for Self-supervised Contrastive Learning

FEAT-KD: Learning Concise Representations for Single and Multi-Target Regression via TabNet Knowledge Distillation

GMAIL: Generative Modality Alignment for generated Image Learning

Improving Multimodal Learning Balance and Sufficiency through Data Remixing

Latent Score-Based Reweighting for Robust Classification on Imbalanced Tabular Data

Learning Single Index Models with Diffusion Priors

Mixed-curvature decision trees and random forests

BEST-Route: Adaptive LLM Routing with Test-Time Optimal Compute

Latent Thought Models with Variational Bayes Inference-Time Computation

DOLPHIN: A Programmable Framework for Scalable Neurosymbolic Learning

On the Importance of Embedding Norms in Self-Supervised Learning

TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration

PlaySlot: Learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning

Revisiting Neural Networks for Few-Shot Learning: A Zero-Cost NAS Perspective

SDMG: Smoothing Your Diffusion Models for Powerful Graph Representation Learning

Self-supervised Masked Graph Autoencoder via Structure-aware Curriculum

TeDS: Joint Learning of Diachronic and Synchronic Perspectives in Quaternion Space for Temporal Knowledge Graph Completion

TINED: GNNs-to-MLPs by Teacher Injection and Dirichlet Energy Distillation

Randomized Dimensionality Reduction for Euclidean Maximization and Diversity Measures

Improved Coresets for Vertical Federated Learning: Regularized Linear and Logistic Regressions

Near-optimal Sketchy Natural Gradients for Physics-Informed Neural Networks

Permutation-Free High-Order Interaction Tests

Geometric Contact Flows: Contactomorphisms for Dynamics and Control

GRAIL: Graph Edit Distance and Node Alignment using LLM-Generated Code

KAN-AD: Time Series Anomaly Detection with Kolmogorov–Arnold Networks

Relational Conformal Prediction for Correlated Time Series

Feedforward Few-shot Species Range Estimation

Residual TPP: A Unified Lightweight Approach for Event Stream Data Analysis

Shifting Time: Time-series Forecasting with Khatri-Rao Neural Operators

VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters

Bipartite Ranking From Multiple Labels: On Loss Versus Label Aggregation

Concentration Distribution Learning from Label Distributions

Cut out and Replay: A Simple yet Versatile Strategy for Multi-Label Online Continual Learning

Enhancing Logits Distillation with Plug&Play Kendall's $\tau$ Ranking Loss

Enhancing Target-unspecific Tasks through a Features Matrix

ENSUR: Equitable and Statistically Unbiased Recommendation

Instance Correlation Graph-based Naive Bayes

Label Distribution Propagation-based Label Completion for Crowdsourcing

LADA: Scalable Label-Specific CLIP Adapter for Continual Learning

Prediction models that learn to avoid missing values

Geometric Representation Condition Improves Equivariant Molecule Generation

Predictive Performance of Deep Quantum Data Re-uploading Models

Pixel-level Certified Explanations via Randomized Smoothing

Preserving AUC Fairness in Learning with Noisy Protected Groups

PhantomWiki: On-Demand Datasets for Reasoning and Retrieval Evaluation

Right Time to Learn: Promoting Generalization via Bio-inspired Spacing Effect in Knowledge Distillation

Tensorized Multi-View Multi-Label Classification via Laplace Tensor Rank

TLLC: Transfer Learning-based Label Completion for Crowdsourcing

Trusted Multi-View Classification with Expert Knowledge Constraints

Be Confident: Uncovering Overfitting in MLLM Multi-Task Tuning

VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data

BECAME: Bayesian Continual Learning with Adaptive Model Merging

BSLoRA: Enhancing the Parameter Efficiency of LoRA with Intra-Layer and Inter-Layer Sharing

CALM: Consensus-Aware Localized Merging for Multi-Task Learning

Does learning the right latent variables necessarily improve in-context learning?

Homophily Enhanced Graph Domain Adaptation

Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers

In-Context Learning and Occam's Razor

Low-Rank Tensor Transitions (LoRT) for Transferable Tensor Regression

NegMerge: Sign-Consensual Weight Merging for Machine Unlearning

Parameter-Efficient Fine-Tuning of State Space Models

Disentangling and Integrating Relational and Sensory Information in Transformer Architectures

Pareto Merging: Multi-Objective Optimization for Preference-Aware Model Merging

Predicting the Susceptibility of Examples to Catastrophic Forgetting

R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts

Representation Surgery in Model Merging with Probabilistic Modeling

FedOne: Query-Efficient Federated Learning for Black-box Discrete Prompt Learning

What Has a Foundation Model Found? Inductive Bias Reveals World Models

$\texttt{I$^2$MoE}$: Interpretable Multimodal Interaction-aware Mixture-of-Experts

Self-Bootstrapping for Versatile Test-Time Adaptation

Socialized Coevolution: Advancing a Better World through Cross-Task Collaboration

Surrogate Prompt Learning: Towards Efficient and Diverse Prompt Learning for Vision-Language Models

Test-time Adaptation on Graphs via Adaptive Subgraph-based Selection and Regularized Prototypes

Test-time Correlation Alignment

All-Purpose Mean Estimation over R: Optimal Sub-Gaussianity with Outlier Robustness and Low Moments Performance

Update Your Transformer to the Latest Release: Re-Basin of Task Vectors

Whoever Started the interference Should End It: Guiding Data-Free Model Merging via Task Vectors

Zero-Shot Adaptation of Parameter-Efficient Fine-Tuning in Diffusion Models

Zero-shot Meta-learning for Tabular Prediction Tasks with Adversarially Pre-trained Transformer

E-LDA: Toward Interpretable LDA Topic Models with Strong Guarantees in Logarithmic Parallel Time

Generalized Category Discovery via Reciprocal Learning and Class-Wise Distribution Regularization

Leveraging Diffusion Model as Pseudo-Anomalous Graph Generator for Graph-Level Anomaly Detection

Score Matching with Missing Data

Self-Discriminative Modeling for Anomalous Graph Detection

Benign Samples Matter! Fine-tuning On Outlier Benign Samples Severely Breaks Safety

Towards an Explainable Comparison and Alignment of Feature Embeddings

BalancEdit: Dynamically Balancing the Generality-Locality Trade-off in Multi-modal Model Editing

Towards the Efficient Inference by Incorporating Automated Computational Phenotypes under Covariate Shift

Universal Neural Optimal Transport

Navigating Conflicting Views: Harnessing Trust for Learning

Aligned Multi Objective Optimization

An in depth look at the Procrustes-Wasserstein distance: properties and barycenters

Armijo Line-search Can Make (Stochastic) Gradient Descent Provably Faster

Convergence of Mean-Field Langevin Stochastic Descent-Ascent for Distributional Minimax Optimization

Learning to Plan & Reason for Evaluation with Thinking-LLM-as-a-Judge

Ensemble Learned Bloom Filters: Two Oracles are Better than One

FedECADO: A Dynamical System Model of Federated Learning

Overcoming Vocabulary Mismatch: Vocabulary-agnostic Teacher Guided Language Modeling

Learning to Generate Projections for Reducing Dimensionality of Heterogeneous Linear Programming Problems

On the Learnability of Distribution Classes with Adaptive Adversaries

Nonlinearly Preconditioned Gradient Methods under Generalized Smoothness

Online Conformal Prediction via Online Optimization

Optimization over Sparse Support-Preserving Sets: Two-Step Projection with Global Optimality Guarantees

CVE-Bench: A Benchmark for AI Agents’ Ability to Exploit Real-World Web Application Vulnerabilities

OptMATH: A Scalable Bidirectional Data Synthesis Framework for Optimization Modeling

PARQ: Piecewise-Affine Regularized Quantization

Revisiting Convergence: Shuffling Complexity Beyond Lipschitz Smoothness

Schwarz–Schur Involution: Lightspeed Differentiable Sparse Linear Solvers

SeedLoRA: A Fusion Approach to Efficient LLM Fine-Tuning

Stochastic Deep Restoration Priors for Imaging Inverse Problems

Triple-Optimistic Learning for Stochastic Contextual Bandits with General Constraints

Geometric Algebra Planes: Convex Implicit Neural Volumes

Improved Last-Iterate Convergence of Shuffling Gradient Methods for Nonsmooth Convex Optimization

Scaffold with Stochastic Gradients: New Analysis with Linear Speed-Up

A Mixed-Curvature based Pre-training Paradigm for Multi-Task Vehicle Routing Solver

PENCIL: Long Thoughts with Short Memory

Breaking Barriers: Combinatorial Algorithms for Non-Monotone Submodular Maximization with Sublinear Adaptivity and $1/e$ Approximation

Hybrid Quantum-Classical Multi-Agent Pathfinding

Multinoulli Extension: A Lossless Yet Effective Probabilistic Framework for Subset Selection over Partition Constraints

Learning Vision and Language Concepts for Controllable Image Generation

SHIELD: Multi-task Multi-distribution Vehicle Routing Solver with Sparsity and Hierarchy

Simple Randomized Rounding for Max-Min Eigenvalue Augmentation

Fast, Accurate Manifold Denoising by Tunneling Riemannian Optimization

Beyond Communication Overhead: A Multilevel Monte Carlo Approach for Mitigating Compression Bias in Distributed Learning

BSemiFL: Semi-supervised Federated Learning via a Bayesian Approach

Causality Inspired Federated Learning for OOD Generalization

Constant Stepsize Local GD for Logistic Regression: Acceleration by Instability

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs

Efficiently Serving Large Multimodal Models Using EPD Disaggregation

FRUGAL: Memory-Efficient Optimization by Reducing State Overhead for Scalable Training

Improving Generalization in Federated Learning with Highly Heterogeneous Data via Momentum-Based Stochastic Controlled Weight Averaging

On the Tension between Byzantine Robustness and No-Attack Accuracy in Distributed Learning

MetaOptimize: A Framework for Optimizing Step Sizes and Other Meta-parameters

Preconditioned Riemannian Gradient Descent Algorithm for Low-Multilinear-Rank Tensor Completion

TSP: A Two-Sided Smoothed Primal-Dual Method for Nonconvex Bilevel Optimization

Clipped SGD Algorithms for Performative Prediction: Tight Bounds for Stochastic Bias and Remedies

Exact risk curves of signSGD in High-Dimensions: quantifying preconditioning and noise-compression effects

Improved Lower Bounds for First-order Stochastic Non-convex Optimization under Markov Sampling

The Global Convergence Time of Stochastic Gradient Descent in Non-Convex Landscapes: Sharp Estimates via Large Deviations

A Likelihood Based Approach to Distribution Regression Using Conditional Deep Generative Models

Variance-Reduced Forward-Reflected-Backward Splitting Methods for Nonmonotone Generalized Equations

Understanding High-Dimensional Bayesian Optimization

Position: Language model developers should report train-test overlap

Position: When Incentives Backfire, Data Stops Being Human

Position: A Theory of Deep Learning Must Include Compositional Sparsity

Position: AI Evaluation Should Learn from How We Test Humans

Position: AI Should Not Be An Imitation Game: Centaur Evaluations

Position: Algebra Unveils Deep Learning - An Invitation to Neuroalgebraic Geometry

Position: An Empirically Grounded Identifiability Theory Will Accelerate Self Supervised Learning Research

Position: Beyond Assistance – Reimagining LLMs as Ethical and Adaptive Co-Creators in Mental Health Care

Position: Causal Machine Learning Requires Rigorous Synthetic Experiments for Broader Adoption

Position: Enough of Scaling LLMs! Lets Focus on Downscaling

Position: Formal Mathematical Reasoning—A New Frontier in AI

Position: Future Research and Challenges Remain Towards AI for Software Engineering

Position: General Intelligence Requires Reward-based Pretraining

Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks

Position: Human Baselines in Model Evaluations Need Rigor and Transparency (With Recommendations & Reporting Checklist)

Position: In-House Evaluation Is Not Enough. Towards Robust Third-Party Evaluation and Flaw Disclosure for General-Purpose AI

Position: Lifetime tuning is incompatible with continual reinforcement learning

Position: LLM Social Simulations Are a Promising Research Method

Learn from Downstream and Be Yourself in Multimodal Large Language Models Fine-Tuning

LLM-Assisted Semantically Diverse Teammate Generation for Efficient Multi-agent Coordination

Position: Medical Large Language Model Benchmarks Should Prioritize Construct Validity

Position: Rethinking LLM Bias Probing Using Lessons from the Social Sciences

Position: Scaling LLM Agents Requires Asymptotic Analysis with LLM Primitives

Position: Stop treating `AGI' as the north-star goal of AI research

Position: Supervised Classifiers Answer the Wrong Questions for OOD Detection

Position: Theory of Mind Benchmarks are Broken for Large Language Models

Position: Truly Self-Improving Agents Require Intrinsic Metacognitive Learning

Core Knowledge Deficits in Multi-Modal Language Models

Position: Trustworthy AI Agents Require the Integration of Large Language Models and Formal Methods

Position: Uncertainty Quantification Needs Reassessment for Large Language Model Agents

Position: We Need Responsible, Application-Driven (RAD) AI Research

Position: The AI Conference Peer Review Crisis Demands Author Feedback and Reviewer Rewards

Position: AI Safety Must Embrace an Antifragile Perspective

Position: Current Model Licensing Practices are Dragging Us into a Quagmire of Legal Noncompliance

Position: Editing Large Language Models Poses Serious Safety Risks

Position: Machine Learning Models Have a Supply Chain Problem

Position: Political Neutrality in AI Is Impossible — But Here Is How to Approximate It

Position: Strong Consumer Protection is an Inalienable Defense for AI Safety in the United States

Position: AI Safety should prioritize the Future of Work

Position: AI's growing due process problem

Position: Build Agent Advocates, Not Platform Agents

Position: Generative AI Regulation Can Learn from Social Media Regulation

Position: Humanity Faces Existential Risk from Gradual Disempowerment

Position: Retrieval-augmented systems can be dangerous medical communicators

Position: Societal Impacts Research Requires Benchmarks for Creative Composition Tasks

Position: The Artificial Intelligence and Machine Learning Community Should Adopt a More Transparent and Regulated Peer Review Process

Position: The Categorization of Race in ML is a Flawed Premise

Position: The Most Expensive Part of an LLM *should* be its Training Data

Position: Deep Learning is Not So Mysterious or Different

Position: Principles of Animal Cognition to Improve LLM Evaluations

Position: Rethinking Explainable Machine Learning as Applied Statistics

Position: We Can’t Understand AI Using our Existing Vocabulary

A Mixture-Based Framework for Guiding Diffusion Models

Adjusting Model Size in Continual Gaussian Processes: How Big is Big Enough?

Can Transformers Learn Full Bayesian Inference in Context?

Decision Making under the Exponential Family: Distributionally Robust Optimisation with Bayesian Ambiguity Sets

KVTuner: Sensitivity-Aware Layer-Wise Mixed-Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference

Fast and Provable Algorithms for Sparse PCA with Improved Sample Complexity

Efficient Optimization with Orthogonality Constraint: a Randomized Riemannian Submanifold Method

Improving the Statistical Efficiency of Cross-Conformal Prediction

Rethinking Aleatoric and Epistemic Uncertainty

TRACE Back from the Future: A Probabilistic Reasoning Approach to Controllable Language Generation

Wrapped Gaussian on the manifold of Symmetric Positive Definite Matrices

A Unified Framework for Entropy Search and Expected Improvement in Bayesian Optimization

Addressing Misspecification in Simulation-based Inference through Data-driven Calibration

Fast Large Language Model Collaborative Decoding via Speculation

BARNN: A Bayesian Autoregressive and Recurrent Neural Network

BILBO: BILevel Bayesian Optimization

Do Bayesian Neural Networks Actually Behave Like Bayesian Models?

Hybrid Batch Normalisation: Resolving the Dilemma of Batch Normalisation in Federated Learning

Linearization Turns Neural Operators into Function-Valued Gaussian Processes

Optimal Sensor Scheduling and Selection for Continuous-Discrete Kalman Filtering with Auxiliary Dynamics

Outsourced Diffusion Sampling: Efficient Posterior Inference in Latent Spaces of Generative Models

Bayesian Inference for Correlated Human Experts and Classifiers

Dequantified Diffusion-Schrödinger Bridge for Density Ratio Estimation

Integration-free Kernels for Equivariant Gaussian Process Modelling

New Bounds for Sparse Variational Gaussian Processes

Return of the Latent Space COWBOYS: Re-thinking the use of VAEs for Bayesian Optimisation of Structured Spaces

Stochastic Poisson Surface Reconstruction with One Solve using Geometric Gaussian Processes

A Generic Family of Graphical Models: Diversity, Efficiency, and Heterogeneity

Scaling Probabilistic Circuits via Monarch Matrices

AutoStep: Locally adaptive involutive MCMC

Conditioning Diffusions Using Malliavin Calculus

Differential Privacy Guarantees of Markov Chain Monte Carlo Algorithms

Importance Corrected Neural JKO Sampling

Non-asymptotic Error Bounds in $\mathcal{W}_2$-Distance with Sqrt(d) Dimension Dependence and First Order Convergence for Langevin Monte Carlo beyond Log-Concavity

The Polynomial Stein Discrepancy for Assessing Moment Convergence

CAT Merging: A Training-Free Approach for Resolving Conflicts in Model Merging

Aggregation of Dependent Expert Distributions in Multimodal Variational Autoencoders

EFDTR: Learnable Elliptical Fourier Descriptor Transformer for Instance Segmentation

Censor Dependent Variational Inference

DriveGPT: Scaling Autoregressive Behavior Models for Driving

Federated Generalised Variational Inference: A Robust Probabilistic Federated Learning Framework

Improving Generalization with Flat Hilbert Bayesian Inference

Neural Guided Diffusion Bridges

Revisiting Unbiased Implicit Variational Inference

A Reduction Framework for Distributionally Robust Reinforcement Learning under Average Reward

Accurate and Efficient World Modeling with Masked Latent Transformers

CommVQ: Commutative Vector Quantization for KV Cache Compression

Action-Dependent Optimality-Preserving Reward Shaping

Adaptive Exploration for Multi-Reward Multi-Policy Evaluation

ADDQ: Adaptive distributional double Q-learning

An Online Learning Approach to Prompt-based Selection of Generative Models and LLMs

Behavioral Exploration: Learning to Explore via In-Context Adaptation

Calibrated Value-Aware Model Learning with Probabilistic Environment Models

Controlling Underestimation Bias in Constrained Reinforcement Learning for Safe Exploration

LLMs can see and hear without any training

Craftium: Bridging Flexibility and Efficiency for Rich 3D Single- and Multi-Agent Environments

Learning to (Learn at Test Time): RNNs with Expressive Hidden States

Learnings from Scaling Visual Tokenizers for Reconstruction and Generation

Enhancing Diversity In Parallel Agents: A Maximum State Entropy Exploration Story

Extreme Value Policy Optimization for Safe Reinforcement Learning

Gap-Dependent Bounds for Federated $Q$-Learning

SelfCite: Self-Supervised Alignment for Context Attribution in Large Language Models

Gradient Boosting Reinforcement Learning

Hyper: Hyperparameter Robust Efficient Exploration in Reinforcement Learning

Improved Off-policy Reinforcement Learning in Biological Sequence Design

Intersectional Fairness in Reinforcement Learning with Large State and Constraint Spaces

KEA: Keeping Exploration Alive by Proactively Coordinating Exploration Strategies

The Missing Alignment Link of In-context Learning on Sequences

Learning Fused State Representations for Control from Multi-View Observations

Unnatural Languages Are Not Bugs but Features for LLMs

Exponential Family Variational Flow Matching for Tabular Data Generation

Controlled Generation with Equivariant Variational Flow Matching

Learning to Reuse Policies in State Evolvable Environments

MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning

Meta-Reinforcement Learning with Adaptation from Human Feedback via Preference-Order-Preserving Task Embedding

Offline-to-Online Reinforcement Learning with Classifier-Free Diffusion Generation

Online Learning in Risk Sensitive constrained MDP

Penalizing Infeasible Actions and Reward Scaling in Reinforcement Learning with Offline Data

Pessimism Principle Can Be Effective: Towards a Framework for Zero-Shot Transfer Reinforcement Learning

Policy Regularization on Globally Accessible States in Cross-Dynamics Reinforcement Learning

Proto Successor Measure: Representing the Behavior Space of an RL Agent

QPRL : Learning Optimal Policies with Quasi-Potential Functions for Asymmetric Traversal

Quantum Algorithms for Finite-horizon Markov Decision Processes

Rank-One Modified Value Iteration

When Bad Data Leads to Good Models

Reward Translation via Reward Machine in Semi-Alignable MDPs

Robust Offline Reinforcement Learning with Linearly Structured $f$-Divergence Regularization

Strategic Planning: A Top-Down Approach to Option Generation

Task-Agnostic Pre-training and Task-Guided Fine-tuning for Versatile Diffusion Planner

Temporal Difference Flows

Wasserstein Policy Optimization

When Maximum Entropy Misleads Policy Optimization

Zero-Shot Offline Imitation Learning via Optimal Transport

Conservative Offline Goal-Conditioned Implicit V-Learning

Constrained Exploitability Descent: An Offline Reinforcement Learning Method for Finding Mixed-Strategy Nash Equilibrium

Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning

Video-Enhanced Offline Reinforcement Learning: A Model-Based Approach

Vintix: Action Model via In-Context Reinforcement Learning

DIME: Diffusion-Based Maximum Entropy Reinforcement Learning

Embedding Safety into RL: A New Take on Trust Region Methods

Enhancing Rating-Based Reinforcement Learning to Effectively Leverage Feedback from Large Vision-Language Models

Hyperspherical Normalization for Scalable Deep Reinforcement Learning

Learning from Suboptimal Data in Continuous Control via Auto-Regressive Soft Q-Network

EAGLES: Towards Effective, Efficient, and Economical Federated Graph Learning via Unified Sparsification

Network Sparsity Unlocks the Scaling Potential of Deep Reinforcement Learning

Optimizing Language Models for Inference Time Objectives using Reinforcement Learning

R*: Efficient Reward Design via Reward Structure Evolution and Parameter Alignment Optimization with Large Language Models

Textural or Textual: How Vision-Language Models Read Text in Images

Reinforcement Learning with Adaptive Reward Modeling for Expensive-to-Evaluate Systems

Return Capping: Sample Efficient CVaR Policy Gradient Optimisation

Robust Autonomy Emerges from Self-Play

Stealing That Free Lunch: Exposing the Limits of Dyna-Style Reinforcement Learning

Task-Aware Virtual Training: Enhancing Generalization in Meta-Reinforcement Learning for Out-of-Distribution Tasks

EvoControl: Multi-Frequency Bi-Level Control for High-Frequency Continuous Control

ZipAR: Parallel Autoregressive Image Generation through Spatial Locality

Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents

Preference Controllable Reinforcement Learning with Advanced Multi-Objective Optimization

Diversifying Robot Locomotion Behaviors with Extrinsic Behavioral Curiosity

IL-SOAR : Imitation Learning with Soft Optimistic Actor cRitic

Robust Reward Alignment via Hypothesis Space Batch Cutting

Ad Hoc Teamwork via Offline Goal-Based Decision Transformers

Ad-Hoc Human-AI Coordination Challenge

Agent-Centric Actor-Critic for Asynchronous Multi-Agent Reinforcement Learning

Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination

DipLLM: Fine-Tuning LLM for Strategic Decision-making in Diplomacy

Graph Diffusion for Robust Multi-Agent Coordination

Learning Imperfect Information Extensive-form Games with Last-iterate Convergence under Bandit Feedback

Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration

Novelty Detection in Reinforcement Learning with World Models

Sleeping Reinforcement Learning

DiLQR: Differentiable Iterative Linear Quadratic Regulator via Implicit Differentiation

LARM: Large Auto-Regressive Model for Long-Horizon Embodied Intelligence

Bayesian Weight Enhancement with Steady-State Adaptation for Test-time Adaptation in Dynamic Environments

Monte Carlo Tree Diffusion for System 2 Planning

Understanding Model Ensemble in Transferable Adversarial Attack

Monte-Carlo Tree Search with Uncertainty Propagation via Optimal Transport

Online Robust Reinforcement Learning Through Monte-Carlo Planning

Rejecting Hallucinated State Targets during Planning

Subgoal-Guided Policy Heuristic Search with Learned Subgoals

A Mathematical Framework for AI-Human Integration in Work

Data-driven Design of Randomized Control Trials with Guaranteed Treatment Effects

How to Evaluate and Mitigate IP Infringement in Visual Generative AI?

Batch List-Decodable Linear Regression via Higher Moments

Bayesian Basis Function Approximation for Scalable Gaussian Process Priors in Deep Generative Models

On the Impact of Performative Risk Minimization for Binary Random Variables

SEMU: Singular Value Decomposition for Efficient Machine Unlearning

Splitting & Integrating: Out-of-Distribution Detection via Adversarial Gradient Attribution

STAMP Your Content: Proving Dataset Membership via Watermarked Rephrasings

(How) Can Transformers Predict Pseudo-Random Numbers?

FlatQuant: Flatness Matters for LLM Quantization

A New Approach to Backtracking Counterfactual Explanations: A Unified Causal Framework for Efficient Model Interpretability

Addressing Concept Mislabeling in Concept Bottleneck Models Through Preference Optimization

Are Sparse Autoencoders Useful? A Case Study in Sparse Probing

Automatically Interpreting Millions of Features in Large Language Models

CoDy: Counterfactual Explainers for Dynamic Graphs

Constrain Alignment with Sparse Autoencoders

SAH-Drive: A Scenario-Aware Hybrid Planner for Closed-Loop Vehicle Trajectory Generation

Discovering Spoofing Attempts on Language Model Watermarks

From Individual Experience to Collective Evidence: A Reporting-Based Framework for Identifying Systemic Harms

GEFA: A General Feature Attribution Framework Using Proxy Gradient Estimation

InfoCons: Identifying Interpretable Critical Concepts in Point Clouds via Information Theory

Learning Multi-Level Features with Matryoshka Sparse Autoencoders

Learning to Route LLMs with Confidence Tokens

Leveraging Predictive Equivalence in Decision Trees

Near-Optimal Decision Trees in a SPLIT Second

Towards Robust Influence Functions with Flat Validation Minima

On Explaining Equivariant Graph Networks via Improved Relevance Propagation

On the Interplay between Graph Structure and Learning Algorithms in Graph Neural Networks

Proactive Agents for Multi-Turn Text-to-Image Generation Under Uncertainty

Generalization and Robustness of the Tilted Empirical Risk

SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretability

SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoders

Sum-of-Parts: Self-Attributing Neural Networks with End-to-End Learning of Feature Groups

The Illusion of Role Separation: Hidden Shortcuts in LLM Role Learning (and How to Fix Them)

Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignment

CollabLLM: From Passive Responders to Active Collaborators

Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective

Self-Consuming Generative Models with Adversarially Curated Data

MaskTwins: Dual-form Complementary Masking for Domain-Adaptive Image Segmentation

The Elicitation Game: Evaluating Capability Elicitation Techniques

Fairness Overfitting in Machine Learning: An Information-Theoretic Perspective

Policy Design for Two-sided Platforms with Participation Dynamics

The Disparate Benefits of Deep Ensembles

The Value of Prediction in Identifying the Worst-Off

Two Tickets are Better than One: Fair and Accurate Hiring Under Strategic LLM Manipulations

A Unified Theoretical Analysis of Private and Robust Offline Alignment: from RLHF to DPO

Approximate Differential Privacy of the $\ell_2$ Mechanism

Auditing $f$-differential privacy in one run

Breaking the $n^{1.5}$ Additive Error Barrier for Private and Efficient Graph Sparsification via Private Expander Decomposition

CAN: Leveraging Clients As Navigators for Generative Replay in Federated Continual Learning

Certified Unlearning for Neural Networks

Distributed Differentially Private Data Analytics via Secure Sketching

FreeMesh: Boosting Mesh Generation with Coordinates Merging

DRAG: Data Reconstruction Attack using Guided Diffusion

Efficient Source-free Unlearning via Energy-Guided Data Synthesis and Discrimination-Aware Multitask Optimization

EgoPrivacy: What Your First-Person Camera Says About You?

Sample Complexity of Correlation Detection in the Gaussian Wigner Model

Federated Disentangled Tuning with Textual Prior Decoupling and Visual Dynamic Adaptation

FOCoOp: Enhancing Out-of-Distribution Robustness in Federated Prompt Learning for Vision-Language Models

GHOST: Generalizable One-Shot Federated Graph Learning with Proxy-Based Topology Knowledge Retention

Lightweight Protocols for Distributed Private Quantile Estimation

Local Pan-privacy for Federated Analytics

On the Private Estimation of Smooth Transport Maps

Plausible Token Amplification for Improving Accuracy of Differentially Private In-Context Learning Based on Implicit Bayesian Inference

Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting

Privacy-Preserving Federated Convex Optimization: Balancing Partial-Participation and Efficiency via Noise Cancellation

Private Model Personalization Revisited

Active Treatment Effect Estimation via Limited Samples

Textual Unlearning Gives a False Sense of Unlearning

Towards Trustworthy Federated Learning with Untrusted Participants

COSDA: Counterfactual-based Susceptibility Risk Framework for Open-Set Domain Adaptation

Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning

Towards Lifelong Model Editing via Simulating Ideal Editor

SafetyAnalyst: Interpretable, Transparent, and Steerable Safety Moderation for AI Behavior

Variance as a Catalyst: Efficient and Transferable Semantic Erasure Adversarial Attack for Customized Diffusion Models

Watch Out Your Album! On the Inadvertent Privacy Memorization in Multi-Modal Large Language Models

unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning

Identifying and Understanding Cross-Class Features in Adversarial Training

Information Bottleneck-guided MLPs for Robust Spatial-temporal Forecasting

Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing

Ranked from Within: Ranking Large Multimodal Models Without Labels

Understanding and Mitigating Miscalibration in Prompt Tuning for Vision-Language Models

Emergent Symbolic Mechanisms Support Abstract Reasoning in Large Language Models

$S^2$FGL: Spatial Spectral Federated Graph Learning

A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment

MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

Assessing Safety Risks and Quantization-aware Safety Patching for Quantized Large Language Models

Improving Out-of-Distribution Detection via Dynamic Covariance Calibration

Just Enough Shifts: Mitigating Over-Refusal in Aligned Language Models with Targeted Representation Fine-Tuning

Mind the Gap: A Practical Attack on GGUF Quantization

SHE: Streaming-media Hashing Retrieval

MTL-UE: Learning to Learn Nothing for Multi-Task Learning

Not All Wrong is Bad: Using Adversarial Examples for Unlearning

Optimizing Adaptive Attacks against Watermarks for Language Models

OR-Bench: An Over-Refusal Benchmark for Large Language Models

POROver: Improving Safety and Reducing Overrefusal in Large Language Models with Overgeneration and Preference Optimization

Scaling Trends in Language Model Robustness

ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning

Speak Easy: Eliciting Harmful Jailbreaks from LLMs with Simple Interactions

The Limits of Predicting Agents from Behaviour

Towards Black-Box Membership Inference Attack for Diffusion Models

TRUST-VLM: Thorough Red-Teaming for Uncovering Safety Threats in Vision-Language Models

Adversarial Inception Backdoor Attacks against Reinforcement Learning

Online Learning with Unknown Constraints

Circumventing Backdoor Space via Weight Symmetry

De-AntiFake: Rethinking the Protective Perturbations Against Voice Cloning Attacks

GSM-$\infty$: How Do your LLMs Behave over Infinitely Increasing Reasoning Complexity and Context Length?

Diffusion-based Adversarial Purification from the Perspective of the Frequency Domain

Exploring and Mitigating Adversarial Manipulation of Voting-Based Leaderboards

ExpProof : Operationalizing Explanations for Confidential Models with ZKPs

MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents

MemFreezing: A Novel Adversarial Attack on Temporal Graph Neural Networks under Limited Future Knowledge

MixBridge: Heterogeneous Image-to-Image Backdoor Attack through Mixture of Schrödinger Bridges

Stay-Positive: A Case for Ignoring Real Image Features in Fake Image Detection

Topological Signatures of Adversaries in Multimodal Alignments

Conformal Tail Risk Control for Large Language Model Alignment

Convergence of Consistency Model with Multistep Sampling under General Data Assumptions

Feature learning from non-Gaussian inputs: the case of Independent Component Analysis in high dimensions

On the Similarities of Embeddings in Contrastive Learning

Private Lossless Multiple Release

Theoretical guarantees on the best-of-n alignment policy

Towards Better-than-2 Approximation for Constrained Correlation Clustering

A Rescaling-Invariant Lipschitz Bound Based on Path-Metrics for Modern ReLU Network Parameterizations

Algorithm Development in Neural Networks: Insights from the Streaming Parity Task

Benefits of Early Stopping in Gradient Descent for Overparameterized Logistic Regression

Curse of High Dimensionality Issue in Transformer for Long Context Modeling

Equivariant Neural Tangent Kernels

Grokking Beyond the Euclidean Norm of Model Parameters

Improved Expressivity of Hypergraph Neural Networks through High-Dimensional Generalized Weisfeiler-Leman Algorithms

LoRA Training Provably Converges to a Low-Rank Global Minimum Or It Fails Loudly (But it Probably Won't Fail)

Mind the Gap: a Spectral Analysis of Rank Collapse and Signal Propagation in Attention Layers

Optimal Task Order for Continual Learning of Multiple Tasks

Permutation Equivariant Neural Networks for Symmetric Tensors

Propagation of Chaos for Mean-Field Langevin Dynamics and its Application to Model Ensemble

Provable In-Context Vector Arithmetic via Retrieving Task Concepts

Representations Shape Weak-to-Strong Generalization: Theoretical Insights and Empirical Predictions

Time to Spike? Understanding the Representational Power of Spiking Neural Networks in Discrete Time

Towards characterizing the value of edge embeddings in Graph Neural Networks

A Closer Look at Multimodal Representation Collapse

A General Representation-Based Approach to Multi-Source Domain Adaptation

Function Encoders: A Principled Approach to Transfer Learning in Hilbert Spaces

The Role of Sparsity for Length Generalization in LLMs

Transfer Learning for Nonparametric Contextual Dynamic Pricing

When Can Proxies Improve the Sample Complexity of Preference Learning?

AKORN: Adaptive Knots generated Online for RegressioN splines

Dimensionality Reduction on Complex Vector Spaces for Euclidean Distance with Dynamic Weights

Exact Recovery of Sparse Binary Vectors from Generalized Linear Measurements

How Distributed Collaboration Influences the Diffusion Model Training? A Theoretical Perspective

Identifying Metric Structures of Deep Latent Variable Models

On Fine-Grained Distinct Element Estimation

Pareto-Optimality, Smoothness, and Stochasticity in Learning-Augmented One-Max-Search

QuEst: Enhancing Estimates of Quantile-Based Distributional Measures Using Model Predictions

The Price of Linear Time: Error Analysis of Structured Kernel Interpolation

The Underlying Universal Statistical Structure of Natural Datasets

A-PSRO: A Unified Strategy Learning Method with Advantage Metric for Normal-form Games

Best of Both Worlds: Regret Minimization versus Minimax Play

Beyond Self-Interest: How Group Strategies Reshape Content Creation in Recommendation Platforms?

Continuous-Time Analysis of Heavy Ball Momentum in Min-Max Games

Convex Markov Games: A New Frontier for Multi-Agent Reinforcement Learning

Observation Interference in Partially Observable Assistance Games

Procurement Auctions via Approximately Optimal Submodular Optimization

Provably Efficient Algorithm for Best Scoring Rule Identification in Online Principal-Agent Information Acquisition

Self-Play $Q$-Learners Can Provably Collude in the Iterated Prisoner's Dilemma

The impact of uncertainty on regularized learning in games

A Generalization Theory for Zero-Shot Prediction

A New Concentration Inequality for Sampling Without Replacement and Its Application for Transductive Learning

Approximation to Smooth Functions by Low-Rank Swish Networks

Certifiably Robust Model Evaluation in Federated Learning under Meta-Distributional Shifts

Design Considerations in Offline Preference-based RL

Discrepancies are Virtue: Weak-to-Strong Generalization through Lens of Intrinsic Dimension

Ehrenfeucht-Haussler Rank and Chain of Thought

Exactly Tight Information-theoretic Generalization Bounds via Binary Jensen-Shannon Divergence

Generalization in Federated Learning: A Conditional Mutual Information Framework

Grokking at the Edge of Linear Separability

Hypothesis Testing for Generalized Thurstone Models

Implicit Regularization for Tubal Tensor Factorizations via Gradient Descent

Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based Learning

Models of Heavy-Tailed Mechanistic Universality

No Free Lunch from Random Feature Ensembles: Scaling Laws and Near-Optimality Conditions

Nonlinear transformers can perform inference-time feature learning

On Learning Parallel Pancakes with Mostly Uniform Weights

On the Convergence of Continuous Single-timescale Actor-critic

On the Generalization Ability of Next-Token-Prediction Pretraining

On the Training Convergence of Transformers for In-Context Classification of Gaussian Mixtures

PDUDT: Provable Decentralized Unlearning under Dynamic Topologies

Provable Benefits of Unsupervised Pre-training and Transfer Learning via Single-Index Models

Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces

Retraining with Predicted Hard Labels Provably Increases Model Accuracy

Sample-Optimal Agnostic Boosting with Unlabeled Data

Theoretical Limitations of Ensembles in the Age of Overparameterization

Transfer Q-Learning with Composite MDP Structures

Understanding Generalization in Quantum Machine Learning with Margins

Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters

Unified Analysis of Continuous Weak Features Learning with Applications to Learning from Missing Data

Weak-to-Strong Generalization Even in Random Feature Networks, Provably

A Classification View on Meta Learning Bandits

A Trichotomy for List Transductive Online Learning

Adaptive Sample Sharing for Multi Agent Linear Bandits

ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning

Beyond Minimax Rates in Group Distributionally Robust Optimization via a Novel Notion of Sparsity

Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret

Constrained Pareto Set Identification with Bandit Feedback

Contextual Bandits for Unbounded Context Distributions

Dimension-Free Adaptive Subgradient Methods with Frequent Directions

Dueling Convex Optimization with General Preferences

Fixing the Loose Brake: Exponential-Tailed Stopping Time in Best Arm Identification

High Probability Bound for Cross-Learning Contextual Bandits with Unknown Context Distributions

High-Dimensional Prediction for Sequential Decision Making

Improved Regret Analysis in Gaussian Process Bandits: Optimality for Noiseless Reward, RKHS norm, and Non-Stationary Variance

Leveraging Offline Data in Linear Latent Contextual Bandits

Linear Bandits with Partially Observable Features

Multi-Armed Bandits with Interference: Bridging Causal Inference and Adversarial Bandits

Near-Optimal Consistency-Robustness Trade-Offs for Learning-Augmented Online Knapsack Problems

Nearly Optimal Algorithms for Contextual Dueling Bandits from Adversarial Feedback

Offline Learning for Combinatorial Multi-armed Bandits

On Mitigating Affinity Bias through Bandits with Evolving Biased Feedback

Optimal Algorithm for Max-Min Fair Bandit

Prediction-Aware Learning in Multi-Agent Systems

Regret-Free Reinforcement Learning for Temporal Logic Specifications

The Batch Complexity of Bandit Pure Exploration

$\mathcal{V}ista\mathcal{DPO}$: Video Hierarchical Spatial-Temporal Direct Preference Optimization for Large Video Models

A Near Linear Query Lower Bound for Submodular Maximization

A Parameter-Free and Near-Optimal Zeroth-Order Algorithm for Stochastic Convex Optimization

Graph-Based Algorithms for Diverse Similarity Search

Learning Mixtures of Experts with EM: A Mirror Descent Perspective

Maximum Coverage in Turnstile Streams with Applications to Fingerprinting Measures

On Differential Privacy for Adaptively Solving Search Problems via Sketching

Polynomial Time Learning Augmented Algorithms for NP-hard Permutation Problems

Radio: Rate–Distortion Optimization for Large Language Model Compression

Sparse-pivot: Dynamic correlation clustering for node insertions

Diffusion models for Gaussian distributions: Exact solutions and Wasserstein errors

Low-Rank Thinning

PAC-Bayes Analysis for Recalibration in Classification

Rectifying Conformity Scores for Better Conditional Coverage

A Reductions Approach to Risk-Sensitive Reinforcement Learning with Optimized Certainty Equivalents

A Sharper Global Convergence Analysis for Average Reward Reinforcement Learning via an Actor-Critic Approach

A Theoretical Justification for Asymmetric Actor-Critic Algorithms

Actor-Critics Can Achieve Optimal Sample Efficiency

Can RLHF be More Efficient with Imperfect Reward Models? A Policy Coverage Perspective

Convergence of Policy Mirror Descent Beyond Compatible Function Approximation

General agents need world models

Incentivize without Bonus: Provably Efficient Model-based Online Multi-agent RL for Markov Games

Minimax Optimal Regret Bound for Reinforcement Learning with Trajectory Feedback

Multi-objective Linear Reinforcement Learning with Lexicographic Rewards

Near-optimal Regret Using Policy Optimization in Online MDPs with Aggregate Bandit Feedback

Overcoming the Curse of Dimensionality in Reinforcement Learning Through Approximate Factorization

Partially Observable Reinforcement Learning with Memory Traces

Polynomial-Time Approximability of Constrained Reinforcement Learning

Square$\chi$PO: Differentially Private and Robust $\chi^2$-Preference Optimization in Offline Direct Alignment

Enhancing Certified Robustness via Block Reflector Orthogonal Layers and Logit Annealing Loss

GRU: Mitigating the Trade-off between Unlearning and Retention for LLMs

X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP

Robust Automatic Modulation Classification with Fuzzy Regularization

DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation

Efficient First-Order Optimization on the Pareto Set for Multi-Objective Learning under Preference Guidance

A First-order Generative Bilevel Optimization Framework for Diffusion Models

MTSTRec: Multimodal Time-Aligned Shared Token Recommender

DTZO: Distributed Trilevel Zeroth Order Learning with Provable Non-Asymptotic Convergence

An Entropy-Based Model for Hierarchical Learning

An Analysis of Quantile Temporal-Difference Learning

On the Impact of Hard Adversarial Instances on Overfitting in Adversarial Training

Compressed and distributed least-squares regression: convergence rates with applications to federated learning

Spherical Rotation Dimension Reduction with Geometric Loss Functions

Goal-Space Planning with Subgoal Models

Depth Degeneracy in Neural Networks: Vanishing Angles in Fully Connected ReLU Networks on Initialization

Semi-Supervised Blind Quality Assessment with Confidence-quantifiable Pseudo-label Learning for Authentic Images

MindCustomer: Multi-Context Image Generation Blended with Brain Signal

COKE: Core Kernel for More Efficient Approximation of Kernel Weights in Multiple Kernel Clustering

Improving the Scaling Laws of Synthetic Data with Deliberate Practice

Learning multivariate Gaussians with imperfect advice

NoLiMa: Long-Context Evaluation Beyond Literal Matching

TabICL: A Tabular Foundation Model for In-Context Learning on Large Data

MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidance

Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Models

MITIGATING OVER-EXPLORATION IN LATENT SPACE OPTIMIZATION USING LES

Density Ratio Estimation-based Bayesian Optimization with Semi-Supervised Learning

Enhancing Graph Invariant Learning from a Negative Inference Perspective

OOD-Chameleon: Is Algorithm Selection for OOD Generalization Learnable?

Complete-Tree Space Favors Data-Efficient Link Prediction

SAFE: Finding Sparse and Flat Minima to Improve Pruning

Revisiting Differentially Private Algorithms for Decentralized Online Learning

Attention Mechanisms Perspective: Exploring LLM Processing of Graph-Structured Data

Navigating the Social Welfare Frontier: Portfolios for Multi-objective Reinforcement Learning

ArrayDPS: Unsupervised Blind Speech Separation with a Diffusion Prior

Wolfpack Adversarial Attack for Robust Multi-Agent Reinforcement Learning

PEINR: A Physics-enhanced Implicit Neural Representation for High-Fidelity Flow Field Reconstruction

Optimistic Algorithms for Adaptive Estimation of the Average Treatment Effect

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reinforcement Learning with Segment Feedback

Rethinking Score Distilling Sampling for 3D Editing and Generation

Test-Time Multimodal Backdoor Detection by Contrastive Prompting

Distillation Scaling Laws

Generalized Random Forests Using Fixed-Point Trees

Positional Encoding meets Persistent Homology on Graphs

PoisonBench: Assessing Large Language Model Vulnerability to Poisoned Preference Data

Doubly Robust Fusion of Many Treatments for Policy Learning

SPD: Sync-Point Drop for Efficient Tensor Parallelism of Large Language Models

SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer

SITCOM: Step-wise Triple-Consistent Diffusion Sampling For Inverse Problems

Commute Graph Neural Networks

KoopSTD: Reliable Similarity Analysis between Dynamical Systems via Approximating Koopman Spectrum with Timescale Decoupling

Improving Reward Model Generalization from Adversarial Process Enhanced Preferences

Mastering Multiple-Expert Routing: Realizable $H$-Consistency and Strong Guarantees for Learning to Defer

Training Deep Learning Models with Norm-Constrained LMOs

Doubly Robust Conformalized Survival Analysis with Right-Censored Data

M³HF: Multi-agent Reinforcement Learning from Multi-phase Human Feedback of Mixed Quality

Overcoming Multi-step Complexity in Multimodal Theory-of-Mind Reasoning: A Scalable Bayesian Planner

Weight matrices compression based on PDB model in deep neural networks

IMTS is Worth Time $\times$ Channel Patches: Visual Masked Autoencoders for Irregular Multivariate Time Series Prediction

Measuring Diversity: Axioms and Challenges

Geometric Generative Modeling with Noise-Conditioned Graph Networks

Autoformulation of Mathematical Optimization Models Using LLMs

The Empirical Mean is Minimax Optimal for Local Glivenko-Cantelli

On the Statistical Mechanisms of Distributional Compositional Generalization

DyCodeEval: Dynamic Benchmarking of Reasoning Capabilities in Code Large Language Models Under Data Contamination

PDE-Transformer: Efficient and Versatile Transformers for Physics Simulations

A Peer-review Look on Multi-modal Clustering: An Information Bottleneck Realization Method

Leveraging Randomness in Model and Data Partitioning for Privacy Amplification

CellFlux: Simulating Cellular Morphology Changes via Flow Matching

Flat-LoRA: Low-Rank Adaptation over a Flat Loss Landscape

Great Models Think Alike and this Undermines AI Oversight

Improving Parallel Program Performance with LLM Optimizers via Agent-System Interfaces

SHARP-Distill: A 68× Faster Recommender System with Hypergraph Neural Networks and Language Models

RE-Bench: Evaluating Frontier AI R&D Capabilities of Language Model Agents against Human Experts

Off-Policy Actor-Critic for Adversarial Observation Robustness: Virtual Alternative Training via Symmetric Policy Evaluation

Prediction-Powered Adaptive Shrinkage Estimation

Scalable First-order Method for Certifying Optimal k-Sparse GLMs

A Dynamical Systems-Inspired Pruning Strategy for Addressing Oversmoothing in Graph Attention Networks

Subspace Optimization for Large Language Models with Convergence Guarantees

Do We Need to Verify Step by Step? Rethinking Process Supervision from a Theoretical Perspective

Contextures: Representations from Contexts

Winner-takes-all for Multivariate Probabilistic Time Series Forecasting

Prices, Bids, Values: One ML-Powered Combinatorial Auction to Rule Them All

DVI:A Derivative-based Vision Network for INR

Compositional Flows for 3D Molecule and Synthesis Pathway Co-design

Random Registers for Cross-Domain Few-Shot Learning

Learning Likelihood-Free Reference Priors

Ultra-Resolution Adaptation with Ease

Premise-Augmented Reasoning Chains Improve Error Identification in Math reasoning with LLMs

RAGGED: Towards Informed Design of Scalable and Stable RAG Systems

Adversarial Robust Generalization of Graph Neural Networks

Raptor: Scalable Train-Free Embeddings for 3D Medical Volumes Leveraging Pretrained 2D Foundation Models

Demystifying Singular Defects in Large Language Models

ToMA: Token Merge with Attention for Diffusion Models

Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach

Can Transformers Reason Logically? A Study in SAT Solving

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

TabPFN Unleashed: A Scalable and Effective Solution to Tabular Classification Problems

Catoni Contextual Bandits are Robust to Heavy-tailed Rewards

UI-Vision: A Desktop-centric GUI Benchmark for Visual Perception and Interaction

Prior Knowledge Guided Neural Architecture Generation

Learning Optimal Multimodal Information Bottleneck Representations

Fast and Robust: Task Sampling with Posterior and Diversity Synergies for Adaptive Decision-Makers in Randomized Environments

Graph Attention is Not Always Beneficial: A Theoretical Analysis of Graph Attention Mechanisms via Contextual Stochastic Block Models

Online Pre-Training for Offline-to-Online Reinforcement Learning

Sorbet: A Neuromorphic Hardware-Compatible Transformer-Based Spiking Language Model

Prune 'n Predict: Optimizing LLM Decision-making with Conformal Prediction

Uniform Mean Estimation for Heavy-Tailed Distributions via Median-of-Means

The Synergy of LLMs & RL Unlocks Offline Learning of Generalizable Language-Conditioned Policies with Low-fidelity Data

CAT: Contrastive Adversarial Training for Evaluating the Robustness of Protective Perturbations in Latent Diffusion Models

Learn to Vaccinate: Combining Structure Learning and Effective Vaccination for Epidemic and Outbreak Control

Riemannian Diffusion Adaptation for Distributed Optimization on Manifolds

A Simple Model of Inference Scaling Laws

To Each Metric Its Decoding: Post-Hoc Optimal Decision Rules of Probabilistic Hierarchical Classifiers

DISCO: learning to DISCover an evolution Operator for multi-physics-agnostic prediction

Cannot See the Forest for the Trees: Invoking Heuristics and Biases to Elicit Irrational Choices of LLMs

Test-Time Selective Adaptation for Uni-Modal Distribution Shift in Multi-Modal Data

Peripheral Memory for LLMs: Integration of Sequential Memory Banks with Adaptive Querying

Understanding Complexity in VideoQA via Visual Program Generation

Efficient Length-Generalizable Attention via Causal Retrieval for Long-Context Language Modeling

Sounding that Object: Interactive Object-Aware Image to Audio Generation

Efficient Distributed Optimization under Heavy-Tailed Noise

Targeted Unlearning with Single Layer Unlearning Gradient

Efficient Multi-modal Long Context Learning for Training-free Adaptation

Large Language Models to Diffusion Finetuning

FDGen: A Fairness-Aware Graph Generation Model

Earley-Driven Dynamic Pruning for Efficient Structured Decoding

Reinforce LLM Reasoning through Multi-Agent Reflection

HybridGS: High-Efficiency Gaussian Splatting Data Compression using Dual-Channel Sparse Representation and Point Cloud Encoder

Sample-specific Noise Injection for Diffusion-based Adversarial Purification

CostFilter-AD: Enhancing Anomaly Detection through Matching Cost Filtering

Stochastic Layer-Wise Shuffle for Improving Vision Mamba Training

Relative Error Fair Clustering in the Weak-Strong Oracle Model

$K^2$VAE: A Koopman-Kalman Enhanced Variational AutoEncoder for Probabilistic Time Series Forecasting

Graph-Assisted Stitching for Offline Hierarchical Reinforcement Learning

RAPID: Long-Context Inference with Retrieval-Augmented Speculative Decoding

A Manifold Perspective on the Statistical Generalization of Graph Neural Networks

CSV-Occ: Fusing Multi-frame Alignment for Occupancy Prediction with Temporal Cross State Space Model and Central Voting Mechanism

On Understanding Attention-Based In-Context Learning for Categorical Data

Decomposition of Graphic Design with Unified Multimodal Model

QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache

TruthFlow: Truthful LLM Generation via Representation Flow Correction

Statistical and Computational Guarantees of Kernel Max-Sliced Wasserstein Distances

Confidence Difference Reflects Various Supervised Signals in Confidence-Difference Classification

Event-Customized Image Generation

FLAM: Frame-Wise Language-Audio Modeling

CSG-ODE: ControlSynth Graph ODE For Modeling Complex Evolution of Dynamic Graphs

Avoiding Leakage Poisoning: Concept Interventions Under Distribution Shifts

LineFlow: A Framework to Learn Active Control of Production Lines

Mechanistic PDE Networks for Discovery of Governing Equations

Confounder-Free Continual Learning via Recursive Feature Normalization

One Arrow, Two Hawks: Sharpness-aware Minimization for Federated Learning via Global Model Trajectory

All-atom Diffusion Transformers: Unified generative modelling of molecules and materials

Controlling Neural Collapse Enhances Out-of-Distribution Detection and Transfer Learning

Robust Noise Attenuation via Adaptive Pooling of Transformer Outputs

TOPLOC: A Locality Sensitive Hashing Scheme for Trustless Verifiable Inference

(How) Do Language Models Track State?

LMAct: A Benchmark for In-Context Imitation Learning with Long Multimodal Demonstrations

It's My Data Too: Private ML for Datasets with Multi-User Training Examples

EffiCoder: Enhancing Code Generation in Large Language Models through Efficiency-Aware Fine-tuning

A Certified Unlearning Approach without Access to Source Data

Improving Out-of-Distribution Detection with Markov Logic Networks

Multiple-policy Evaluation via Density Estimation

Self-Consistency Preference Optimization

BinauralFlow: A Causal and Streamable Approach for High-Quality Binaural Speech Synthesis with Flow Matching Models

C2IQL: Constraint-Conditioned Implicit Q-learning for Safe Offline Reinforcement Learning

Clustering Items through Bandit Feedback: Finding the Right Feature out of Many

Nesterov Method for Asynchronous Pipeline Parallel Optimization

Balanced Learning for Domain Adaptive Semantic Segmentation

SERENA: A Unified Stochastic Recursive Variance Reduced Gradient Framework for Riemannian Non-Convex Optimization

Safe-EF: Error Feedback for Non-smooth Constrained Optimization

Flow-based Domain Randomization for Learning and Sequencing Robotic Skills

Pairwise Maximum Likelihood For Multi-Class Logistic Regression Model With Multiple Rare Classes

Feature-Mapping Topology Optimization with Neural Heaviside Signed Distance Functions

WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs

TeLoGraF: Temporal Logic Planning via Graph-encoded Flow Matching

Pareto-frontier Entropy Search with Variational Lower Bound Maximization

One-dimensional Path Convolution

Improving Memory Efficiency for Training KANs via Meta Learning

Copilot Arena: A Platform for Code LLM Evaluation in the Wild

TCP-Diffusion: A Multi-modal Diffusion Model for Global Tropical Cyclone Precipitation Forecasting with Change Awareness

Segment Anyword: Mask Prompt Inversion for Open-Set Grounded Segmentation

Ab Initio Nonparametric Variable Selection for Scalable Symbolic Regression with Large $p$

MODA: MOdular Duplex Attention for Multimodal Perception, Cognition, and Emotion Understanding

Accelerating Unbiased LLM Evaluation via Synthetic Feedback

Robust Consensus Anchor Learning for Efficient Multi-view Subspace Clustering

Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models

Clustering via Self-Supervised Diffusion

IMPACT: Iterative Mask-based Parallel Decoding for Text-to-Audio Generation with Diffusion Modeling

OneForecast: A Universal Framework for Global and Regional Weather Forecasting

TMetaNet: Topological Meta-Learning Framework for Dynamic Link Prediction

Harmonizing Geometry and Uncertainty: Diffusion with Hyperspheres

Safety Certificate against Latent Variables with Partially Unidentifiable Dynamics

GenZSL: Generative Zero-Shot Learning Via Inductive Variational Autoencoder

3D-LMVIC: Learning-based Multi-View Image Compression with 3D Gaussian Geometric Priors

David and Goliath: Small One-step Model Beats Large Diffusion with Score Post-training

Learning Progress Driven Multi-Agent Curriculum

Exogenous Isomorphism for Counterfactual Identifiability

Distributed Nonparametric Estimation: from Sparse to Dense Samples per Terminal

Teaching Physical Awareness to LLMs through Sounds

Knowledge Swapping via Learning and Unlearning

PAC Learning with Improvements

Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing

Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM Reasoning

A Model of Place Field Reorganization During Reward Maximization

iN2V: Bringing Transductive Node Embeddings to Inductive Graphs

RBench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation

The Role of Randomness in Stability

OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning

A Online Statistical Framework for Out-of-Distribution Detection

PolyConf: Unlocking Polymer Conformation Generation through Hierarchical Generative Models

Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning

Revisiting Noise Resilience Strategies in Gesture Recognition: Short-Term Enhancement in sEMG Analysis

Detecting Strategic Deception with Linear Probes

Which Attention Heads Matter for In-Context Learning?

Clients Collaborate: Flexible Differentially Private Federated Learning with Guaranteed Improvement of Utility-Privacy Trade-off

Devil is in the Details: Density Guidance for Detail-Aware Generation with Flow Models

Sable: a Performant, Efficient and Scalable Sequence Model for MARL

Bellman Unbiasedness: Toward Provably Efficient Distributional Reinforcement Learning with General Value Function Approximation

LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs – No Silver Bullet for LC or RAG Routing

Learning from True-False Labels via Multi-modal Prompt Retrieving

VCT: Training Consistency Models with Variational Noise Coupling

Tracking Most Significant Shifts in Infinite-Armed Bandits

More Than Meets the Eye: Enhancing Multi-Object Tracking Even with Prolonged Occlusions

Near-Optimal Sample Complexity for MDPs via Anchoring

Emergent Response Planning in LLMs

UniSim: A Unified Simulator for Time-Coarsened Dynamics of Biomolecules

Heavy-Tailed Linear Bandits: Huber Regression with One-Pass Update

Exploiting Curvature in Online Convex Optimization with Delayed Feedback

Comparing Comparisons: Informative and Easy Human Feedback with Distinguishability Queries

Tensor Product Neural Networks for Functional ANOVA Model

Conformal Anomaly Detection in Event Sequences

MultiPDENet: PDE-embedded Learning with Multi-time-stepping for Accelerated Flow Simulation

Distilling the Knowledge in Data Pruning

Re-ranking Reasoning Context with Tree Search Makes Large Vision-Language Models Stronger

How Far Is Video Generation from World Model: A Physical Law Perspective

Zero-Inflated Bandits

Diffusion Instruction Tuning

Robust Multi-bit Text Watermark with LLM-based Paraphrasers

Text-to-CAD Generation Through Infusing Visual Feedback in Large Language Models

A Closer Look at Backdoor Attacks on CLIP

Contract Design Under Approximate Best Responses

RobustZero: Enhancing MuZero Reinforcement Learning Robustness to State Perturbations

From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models

Multilayer Matrix Factorization via Dimension-Reducing Diffusion Variational Inference

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents

FlexTok: Resampling Images into 1D Token Sequences of Flexible Length

Angle Domain Guidance: Latent Diffusion Requires Rotation Rather Than Extrapolation

Omni-Angle Assault: An Invisible and Powerful Physical Adversarial Attack on Face Recognition

Spectral-Aware Reservoir Computing for Fast and Accurate Time Series Classification

Stochastic Control for Fine-tuning Diffusion Models: Optimality, Regularity, and Convergence

Enforcing Latent Euclidean Geometry in Single-Cell VAEs for Manifold Interpolation

Inverse Flow and Consistency Models

Distributionally Robust Multi-Agent Reinforcement Learning for Dynamic Chute Mapping

When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series

Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions

Flopping for FLOPs: Leveraging Equivariance for Computational Efficiency

Perceptually Constrained Precipitation Nowcasting Model

The Logical Implication Steering Method for Conditional Interventions on Transformer Generation

Constrained Online Convex Optimization with Polyak Feasibility Steps

Hierarchical Refinement: Optimal Transport to Infinity and Beyond

Efficient LiDAR Reflectance Compression via Scanning Serialization

Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection

An Architecture Search Framework for Inference-Time Techniques

Active Evaluation Acquisition for Efficient LLM Benchmarking

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning

MuLan: Adapting Multilingual Diffusion Models for Hundreds of Languages with Negligible Cost

Uncertainty-Based Extensible Codebook for Discrete Federated Learning in Heterogeneous Data Silos

Mitigating Local Cohesion and Global Sparseness in Graph Contrastive Learning with Fuzzy Boundaries

Learning-Order Autoregressive Models with Application to Molecular Graph Generation

Statistical Query Hardness of Multiclass Linear Classification with Random Classification Noise

CFP-Gen: Combinatorial Functional Protein Generation via Diffusion Language Models

Mitigating Plasticity Loss in Continual Reinforcement Learning by Reducing Churn

On-Device Collaborative Language Modeling via a Mixture of Generalists and Specialists

Reinforcement Learning for Quantum Control under Physical Constraints

Emotional Face-to-Speech

The Price of Freedom: Exploring Expressivity and Runtime Tradeoffs in Equivariant Tensor Products

TAROT: Targeted Data Selection via Optimal Transport

Demeaned Sparse: Efficient Anomaly Detection by Residual Estimate

In-Context Denoising with One-Layer Transformers: Connections between Attention and Associative Memory Retrieval

Layer-wise Alignment: Examining Safety Alignment Across Image Encoder Layers in Vision Language Models

Eigen Analysis of Conjugate Kernel and Neural Tangent Kernel

Is Complex Query Answering Really Complex?

Idiosyncrasies in Large Language Models

Model Swarms: Collaborative Search to Adapt LLM Experts via Swarm Intelligence

Physics-Informed Generative Modeling of Wireless Channels

Whitened CLIP as a Likelihood Surrogate of Images and Captions

Generative Modeling Reinvents Supervised Learning: Label Repurposing with Predictive Consistency Learning

CombiMOTS: Combinatorial Multi-Objective Tree Search for Dual-Target Molecule Generation

Learning-Augmented Hierarchical Clustering

Improved Online Confidence Bounds for Multinomial Logistic Bandits

From Pixels to Perception: Interpretable Predictions via Instance-wise Grouped Feature Selection

O-MAPL: Offline Multi-agent Preference Learning

Unifying 2D and 3D Vision-Language Understanding

Improving Flow Matching by Aligning Flow Divergence

Learning Safety Constraints for Large Language Models

Exploring Large Action Sets with Hyperspherical Embeddings using von Mises-Fisher Sampling

LowRA: Accurate and Efficient LoRA Fine-Tuning of LLMs under 2 Bits

Graph Generative Pre-trained Transformer

Graph-constrained Reasoning: Faithful Reasoning on Knowledge Graphs with Large Language Models

Scaling Collapse Reveals Universal Dynamics in Compute-Optimally Trained Neural Networks

Unveiling Markov heads in Pretrained Language Models for Offline Reinforcement Learning

Smoothed Preference Optimization via ReNoise Inversion for Aligning Diffusion Models with Varied Human Preferences

UncertainSAM: Fast and Efficient Uncertainty Quantification of the Segment Anything Model

LRA-QViT: Integrating Low-Rank Approximation and Quantization for Robust and Efficient Vision Transformers

Towards Rationale-Answer Alignment of LVLMs via Self-Rationale Calibration

A Unified Comparative Study with Generalized Conformity Scores for Multi-Output Conformal Regression

Learning Classifiers That Induce Markets

A Parametric Contextual Online Learning Theory of Brokerage

Diss-l-ECT: Dissecting Graph Data with Local Euler Characteristic Transforms

Learning Invariant Causal Mechanism from Vision-Language Models

ITFormer: Bridging Time Series and Natural Language for Multi-Modal QA with Large-Scale Multitask Dataset

Adversaries Can Misuse Combinations of Safe Models

ProofAug: Efficient Neural Theorem Proving via Fine-grained Proof Structure Analysis

Uncertainty Estimation for Heterophilic Graphs Through the Lens of Information Theory

Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection

A Variational Information Theoretic Approach to Out-of-Distribution Detection

Best of Both Worlds: Advantages of Hybrid Graph Sequence Models

The Noisy Laplacian: a Threshold Phenomenon for Non-Linear Dimension Reduction

Shortcut-connected Expert Parallelism for Accelerating Mixture of Experts

Active Reward Modeling: Adaptive Preference Labeling for Large Language Model Alignment

Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems

Morse: Dual-Sampling for Lossless Acceleration of Diffusion Models

Model-Based Exploration in Monitored Markov Decision Processes

Efficient Quantification of Multimodal Interaction at Sample Level

Deep Streaming View Clustering

On the Power of Context-Enhanced Learning in LLMs

Adaptive Multi-prompt Contrastive Network for Few-shot Out-of-distribution Detection

Towards Robustness and Explainability of Automatic Algorithm Selection

Feasible Action Search for Bandit Linear Programs via Thompson Sampling

Diagonal Symmetrization of Neural Network Solvers for the Many-Electron Schrödinger Equation

Structured Preconditioners in Adaptive Optimization: A Unified Analysis

Fragments to Facts: Partial-Information Fragment Inference from LLMs

Benign Overfitting in Token Selection of Attention Mechanism

Probabilistic Group Mask Guided Discrete Optimization for Incremental Learning

Towards Efficient Online Tuning of VLM Agents via Counterfactual Soft Reinforcement Learning

Dendritic Localized Learning: Toward Biologically Plausible Algorithm

Multiobjective distribution matching

Activation Space Interventions Can Be Transferred Between Large Language Models

Efficient Parallel Training Methods for Spiking Neural Networks with Constant Time Complexity

Learning Robust Neural Processes with Risk-Averse Stochastic Optimization

Exploring Invariance in Images through One-way Wave Equations

Overcoming Non-monotonicity in Transducer-based Streaming Generation

WMarkGPT: Watermarked Image Understanding via Multimodal Large Language Models

What If We Recaption Billions of Web Images with LLaMA-3?

FSL-SAGE: Accelerating Federated Split Learning via Smashed Activation Gradient Estimation

Universal Biological Sequence Reranking for Improved De Novo Peptide Sequencing

CoMemo: LVLMs Need Image Context with Image Memory

Density Ratio Estimation with Conditional Probability Paths

Log-Sum-Exponential Estimator for Off-Policy Evaluation and Learning

Flowing Datasets with Wasserstein over Wasserstein Gradient Flows

Targeted control of fast prototyping through domain-specific interface

The Number of Trials Matters in Infinite-Horizon General-Utility Markov Decision Processes

Quantifying Treatment Effects: Estimating Risk Ratios via Observational Studies

DiMa: Understanding the Hardness of Online Matching Problems via Diffusion Models

Beyond Confidence: Exploiting Homogeneous Pattern for Semi-Supervised Semantic Segmentation

Boost-and-Skip: A Simple Guidance-Free Diffusion for Minority Generation

Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coverage

PCEvolve: Private Contrastive Evolution for Synthetic Dataset Generation via Few-Shot Private Data and Generative APIs

Unifews: You Need Fewer Operations for Efficient Graph Neural Networks

Self-Supervised Transformers as Iterative Solution Improvers for Constraint Satisfaction

Contextual Linear Bandits with Delay as Payoff

Improving Transformer World Models for Data-Efficient RL

Stochastic Online Conformal Prediction with Semi-Bandit Feedback

Understanding Chain-of-Thought in LLMs through Information Theory

Discrete Neural Algorithmic Reasoning

How to set AdamW's weight decay as you scale model and dataset size

MedXpertQA: Benchmarking Expert-Level Medical Reasoning and Understanding

Are Large Brainwave Foundation Models Capable Yet ? Insights from Fine-Tuning

Enhancing Decision-Making of Large Language Models via Actor-Critic

Graph Neural Network Generalization With Gaussian Mixture Model Based Augmentation

Diff-MoE: Diffusion Transformer with Time-Aware and Space-Adaptive Experts

Symmetry-Aware GFlowNets

Deep Ridgelet Transform and Unified Universality Theorem for Deep and Shallow Joint-Group-Equivariant Machines

Bi-perspective Splitting Defense: Achieving Clean-Seed-Free Backdoor Security

Test-Time Canonicalization by Foundation Models for Robust Perception

When, Where and Why to Average Weights?

Visual Autoregressive Modeling for Image Super-Resolution

Revisiting Cooperative Off-Policy Multi-Agent Reinforcement Learning

Trustworthy Machine Learning through Data-Specific Indistinguishability

Step-DAD: Semi-Amortized Policy-Based Bayesian Experimental Design

Fast Estimation of Partial Dependence Functions using Trees

Spherical-Nested Diffusion Model for Panoramic Image Outpainting

CSTrack: Enhancing RGB-X Tracking via Compact Spatiotemporal Features

Learning Event Completeness for Weakly Supervised Video Anomaly Detection

Efficient Graph Continual Learning via Lightweight Graph Neural Tangent Kernels-based Dataset Distillation

Disparate Conditional Prediction in Multiclass Classifiers

Continual Generalized Category Discovery: Learning and Forgetting from a Bayesian Perspective

Compositional Scene Understanding through Inverse Generative Modeling

Optimization Proxies using Limited Labeled Data and Training Time -- A Semi-Supervised Bayesian Neural Network Approach

A General Framework for Inference-time Scaling and Steering of Diffusion Models

Inference-Time Alignment of Diffusion Models with Direct Noise Optimization

Understanding and Mitigating Memorization in Diffusion Models for Tabular Data

Preference Optimization for Combinatorial Optimization Problems

Adaptive Learn-then-Test: Statistically Valid and Efficient Hyperparameter Selection

Learning Changes in Graphon Attachment Network Models

WeGeFT: Weight‑Generative Fine‑Tuning for Multi‑Faceted Efficient Adaptation of Large Models

LLM Alignment as Retriever Optimization: An Information Retrieval Perspective

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

Efficient and Scalable Density Functional Theory Hamiltonian Prediction through Adaptive Sparsity

Test-Time Adaptation for Online Vision-Language Navigation with Feedback-based Reinforcement Learning

Improving Model Alignment Through Collective Intelligence of Open-Source Models

GradPS: Resolving Futile Neurons in Parameter Sharing Network for Multi-Agent Reinforcement Learning

XAttention: Block Sparse Attention with Antidiagonal Scoring

Towards Understanding Parametric Generalized Category Discovery on Graphs

DeFoG: Discrete Flow Matching for Graph Generation

An Expressive and Self-Adaptive Dynamical System for Efficient Function Learning

UniMoMo: Unified Generative Modeling of 3D Molecules for De Novo Binder Design

ADIOS: Antibody Development via Opponent Shaping

FuseUNet: A Multi-Scale Feature Fusion Method for U-like Networks

Towards Memorization Estimation: Fast, Formal and Free

LipsNet++: Unifying Filter and Controller into a Policy Network

Counting in Small Transformers: The Delicate Interplay between Attention and Feed-Forward Layers

Learning Time-Aware Causal Representation for Model Generalization in Evolving Domains

Optimal Transfer Learning for Missing Not-at-Random Matrix Completion

A Hitchhiker's Guide to Scaling Law Estimation

LoRA-One: One-Step Full Gradient Could Suffice for Fine-Tuning Large Language Models, Provably and Efficiently

The Butterfly Effect: Neural Network Training Trajectories Are Highly Sensitive to Initial Conditions

Flexible and Efficient Grammar-Constrained Decoding

Policy Filtration for RLHF to Mitigate Noise in Reward Models

On the Diversity of Adversarial Ensemble Learning

LangDAug: Langevin Data Augmentation for Multi-Source Domain Generalization in Medical Image Segmentation

A Novel Characterization of the Population Area Under the Risk Coverage Curve (AURC) and Rates of Finite Sample Estimators

Expected Variational Inequalities

Preference Adaptive and Sequential Text-to-Image Generation

Accelerating Quantum Reinforcement Learning with a Quantum Natural Policy Gradient Based Approach

LETS Forecast: Learning Embedology for Time Series Forecasting

HALoS: Hierarchical Asynchronous Local SGD over Slow Networks for Geo-Distributed Large Language Model Training

Clipping Improves Adam-Norm and AdaGrad-Norm when the Noise Is Heavy-Tailed

Sundial: A Family of Highly Capable Time Series Foundation Models

Context-Informed Neural ODEs Unexpectedly Identify Broken Symmetries: Insights from the Poincaré–Hopf Theorem

Generalized Smooth Bilevel Optimization with Nonconvex Lower-Level

Provably Improving Generalization of Few-shot models with Synthetic Data

From Crowdsourced Data to High-quality Benchmarks: Arena-Hard and Benchbuilder Pipeline

Learning dynamics in linear recurrent neural networks

Reconstructing Cell Lineage Trees from Phenotypic Features with Metric Learning

KV Shifting Attention Enhances Language Modeling

Variational Learning of Fractional Posteriors

FireFlow: Fast Inversion of Rectified Flow for Image Semantic Editing

Incremental Gradient Descent with Small Epoch Counts is Surprisingly Slow on Ill-Conditioned Problems

3D Question Answering via only 2D Vision-Language Models

CACTI: Leveraging Copy Masking and Contextual Information to Improve Tabular Data Imputation

Causal Attribution Analysis for Continuous Outcomes

LLMs on the Line: Data Determines Loss-to-Loss Scaling Laws

When and How Does CLIP Enable Domain and Compositional Generalization?

AffinityFlow: Guided Flows for Antibody Affinity Maturation

Adaptive Elicitation of Latent Information Using Natural Language

QuEST: Stable Training of LLMs with 1-Bit Weights and Activations

LEAPS: A discrete neural sampler via locally equivariant networks

CEGA: A Cost-Effective Approach for Graph-Based Model Extraction and Acquisition

Hyperband-based Bayesian Optimization for Black-box Prompt Selection

CoreMatching: A Co-adaptive Sparse Inference Framework with Token and Neuron Pruning for Comprehensive Acceleration of Vision-Language Models

Laplace Transform Based Low-Complexity Learning of Continuous Markov Semigroups

Efficient Multivariate Robust Mean Estimation Under Mean-Shift Contamination

KoNODE: Koopman-Driven Neural Ordinary Differential Equations with Evolving Parameters for Time Series Analysis

Structure-informed Risk Minimization for Robust Ensemble Learning

Learning curves theory for hierarchically compositional data with power-law distributed features

Improving Consistency Models with Generator-Augmented Flows

Stabilizing Sample Similarity in Representation via Mitigating Random Consistency

Fundamental Bias in Inverting Random Sampling Matrices with Application to Sub-sampled Newton

An Asymptotically Optimal Approximation Algorithm for Multiobjective Submodular Maximization at Scale

Pfeife: Automatic Pipeline Parallelism for PyTorch

BOPO: Neural Combinatorial Optimization via Best-anchored and Objective-guided Preference Optimization

AutoAdvExBench: Benchmarking Autonomous Exploitation of Adversarial Example Defenses

Does Data Scaling Lead to Visual Compositional Generalization?

Graph4MM: Weaving Multimodal Learning with Structural Information

Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging

How Do Images Align and Complement LiDAR? Towards a Harmonized Multi-modal 3D Panoptic Segmentation

Signed Laplacians for Constrained Graph Clustering

Understanding the Statistical Accuracy-Communication Trade-off in Personalized Federated Learning with Minimax Guarantees

Flow-of-Options: Diversified and Improved LLM Reasoning by Thinking Through Options

MindLLM: A Subject-Agnostic and Versatile Model for fMRI-to-text Decoding

Posterior Inference with Diffusion Models for High-dimensional Black-box Optimization

Generative Social Choice: The Next Generation

System-Aware Unlearning Algorithms: Use Lesser, Forget Faster

Pointwise Information Measures as Confidence Estimators in Deep Neural Networks: A Comparative Study

Auto-reconfiguration for Latency Minimization in CPU-based DNN Serving

Hypo3D: Exploring Hypothetical Reasoning in 3D

Visual Abstraction: A Plug-and-Play Approach for Text-Visual Retrieval

Strategic A/B testing via Maximum Probability-driven Two-armed Bandit

Context Matters: Query-aware Dynamic Long Sequence Modeling of Gigapixel Images

Risk-Sensitive Theory of Mind: Coordinating with Agents of Unknown Bias using Cumulative Prospect Theory

Supercharging Graph Transformers with Advective Diffusion

Improved Theoretically-Grounded Evolutionary Algorithms for Subset Selection with a Linear Cost Constraint

Quantifying Prediction Consistency Under Fine-tuning Multiplicity in Tabular LLMs

KABB: Knowledge-Aware Bayesian Bandits for Dynamic Expert Coordination in Multi-Agent Systems

SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator

An End-to-End Model for Logits-Based Large Language Models Watermarking

Lightweight Dataset Pruning without Full Training via Example Difficulty and Prediction Uncertainty

Decision-aware Training of Spatiotemporal Forecasting Models to Select a Top-K Subset of Sites for Intervention

Weisfeiler and Leman Go Gambling: Why Expressive Lottery Tickets Win

Weak-to-Strong Jailbreaking on Large Language Models

Stochastic Encodings for Active Feature Acquisition

Exploiting Presentative Feature Distributions for Parameter-Efficient Continual Learning of Large Language Models

Cross-Modal Alignment via Variational Copula Modelling

Scaling Sparse Feature Circuits For Studying In-Context Learning

Logarithmic Regret for Online KL-Regularized Reinforcement Learning

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking

Improving Your Model Ranking on Chatbot Arena by Vote Rigging

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention

Human Body Restoration with One-Step Diffusion Model and A New Benchmark

What Limits Virtual Agent Application? OmniBench: A Scalable Multi-Dimensional Benchmark for Essential Virtual Agent Capabilities

GraphGPT: Generative Pre-trained Graph Eulerian Transformer

Sparse Autoencoders for Hypothesis Generation

NestQuant: nested lattice quantization for matrix products and LLMs

Origin Identification for Text-Guided Image-to-Image Diffusion Models

LLM Data Selection and Utilization via Dynamic Bi-level Optimization

NICE Data Selection for Instruction Tuning in LLMs with Non-differentiable Evaluation Metric

WGFormer: An SE(3)-Transformer Driven by Wasserstein Gradient Flows for Molecular Ground-State Conformation Prediction

Since Faithfulness Fails: The Performance Limits of Neural Causal Discovery

A Selective Learning Method for Temporal Graph Continual Learning

Improving the Diffusability of Autoencoders

On Measuring Long-Range Interactions in Graph Neural Networks

VinePPO: Refining Credit Assignment in RL Training of LLMs

Modalities Contribute Unequally: Enhancing Medical Multi-modal Learning through Adaptive Modality Token Re-balancing

Interchangeable Token Embeddings for Extendable Vocabulary and Alpha-Equivalence

End-to-End Learning Framework for Solving Non-Markovian Optimal Control

Liger: Linearizing Large Language Models to Gated Recurrent Structures

Scaling Video-Language Models to 10K Frames via Hierarchical Differential Distillation

DLP: Dynamic Layerwise Pruning in Large Language Models

ReinboT: Amplifying Robot Visual-Language Manipulation with Reinforcement Learning

Leveraging Per-Instance Privacy for Machine Unlearning

Learning Strategic Language Agents in the Werewolf Game with Iterative Latent Space Policy Optimization

NMA-tune: Generating Highly Designable and Dynamics Aware Protein Backbones

Distributional Diffusion Models with Scoring Rules

Identifying biological perturbation targets through causal differential networks

Floating-Point Neural Networks Can Represent Almost All Floating-Point Functions

Reasoning Limitations of Multimodal Large Language Models. A case study of Bongard Problems

Training Diffusion-based Generative Models with Limited Data

Finite-Sample Convergence Bounds for Trust Region Policy Optimization in Mean Field Games

Scalable Private Partition Selection via Adaptive Weighting

Self-Organizing Visual Prototypes for Non-Parametric Representation Learning

MGD$^3$ : Mode-Guided Dataset Distillation using Diffusion Models

Adaptive kernel predictors from feature-learning infinite limits of neural networks

Guided Search Strategies in Non-Serializable Environments with Applications to Software Engineering Agents

Towards Universal Offline Black-Box Optimization via Learning Language Model Embeddings

HarmoniCa: Harmonizing Training and Inference for Better Feature Caching in Diffusion Transformer Acceleration

Scalable Meta-Learning via Mixed-Mode Differentiation

CogReact: A Reinforced Framework to Model Human Cognitive Reaction Modulated by Dynamic Intervention

Online Laplacian-Based Representation Learning in Reinforcement Learning

Action-Constrained Imitation Learning

Understanding Model Reprogramming for CLIP via Decoupling Visual Prompts

Mitigating Over-Squashing in Graph Neural Networks by Spectrum-Preserving Sparsification

Agent-as-a-Judge: Evaluate Agents with Agents

Sparse Spectral Training and Inference on Euclidean and Hyperbolic Neural Networks

MARS: Unleashing the Power of Variance Reduction for Training Large Models

Balancing Preservation and Modification: A Region and Semantic Aware Metric for Instruction-Based Image Editing

Scalable Gaussian Processes with Latent Kronecker Structure

Understanding the Logic of Direct Preference Alignment through Logic

A Comprehensive Framework for Analyzing the Convergence of Adam: Bridging the Gap with SGD

TinyMIG: Transferring Generalization from Vision Foundation Models to Single-Domain Medical Imaging

Bridging Fairness and Efficiency in Conformal Inference: A Surrogate-Assisted Group-Clustered Approach

PipeOffload: Improving Scalability of Pipeline Parallelism with Memory Optimization

Gridded Transformer Neural Processes for Spatio-Temporal Data

Accurate Identification of Communication Between Multiple Interacting Neural Populations

SafeMap: Robust HD Map Construction from Incomplete Observations

Optimal Transport Barycenter via Nonconvex-Concave Minimax Optimization

Differentially Private Space-Efficient Algorithms for Counting Distinct Elements in the Turnstile Model

CurvGAD: Leveraging Curvature for Enhanced Graph Anomaly Detection

When Do LLMs Help With Node Classification? A Comprehensive Analysis

Local Identifying Causal Relations in the Presence of Latent Variables

The dark side of the forces: assessing non-conservative force models for atomistic machine learning

WyckoffDiff -- A Generative Diffusion Model for Crystal Symmetry

Direct Discriminative Optimization: Your Likelihood-Based Visual Generative Model is Secretly a GAN Discriminator

Fundamental limits of learning in sequence multi-index models and deep attention networks: high-dimensional asymptotics and sharp thresholds

Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark

Physics-informed Temporal Alignment for Auto-regressive PDE Foundation Models

Demystifying Long Chain-of-Thought Reasoning

Unified Breakdown Analysis for Byzantine Robust Gossip

QT-DoG: Quantization-Aware Training for Domain Generalization

AMPO: Active Multi Preference Optimization for Self-play Preference Selection

AdaSplash: Adaptive Sparse Flash Attention

One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Framework

Algorithms with Calibrated Machine Learning Predictions

Task-Gated Multi-Expert Collaboration Network for Degraded Multi-Modal Image Fusion

Reinforced Learning Explicit Circuit Representations for Quantum State Characterization from Local Measurements

Q-VDiT: Towards Accurate Quantization and Distillation of Video-Generation Diffusion Transformers

Internal Causal Mechanisms Robustly Predict Language Model Out-of-Distribution Behaviors

An Efficient Private GPT Never Autoregressively Decodes

Towards Understanding Gradient Dynamics of the Sliced-Wasserstein Distance via Critical Point Analysis

Revisiting Continuity of Image Tokens for Cross-domain Few-shot Learning

Learn Singularly Perturbed Solutions via Homotopy Dynamics

Heterogeneous Treatment Effect in Time-to-Event Outcomes: Harnessing Censored Data with Recursively Imputed Trees

ExPLoRA: Parameter-Efficient Extended Pre-Training to Adapt Vision Transformers under Domain Shifts

BiAssemble: Learning Collaborative Affordance for Bimanual Geometric Assembly

Subgroups Matter for Robust Bias Mitigation

Taming Diffusion for Dataset Distillation with High Representativeness

Synthesizing Software Engineering Data in a Test-Driven Manner

Heterogeneous Label Shift: Theory and Algorithm

PILAF: Optimal Human Preference Sampling for Reward Modeling

Score-based Pullback Riemannian Geometry: Extracting the Data Manifold Geometry using Anisotropic Flows

Statistical Test for Feature Selection Pipelines by Selective Inference

Model Uncertainty Quantification by Conformal Prediction in Continual Learning

Inducing, Detecting and Characterising Neural Modules: A Pipeline for Functional Interpretability in Reinforcement Learning

Conformal Prediction as Bayesian Quadrature

SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language Models

FlexControl: Computation-Aware Conditional Control with Differentiable Router for Text-to-Image Generation

QEM-Bench: Benchmarking Learning-based Quantum Error Mitigation and QEMFormer as a Multi-ranged Context Learning Baseline

Adaptive Median Smoothing: Adversarial Defense for Unlearned Text-to-Image Diffusion Models at Inference Time

e-GAI: e-value-based Generalized $\alpha$-Investing for Online False Discovery Rate Control

How Much Can We Forget about Data Contamination?

De-coupled NeuroGF for Shortest Path Distance Approximations on Large Terrain Graphs

Explanatory Instructions: Towards Unified Vision Tasks Understanding and Zero-shot Generalization

Representative Ranking for Deliberation in the Public Sphere

Tuning Sequential Monte Carlo Samplers via Greedy Incremental Divergence Minimization

Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction

A Generalization Result for Convergence in Learning-to-Optimize

Conformity Score Averaging for Classification

M2PDE: Compositional Generative Multiphysics and Multi-component PDE Simulation

Trajectory World Models for Heterogeneous Environments

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning

SynEVO: A neuro-inspired spatiotemporal evolutional framework for cross-domain adaptation

Optimizing Social Network Interventions via Hypergradient-Based Recommender System Design

Revisiting the Predictability of Performative, Social Events

Model Steering: Learning with a Reference Model Improves Generalization Bounds and Scaling Laws

The Double-Ellipsoid Geometry of CLIP

Solving Linear-Gaussian Bayesian Inverse Problems with Decoupled Diffusion Sequential Monte Carlo

Evolving Minds: Logic-Informed Inference from Temporal Action Patterns

Mechanisms of Projective Composition of Diffusion Models

Generative Human Trajectory Recovery via Embedding-Space Conditional Diffusion

Componential Prompt-Knowledge Alignment for Domain Incremental Learning

REINFORCE Adversarial Attacks on Large Language Models: An Adaptive, Distributional, and Semantic Objective

Star Attention: Efficient LLM Inference over Long Sequences

Boosting Protein Graph Representations through Static-Dynamic Fusion

Hidden No More: Attacking and Defending Private Third-Party LLM Inference

Propagate and Inject: Revisiting Propagation-Based Feature Imputation for Graphs with Partially Observed Features

Causal-PIK: Causality-based Physical Reasoning with a Physics-Informed Kernel

Implicit Subgraph Neural Network

Federated Causal Structure Learning with Non-identical Variable Sets

Kona: An Efficient Privacy-Preservation Framework for KNN Classification by Communication Optimization

Is Best-of-N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment

Emergence in non-neural models: grokking modular arithmetic via average gradient outer product

ALMTokenizer: A Low-bitrate and Semantic-rich Audio Codec Tokenizer for Audio Language Modeling

"Who experiences large model decay and why?" A Hierarchical Framework for Diagnosing Heterogeneous Performance Drift

Improved and Oracle-Efficient Online $\ell_1$-Multicalibration

KBQA-o1: Agentic Knowledge Base Question Answering with Monte Carlo Tree Search

Synthetic Face Datasets Generation via Latent Space Exploration from Brownian Identity Diffusion

RollingQ: Reviving the Cooperation Dynamics in Multimodal Transformer

Editable Concept Bottleneck Models

Bongard in Wonderland: Visual Puzzles that Still Make AI Go Mad?

Theoretically Unmasking Inference Attacks Against LDP-Protected Clients in Federated Vision Models

Distributed Conformal Prediction via Message Passing

Zero-Shot Generalization of GNNs over Distinct Attribute Domains

Advancing Constrained Monotonic Neural Networks: Achieving Universal Approximation Beyond Bounded Activations

EpiCoder: Encompassing Diversity and Complexity in Code Generation

Categorical Schrödinger Bridge Matching

Generalization Performance of Ensemble Clustering: From Theory to Algorithm

Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage

PISA Experiments: Exploring Physics Post-Training for Video Diffusion Models by Watching Stuff Drop

Core Context Aware Transformers for Long Context Language Modeling

PINNsAgent: Automated PDE Surrogation with Large Language Models

Neural Representational Consistency Emerges from Probabilistic Neural-Behavioral Representation Alignment

Residual Matrix Transformers: Scaling the Size of the Residual Stream

Improving the Effective Receptive Field of Message-Passing Neural Networks

Adapting to Evolving Adversaries with Regularized Continual Robust Training

Enhancing Spectral GNNs: From Topology and Perturbation Perspectives

Puzzle: Distillation-Based NAS for Inference-Optimized LLMs

Sparse Causal Discovery with Generative Intervention for Unsupervised Graph Domain Adaptation

ConceptAttention: Diffusion Transformers Learn Highly Interpretable Features

Haste Makes Waste: A Simple Approach for Scaling Graph Neural Networks

SCISSOR: Mitigating Semantic Bias through Cluster-Aware Siamese Networks for Robust Classification

Ringmaster ASGD: The First Asynchronous SGD with Optimal Time Complexity

Instruct2See: Learning to Remove Any Obstructions Across Distributions

Customizing the Inductive Biases of Softmax Attention using Structured Matrices

Temperature-Annealed Boltzmann Generators

Improving Diversity in Language Models: When Temperature Fails, Change the Loss

Fast Inference with Kronecker-Sparse Matrices

MASS: Mathematical Data Selection via Skill Graphs for Pretraining Large Language Models

Adapting to Linear Separable Subsets with Large-Margin in Differentially Private Learning

DSBRouter: End-to-end Global Routing via Diffusion Schr\"{o}dinger Bridge

Chaos Meets Attention: Transformers for Large-Scale Dynamical Prediction

RocketKV: Accelerating Long-Context LLM Inference via Two-Stage KV Cache Compression

Stable Offline Value Function Learning with Bisimulation-based Representations

Hierarchical Reinforcement Learning with Uncertainty-Guided Diffusional Subgoals

TopoTune: A Framework for Generalized Combinatorial Complex Neural Networks

Directly Forecasting Belief for Reinforcement Learning with Delays

Provable Efficiency of Guidance in Diffusion Models for General Data Distribution

A Theoretical Framework For Overfitting In Energy-based Modeling

Analyze Feature Flow to Enhance Interpretation and Steering in Language Models

Simple Policy Optimization

STD-FD: Spatio-Temporal Distribution Fitting Deviation for AIGC Forgery Identification

KIND: Knowledge Integration and Diversion for Training Decomposable Models

Breaking the Barrier of Hard Samples: A Data-Centric Approach to Synthetic Data for Medical Tasks

DANCE: Dual Unbiased Expansion with Group-acquired Alignment for Out-of-distribution Graph Fairness Learning

On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domains

Learning Distances from Data with Normalizing Flows and Score Matching

Robust Multimodal Large Language Models Against Modality Conflict

The Case for Learned Provenance-based System Behavior Baseline

Average Certified Radius is a Poor Metric for Randomized Smoothing

Finding Wasserstein Ball Center: Efficient Algorithm and The Applications in Fairness

Time Series Representations with Hard-Coded Invariances

RepLoRA: Reparameterizing Low-rank Adaptation via the Perspective of Mixture of Experts

Differential Privacy Under Class Imbalance: Methods and Empirical Insights

Synthesizing Images on Perceptual Boundaries of ANNs for Uncovering and Manipulating Human Perceptual Variability

The Perils of Optimizing Learned Reward Functions: Low Training Error Does Not Guarantee Low Regret

PokéChamp: an Expert-level Minimax Language Agent

Off-Policy Evaluation under Nonignorable Missing Data

Simultaneous Multi-Robot Motion Planning with Projected Diffusion Models

No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces

NETS: A Non-equilibrium Transport Sampler

LIVS: A Pluralistic Alignment Dataset for Inclusive Public Spaces

AdaWorld: Learning Adaptable World Models with Latent Actions

AI for Global Climate Cooperation: Modeling Global Climate Negotiations, Agreements, and Long-Term Cooperation in RICE-N

A Theory for Conditional Generative Modeling on Multiple Data Sources

Lower Bounds for Chain-of-Thought Reasoning in Hard-Attention Transformers

Towards Theoretical Understanding of Sequential Decision Making with Preference Feedback

An Efficient Matrix Multiplication Algorithm for Accelerating Inference in Binary and Ternary Neural Networks

Tree-Sliced Wasserstein Distance: A Geometric Perspective

ResKoopNet: Learning Koopman Representations for Complex Dynamics with Spectral Residuals

Tree-Sliced Wasserstein Distance with Nonlinear Projection

LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Models

Robot-Gated Interactive Imitation Learning with Adaptive Intervention Mechanism

Enhancing Cooperative Multi-Agent Reinforcement Learning with State Modelling and Adversarial Exploration

Tokenized Bandit for LLM Decoding and Alignment

Retrieval-Augmented Language Model for Knowledge-aware Protein Encoding

ROS: A GNN-based Relax-Optimize-and-Sample Framework for Max-$k$-Cut Problems

TGDPO: Harnessing Token-Level Reward Guidance for Enhancing Direct Preference Optimization

Learning Latent Graph Structures and their Uncertainty

Privacy Amplification Through Synthetic Data: Insights from Linear Regression

Layer-wise Quantization for Quantized Optimistic Dual Averaging

Unlocking the Power of SAM 2 for Few-Shot Segmentation

A Physics-Augmented Deep Learning Framework for Classifying Single Molecule Force Spectroscopy Data

Stacey: Promoting Stochastic Steepest Descent via Accelerated $\ell_p$-Smooth Nonconvex Optimization

CLIMB: Data Foundations for Large Scale Multimodal Clinical Foundation Models

Non-stationary Online Learning for Curved Losses: Improved Dynamic Regret via Mixability

Measuring Variable Importance in Heterogeneous Treatment Effects with Confidence

GLGENN: A Novel Parameter-Light Equivariant Neural Networks Architecture Based on Clifford Geometric Algebras

Proxy-FDA: Proxy-based Feature Distribution Alignment for Fine-tuning Vision Foundation Models without Forgetting

Discrepancy Minimization in Input-Sparsity Time

LAST SToP for Modeling Asynchronous Time Series

The Emperor's New Clothes in Benchmarking? A Rigorous Examination of Mitigation Strategies for LLM Benchmark Data Contamination

Ergodic Generative Flows

Griffin: Towards a Graph-Centric Relational Database Foundation Model

Fast Exact Unlearning for In-Context Learning Data for LLMs

Understanding the difficulties of posterior predictive estimation

Learning Curves of Stochastic Gradient Descent in Kernel Regression

Scalable Equilibrium Sampling with Sequential Boltzmann Generators

UnHiPPO: Uncertainty-aware Initialization for State Space Models

Preference learning made easy: Everything should be understood through win rate

Inverse Bridge Matching Distillation

Super Deep Contrastive Information Bottleneck for Multi-modal Clustering

PTTA: Purifying Malicious Samples for Test-Time Model Adaptation

Identifying Causal Direction via Variational Bayesian Compression

Dialogue Without Limits: Constant-Sized KV Caches for Extended Response in LLMs

am-ELO: A Stable Framework for Arena-based LLM Evaluation

The Complexity of Learning Sparse Superposed Features with Feedback

Logits are All We Need to Adapt Closed Models

Reward-Guided Prompt Evolving in Reinforcement Learning for LLMs

Best Subset Selection: Optimal Pursuit for Feature Selection and Elimination

Leveraging Model Guidance to Extract Training Data from Personalized Diffusion Models

A Versatile Influence Function for Data Attribution with Non-Decomposable Loss

Distribution-aware Fairness Learning in Medical Image Segmentation From A Control-Theoretic Perspective

Eliciting Language Model Behaviors with Investigator Agents

Teaching Transformers Causal Reasoning through Axiomatic Training

LaCache: Ladder-Shaped KV Caching for Efficient Long-Context Modeling of Large Language Models

Introducing 3D Representation for Dense Volume-to-Volume Translation via Score Fusion

Hyperbolic-PDE GNN: Spectral Graph Neural Networks in the Perspective of A System of Hyperbolic Partial Differential Equations

Anytime-Constrained Equilibria in Polynomial Time

Flexibility-conditioned protein structure design with flow matching

Fully Heteroscedastic Count Regression with Deep Double Poisson Networks

Probably Approximately Global Robustness Certification

Natural Perturbations for Black-box Training of Neural Networks by Zeroth-Order Optimization

FAB-PPI: Frequentist, Assisted by Bayes, Prediction-Powered Inference

Imagine While Reasoning in Space: Multimodal Visualization-of-Thought

Global Optimization with a Power-Transformed Objective and Gaussian Smoothing

EduLLM: Leveraging Large Language Models and Framelet-Based Signed Hypergraph Neural Networks for Student Performance Prediction

MM-RLHF: The Next Step Forward in Multimodal LLM Alignment

SDP-CROWN: Efficient Bound Propagation for Neural Network Verification with Tightness of Semidefinite Programming

MF-LAL: Drug Compound Generation Using Multi-Fidelity Latent Space Active Learning

COGNATE: Acceleration of Sparse Tensor Programs on Emerging Hardware using Transfer Learning

Learning Compact Semantic Information for Incomplete Multi-View Missing Multi-Label Classification

Do Not Mimic My Voice : Speaker Identity Unlearning for Zero-Shot Text-to-Speech

Taming Knowledge Conflicts in Language Models

Universal Length Generalization with Turing Programs

Large Language-Geometry Model: When LLM meets Equivariance

Automatic Reward Shaping from Confounded Offline Data

RE-IMAGINE: Symbolic Benchmark Synthesis for Reasoning Evaluation

Adapter Naturally Serves as Decoupler for Cross-Domain Few-Shot Semantic Segmentation

EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption

Smooth Interpolation for Improved Discrete Graph Generative Models

Collaborative Mean Estimation Among Heterogeneous Strategic Agents: Individual Rationality, Fairness, and Truthful Contribution

Learning In-context $n$-grams with Transformers: Sub-$n$-grams Are Near-Stationary Points

RestoreGrad: Signal Restoration Using Conditional Denoising Diffusion Models with Jointly Learned Prior

Policy Guided Tree Search for Enhanced LLM Reasoning

Do NOT Think That Much for 2+3=? On the Overthinking of Long Reasoning Models

G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks

Self-Disentanglement and Re-Composition for Cross-Domain Few-Shot Segmentation

Accelerating Linear Recurrent Neural Networks for the Edge with Unstructured Sparsity

xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference

Improving LLM Safety Alignment with Dual-Objective Optimization

The Generalized Skew Spectrum of Graphs

Mastering Board Games by External and Internal Planning with Language Models

SKIM: Any-bit Quantization Pushing The Limits of Post-Training Quantization

DUNIA: Pixel-Sized Embeddings via Cross-Modal Alignment for Earth Observation Applications

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks

Foundation Molecular Grammar: Multi-Modal Foundation Models Induce Interpretable Molecular Graph Languages

Hessian Geometry of Latent Space in Generative Models

Noise Conditional Variational Score Distillation

KGMark: A Diffusion Watermark for Knowledge Graphs

Faster Global Minimum Cut with Predictions

PepTune: De Novo Generation of Therapeutic Peptides with Multi-Objective-Guided Discrete Diffusion

Multimodal Medical Code Tokenizer

Predicting mutational effects on protein binding from folding energy

UGPhysics: A Comprehensive Benchmark for Undergraduate Physics Reasoning with Large Language Models

Designing Cyclic Peptides via Harmonic SDE with Atom-Bond Modeling

One-Step Diffusion Policy: Fast Visuomotor Policies via Diffusion Distillation

BARK: A Fully Bayesian Tree Kernel for Black-box Optimization

An Empirical Study on Configuring In-Context Learning Demonstrations for Unleashing MLLMs' Sentimental Perception Capability

Learning Input Encodings for Kernel-Optimal Implicit Neural Representations

EARTH: Epidemiology-Aware Neural ODE with Continuous Disease Transmission Graph

Empower Structure-Based Molecule Optimization with Gradient Guided Bayesian Flow Networks

ParallelComp: Parallel Long-Context Compressor for Length Extrapolation

DCBM: Data-Efficient Visual Concept Bottleneck Models

Wait-Less Offline Tuning and Re-solving for Online Decision Making

Distributed Retraction-Free and Communication-Efficient Optimization on the Stiefel Manifold

CoSER: Coordinating LLM-Based Persona Simulation of Established Roles

A Sub-Problem Quantum Alternating Operator Ansatz for Correlation Clustering

Safety Reasoning with Guidelines

Training a Generally Curious Agent

MA-LoT: Model-Collaboration Lean-based Long Chain-of-Thought Reasoning enhances Formal Theorem Proving

Efficient Core-set Selection for Deep Learning Through Squared Loss Minimization

How Transformers Learn Structured Data: Insights From Hierarchical Filtering

Token Signature: Predicting Chain-of-Thought Gains with Token Decoding Feature in Large Language Models

Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment

Towards the Causal Complete Cause of Multi-Modal Representation Learning

GIVE: Structured Reasoning of Large Language Models with Knowledge Graph Inspired Veracity Extrapolation

Compressing tree ensembles through Level-wise Optimization and Pruning

Free Process Rewards without Process Labels

OmiAD: One-Step Adaptive Masked Diffusion Model for Multi-class Anomaly Detection via Adversarial Distillation

Disentangled Graph Spectral Domain Adaptation

OmniArch: Building Foundation Model for Scientific Computing

SafeArena: Evaluating the Safety of Autonomous Web Agents

Joint Localization and Activation Editing for Low-Resource Fine-Tuning

Enhancing Foundation Models with Federated Domain Knowledge Infusion

Interpreting CLIP with Hierarchical Sparse Autoencoders

Efficient Skill Discovery via Regret-Aware Optimization

Human Cognition-Inspired Hierarchical Fuzzy Learning Machine

Understanding the Limits of Deep Tabular Methods with Temporal Shift

Reward Modeling with Ordinal Feedback: Wisdom of the Crowd

TabSDS: a Lightweight, Fully Non-Parametric, and Model Free Approach for Generating Synthetic Tabular Data

Mutual Learning for SAM Adaptation: A Dual Collaborative Network Framework for Source-Free Domain Transfer

La RoSA: Enhancing LLM Efficiency via Layerwise Rotated Sparse Activation

Upweighting Easy Samples in Fine-Tuning Mitigates Forgetting

BoA: Attention-aware Post-training Quantization without Backpropagation

MoEQuant: Enhancing Quantization for Mixture-of-Experts Large Language Models via Expert-Balanced Sampling and Affinity Guidance

Autonomy-of-Experts Models

Maximizing Intermediate Checkpoint Value in LLM Pretraining with Bayesian Optimization

Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settings

FedSMU: Communication-Efficient and Generalization-Enhanced Federated Learning through Symbolic Model Updates

Explaining the role of Intrinsic Dimensionality in Adversarial Training

Loss Functions and Operators Generated by f-Divergences

LotteryCodec: Searching the Implicit Representation in a Random Network for Low-Complexity Image Compression

Score as Action: Fine Tuning Diffusion Generative Models by Continuous-time Reinforcement Learning

Aligning with Logic: Measuring, Evaluating and Improving Logical Preference Consistency in Large Language Models

FIC-TSC: Learning Time Series Classification with Fisher Information Constraint

AutoAL: Automated Active Learning with Differentiable Query Strategy Search

Regress, Don't Guess: A Regression-like Loss on Number Tokens for Language Models

Reaction Graph: Towards Reaction-Level Modeling for Chemical Reactions with 3D Structures

Fairness on Principal Stratum: A New Perspective on Counterfactual Fairness

Sub-Sequential Physics-Informed Learning with State Space Model

Finite-Time Analysis of Discrete-Time Stochastic Interpolants

Improved Learning via k-DTW: A Novel Dissimilarity Measure for Curves

Beyond Self-Repellent Kernels: History-Driven Target Towards Efficient Nonlinear MCMC on General Graphs

Computing Voting Rules with Improvement Feedback

Online Linear Classification with Massart Noise

MuseControlLite: Multifunctional Music Generation with Lightweight Conditioners

The Four Color Theorem for Cell Instance Segmentation

Info-Coevolution: An Efficient Framework for Data Model Coevolution

Maximum Entropy Reinforcement Learning with Diffusion Policy

Trajectory Inference with Smooth Schrödinger Bridges

Mirror, Mirror of the Flow: How Does Regularization Shape Implicit Bias?

Symmetry-Driven Discovery of Dynamical Variables in Molecular Simulations

R3DM: Enabling Role Discovery and Diversity Through Dynamics Models in Multi-agent Reinforcement Learning

Learning Monotonic Probabilities with a Generative Cost Model

Attributes Shape the Embedding Space of Face Recognition Models

DiffusionVLA: Scaling Robot Foundation Models via Unified Diffusion and Autoregression

Probabilistic Factorial Experimental Design for Combinatorial Interventions

Evaluating Judges as Evaluators: The JETTS Benchmark of LLM-as-Judges as Test-Time Scaling Evaluators

Trust-Region Twisted Policy Improvement

LangTime: A Language-Guided Unified Model for Time Series Forecasting with Proximal Policy Optimization

Expressive Score-Based Priors for Distribution Matching with Geometry-Preserving Regularization

CASE-Bench: Context-Aware SafEty Benchmark for Large Language Models

Scalable Sobolev IPM for Probability Measures on a Graph

A Forget-and-Grow Strategy for Deep Reinforcement Learning Scaling in Continuous Control

Variational Rectified Flow Matching

Retrieval Augmented Time Series Forecasting

AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence

Polynomial-Delay MAG Listing with Novel Locally Complete Orientation Rules

Catch Your Emotion: Sharpening Emotion Perception in Multimodal Large Language Models

Efficiently Vectorized MCMC on Modern Accelerators

Learning Smooth and Expressive Interatomic Potentials for Physical Property Prediction

PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis

AdaDecode: Accelerating LLM Decoding with Adaptive Layer Parallelism

Think Smarter not Harder: Adaptive Reasoning with Inference Aware Optimization

Understanding Nonlinear Implicit Bias via Region Counts in Input Space

Gumiho: A Hybrid Architecture to Prioritize Early Tokens in Speculative Decoding

Optimal Auction Design in the Joint Advertising

Understanding Fixed Predictions via Confined Regions

Going Deeper into Locally Differentially Private Graph Neural Networks

Normalizing Flows are Capable Generative Models

SlimLLM: Accurate Structured Pruning for Large Language Models

Improving LLM Video Understanding with 16 Frames Per Second

AlphaQCM: Alpha Discovery in Finance with Distributional Reinforcement Learning

Statistical Collusion by Collectives on Learning Platforms

TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting

Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale

Attention-Level Speculation

Revolve: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization

Implicit Language Models are RNNs: Balancing Parallelization and Expressivity

NTPP: Generative Speech Language Modeling for Dual-Channel Spoken Dialogue via Next-Token-Pair Prediction

SpikF: Spiking Fourier Network for Efficient Long-term Prediction

Pruning for GNNs: Lower Complexity with Comparable Expressiveness

Partition First, Embed Later: Laplacian-Based Feature Partitioning for Refined Embedding and Visualization of High-Dimensional Data

Adjoint Sampling: Highly Scalable Diffusion Samplers via Adjoint Matching

Geometric and Physical Constraints Synergistically Enhance Neural PDE Surrogates

Thermalizer: Stable autoregressive neural emulation of spatiotemporal chaos

The Sparse-Plus-Low-Rank Quasi-Newton Method for Entropic-Regularized Optimal Transport

When Diffusion Models Memorize: Inductive Biases in Probability Flow of Minimum-Norm Shallow Neural Nets

Maximum Total Correlation Reinforcement Learning

Progressively Label Enhancement for Large Language Model Alignment

A Generalizable Physics-Enhanced State Space Model for Long-Term Dynamics Forecasting in Complex Environments

Reward-Guided Iterative Refinement in Diffusion Models at Test-Time with Applications to Protein and DNA Design

Learning Distribution-wise Control in Representation Space for Language Models

A Unified Approach to Routing and Cascading for LLMs

MetricEmbedding: Accelerate Metric Nearness by Tropical Inner Product

SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training

ResearchTown: Simulator of Human Research Community

Sample Complexity of Branch-length Estimation by Maximum Likelihood

Gaussian Mixture Flow Matching Models

Scaling Laws for Differentially Private Language Models

Layer by Layer: Uncovering Hidden Representations in Language Models

Universal Approximation Theorem of Deep Q-Networks

Online Clustering of Dueling Bandits

WATCH: Adaptive Monitoring for AI Deployments via Weighted-Conformal Martingales

Protriever: End-to-End Differentiable Protein Homology Search for Fitness Prediction

Mixture of Hidden-Dimensions: Not All Hidden-States’ Dimensions are Needed in Transformer

Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction

ExLM: Rethinking the Impact of $\texttt{[MASK]}$ Tokens in Masked Language Models

Feature Importance Metrics in the Presence of Missing Data

The Relationship Between No-Regret Learning and Online Conformal Prediction

Learning the Electronic Hamiltonian of Large Atomic Structures

Sort Before You Prune: Improved Worst-Case Guarantees of the DiskANN Family of Graphs

Online Learning in the Random-Order Model

AnalogGenie-Lite: Enhancing Scalability and Precision in Circuit Topology Discovery through Lightweight Graph Modeling

Should Decision-Makers Reveal Classifiers in Online Strategic Classification?

Avoiding Catastrophe in Online Learning by Asking for Help

Navigating Semantic Drift in Task-Agnostic Class-Incremental Learning

Impossible Videos

BackSlash: Rate Constrained Optimized Training of Large Language Models

Stability and Generalization Capability of Subgraph Reasoning Models for Inductive Knowledge Graph Completion

MERIT: Maximum-normalized Element-wise Ratio for Language Model Large-batch Training

Provable and Practical Online Learning Rate Adaptation with Hypergradient Descent

DMM: Distributed Matrix Mechanism for Differentially-Private Federated Learning Based on Constant-Overhead Linear Secret Resharing

DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference Optimization

Token Coordinated Prompt Attention is Needed for Visual Prompting

Diversified Flow Matching with Translation Identifiability

Emoji Attack: Enhancing Jailbreak Attacks Against Judge LLM Detection

Edge-Colored Clustering in Hypergraphs: Beyond Minimizing Unsatisfied Edges

GPTAQ: Efficient Finetuning-Free Quantization for Asymmetric Calibration

Rapid Overfitting of Multi-Pass SGD in Stochastic Convex Optimization

Blink of an eye: a simple theory for feature localization in generative models

Bivariate Causal Discovery with Proxy Variables: Integral Solving and Beyond

Selective Preference Aggregation

Softmax is not Enough (for Sharp Size Generalisation)

Selective Response Strategies for GenAI

PhySpec: Physically Consistent Spectral Reconstruction via Orthogonal Subspace Decomposition and Self-Supervised Meta-Auxiliary Learning

DocKS-RAG: Optimizing Document-Level Relation Extraction through LLM-Enhanced Hybrid Prompt Tuning

Does Graph Prompt Work? A Data Operation Perspective with Theoretical Analysis

Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts

Improving Zero-Shot Adversarial Robustness in Vision-Language Models by Closed-form Alignment of Adversarial Path Simplices

Adapting Precomputed Features for Efficient Graph Condensation

Aequa: Fair Model Rewards in Collaborative Learning via Slimmable Networks

SE(3)-Equivariant Diffusion Policy in Spherical Fourier Space

Achieving Linear Speedup and Near-Optimal Complexity for Decentralized Optimization over Row-stochastic Networks

FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Making

Teaching Language Models to Critique via Reinforcement Learning

BSO: Binary Spiking Online Optimization Algorithm

Permutation-based Rank Test in the Presence of Discretization and Application in Causal Discovery with Mixed Data

The Hidden Joules: Evaluating the Energy Consumption of Vision Backbones for Progress Towards More Efficient Model Inference

Communicating Activations Between Language Model Agents

Compute or Load KV Cache? Why Not Both?

D-Fusion: Direct Preference Optimization for Aligning Diffusion Models with Visually Consistent Samples

Approximating Latent Manifolds in Neural Networks via Vanishing Ideals

Hybrid Spiking Vision Transformer for Object Detection with Event Cameras

Generating Hypotheses of Dynamic Causal Graphs in Neuroscience: Leveraging Generative Factor Models of Observed Time Series

Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing

HealthGPT: A Medical Large Vision-Language Model for Unifying Comprehension and Generation via Heterogeneous Knowledge Adaptation

TabNAT: A Continuous-Discrete Joint Generative Framework for Tabular Data

Optimal and Practical Batched Linear Bandit Algorithm

Beyond Atoms: Enhancing Molecular Pretrained Representations with 3D Space Modeling

Broadband Ground Motion Synthesis by Diffusion Model with Minimal Condition

Open-Det: An Efficient Learning Framework for Open-Ended Detection

Tracking The Best Expert Privately

Efficient and Privacy-Preserving Soft Prompt Transfer for LLMs

Diffusion Counterfactual Generation with Semantic Abduction

Stochastic Forward–Backward Deconvolution: Training Diffusion Models with Finite Noisy Datasets

Scaling Laws for Task-Optimized Models of the Primate Visual Ventral Stream

Faster and Stronger: When ANN-SNN Conversion Meets Parallel Spiking Calculation

Diving into Self-Evolving Training for Multimodal Reasoning

Feature Shift Localization Network

Learning Safe Control via On-the-Fly Bandit Exploration

When to retrain a machine learning model

UniMate: A Unified Model for Mechanical Metamaterial Generation, Property Prediction, and Condition Confirmation

Retrieval-Augmented Perception: High-resolution Image Perception Meets Visual RAG

Non-Asymptotic and Non-Lipschitzian Bounds on Optimal Values in Stochastic Optimization Under Heavy Tails

BAME: Block-Aware Mask Evolution for Efficient N:M Sparse Training

Shielded Diffusion: Generating Novel and Diverse Images using Sparse Repellency

DeepLayout: Learning Neural Representations of Circuit Placement Layout

Speculate, then Collaborate: Fusing Knowledge of Language Models during Decoding

Convergence Analysis of Policy Gradient Methods with Dynamic Stochasticity

PIPA: Preference Alignment as Prior-Informed Statistical Estimation

A Non-Asymptotic Convergent Analysis for Scored-Based Graph Generative Model via a System of Stochastic Differential Equations

LIFT the Veil for the Truth: Principal Weights Emerge after Rank Reduction for Reasoning-Focused Supervised Fine-Tuning

Habitizing Diffusion Planning for Efficient and Effective Decision Making

Fast Tensor Completion via Approximate Richardson Iteration

K$^2$IE: Kernel Method-based Kernel Intensity Estimators for Inhomogeneous Poisson Processes

Consensus Is All You Get: The Role of Attention in Transformers

Large Continual Instruction Assistant

EARL-BO: Reinforcement Learning for Multi-Step Lookahead, High-Dimensional Bayesian Optimization

SUICA: Learning Super-high Dimensional Sparse Implicit Neural Representations for Spatial Transcriptomics

Learning the RoPEs: Better 2D and 3D Position Encodings with STRING

RelGNN: Composite Message Passing for Relational Deep Learning

No Metric to Rule Them All: Toward Principled Evaluations of Graph-Learning Datasets

Speeding up Policy Simulation in Supply Chain RL

Quadruple Attention in Many-body Systems for Accurate Molecular Property Predictions

SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting

FunBO: Discovering Acquisition Functions for Bayesian Optimization with FunSearch

Arbitrarily-Conditioned Multi-Functional Diffusion for Multi-Physics Emulation

Bayesian Neural Scaling Law Extrapolation with Prior-Data Fitted Networks

A Memory Efficient Randomized Subspace Optimization Method for Training Large Language Models

Test-time Adapted Reinforcement Learning with Action Entropy Regularization

Beyond Induction Heads: In-Context Meta Learning Induces Multi-Phase Circuit Emergence

LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding

Activation by Interval-wise Dropout: A Simple Way to Prevent Neural Networks from Plasticity Loss

LaMAGIC2: Advanced Circuit Formulations for Language Model-Based Analog Topology Generation

Weakly-Supervised Contrastive Learning for Imprecise Class Labels

Efficient Time Series Processing for Transformers and State-Space Models through Token Merging

Latent Mamba Operator for Partial Differential Equations

Maintaining Proportional Committees with Dynamic Candidate Sets

Gandalf the Red: Adaptive Security for LLMs

OV-MER: Towards Open-Vocabulary Multimodal Emotion Recognition

Improving Compositional Generation with Diffusion Models Using Lift Scores

Unified K-Means Clustering with Label-Guided Manifold Learning

Rethink the Role of Deep Learning towards Large-scale Quantum Systems

GaussMark: A Practical Approach for Structural Watermarking of Language Models

Robust and Conjugate Spatio-Temporal Gaussian Processes

RobustLight: Improving Robustness via Diffusion Reinforcement Learning for Traffic Signal Control

LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models

Adaptive Flow Matching for Resolving Small-Scale Physics

Exact Upper and Lower Bounds for the Output Distribution of Neural Networks with Random Inputs

Quantifying Memory Utilization with Effective State-Size

When can in-context learning generalize out of task distribution?

Fixing the Double Penalty in Data-Driven Weather Forecasting Through a Modified Spherical Harmonic Loss Function

EcoMapper: Generative Modeling for Climate-Aware Satellite Imagery

Imitation Learning from a Single Temporally Misaligned Video

Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimation

Adaptive Message Passing: A General Framework to Mitigate Oversmoothing, Oversquashing, and Underreaching

MME-CoT: Benchmarking Chain-of-Thought in Large Multimodal Models for Reasoning Quality, Robustness, and Efficiency

Thickness-aware E(3)-Equivariant 3D Mesh Neural Networks

Disentangling Invariant Subgraph via Variance Contrastive Estimation under Distribution Shifts

TUMTraf VideoQA: Dataset and Benchmark for Unified Spatio-Temporal Video Understanding in Traffic Scenes

Symmetric Reinforcement Learning Loss for Robust Learning on Diverse Tasks and Model Scales

ProSec: Fortifying Code LLMs with Proactive Security Alignment

Principled Algorithms for Optimizing Generalized Metrics in Binary Classification

Perception in Reflection

Catching Two Birds with One Stone: Reward Shaping with Dual Random Networks for Balancing Exploration and Exploitation

Incorporating Arbitrary Matrix Group Equivariance into KANs

Not all solutions are created equal: An analytical dissociation of functional and representational similarity in deep linear neural networks

On the Emergence of Position Bias in Transformers

PASS: Private Attributes Protection with Stochastic Data Substitution

Scaling Value Iteration Networks to 5000 Layers for Extreme Long-Term Planning

Variational Control for Guidance in Diffusion Models

Isolated Causal Effects of Natural Language

Calibrated Physics-Informed Uncertainty Quantification

Settling the Maximin Share Fairness for Scheduling among Groups of Machines

Flexible Tails for Normalizing Flows

CLOVER: Cross-Layer Orthogonal Vectors Pruning

ETTA: Elucidating the Design Space of Text-to-Audio Models

BounDr.E: Predicting Drug-likeness via Biomedical Knowledge Alignment and EM-like One-Class Boundary Optimization

Equivalence is All: A Unified View for Self-supervised Graph Learning

Backdoor Attacks in Token Selection of Attention Mechanism

Prediction via Shapley Value Regression

ConfPO: Exploiting Policy Model Confidence for Critical Token Selection in Preference Optimization

Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence

Improved Sample Complexity for Private Nonsmooth Nonconvex Optimization

Flow-field inference from neural data using deep recurrent networks

Multi-Marginal Stochastic Flow Matching for High-Dimensional Snapshot Data at Irregular Time Points

Physics Aware Neural Networks for Unsupervised Binding Energy Prediction

Drug-TTA: Test-Time Adaptation for Drug Virtual Screening via Multi-task Meta-Auxiliary Learning

Volume-Aware Distance for Robust Similarity Learning

Concept-Based Unsupervised Domain Adaptation

Decoupled SGDA for Games with Intermittent Strategy Communication

EvoMesh: Adaptive Physical Simulation with Hierarchical Graph Evolutions

GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidance

Goal-Oriented Skill Abstraction for Offline Multi-Task Reinforcement Learning

Fourier Position Embedding: Enhancing Attention’s Periodic Extension for Length Generalization

Learning to Keep a Promise: Scaling Language Model Decoding Parallelism with Learned Asynchronous Decoding

Come Together, But Not Right Now: A Progressive Strategy to Boost Low-Rank Adaptation

MoHAVE: Mixture of Hierarchical Audio-Visual Experts for Robust Speech Recognition

Quamba2: A Robust and Scalable Post-training Quantization Framework for Selective State Space Models

Harnessing Heterogeneous Statistical Strength for Personalized Federated Learning via Hierarchical Bayesian Inference

Neural Collapse Beyond the Unconstrained Features Model: Landscape, Dynamics, and Generalization in the Mean-Field Regime

Private Federated Learning using Preference-Optimized Synthetic Data

In-Context Learning as Conditioned Associative Memory Retrieval

BiMark: Unbiased Multilayer Watermarking for Large Language Models

Unlocking Post-hoc Dataset Inference with Synthetic Data

Stochastic Smoothed Primal-Dual Algorithms for Nonconvex Optimization with Linear Inequality Constraints

Verification Learning: Make Unsupervised Neuro-Symbolic System Feasible

Efficient Motion Prompt Learning for Robust Visual Tracking

Training Dynamics of In-Context Learning in Linear Attention

STAIR: Improving Safety Alignment with Introspective Reasoning

Long-Term TalkingFace Generation via Motion-Prior Conditional Diffusion Model

Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

Learning Minimum-Size BDDs: Towards Efficient Exact Algorithms

Differentiable Quadratic Optimization For the Maximum Independent Set Problem

Gradual Transition from Bellman Optimality Operator to Bellman Operator in Online Reinforcement Learning

Resolving Lexical Bias in Model Editing

On The Concurrence of Layer-wise Preconditioning Methods and Provable Feature Learning

Transformative or Conservative? Conservation laws for ResNets and Transformers

Be a Goldfish: Forgetting Bad Conditioning in Sparse Linear Regression via Variational Autoencoders

Unveiling AI's Blind Spots: An Oracle for In-Domain, Out-of-Domain, and Adversarial Errors

Adversarial Reasoning at Jailbreaking Time

Calibrating Video Watch-time Predictions with Credible Prototype Alignment

TabFSBench: Tabular Benchmark for Feature Shifts in Open Environments

Enhancing the Influence of Labels on Unlabeled Nodes in Graph Convolutional Networks

Non-stationary Diffusion For Probabilistic Time Series Forecasting

Lightweight-Mark: Rethinking Deep Learning-Based Watermarking

ROME is Forged in Adversity: Robust Distilled Datasets via Information Bottleneck

Programming Every Example: Lifting Pre-training Data Quality Like Experts at Scale

BCE vs. CE in Deep Feature Learning

The Panaceas for Improving Low-Rank Decomposition in Communication-Efficient Federated Learning

Discriminative Policy Optimization for Token-Level Reward Models

The Surprising Effectiveness of Test-Time Training for Few-Shot Learning

On Zero-Initialized Attention: Optimal Prompt and Gating Factor Estimation

Continuous Visual Autoregressive Generation via Score Maximization

Otter: Generating Tests from Issues to Validate SWE Patches

IRBridge: Solving Image Restoration Bridge with Pre-trained Generative Diffusion Models

Time-VLM: Exploring Multimodal Vision-Language Models for Augmented Time Series Forecasting

DragLoRA: Online Optimization of LoRA Adapters for Drag-based Image Editing in Diffusion Model

The Surprising Agreement Between Convex Optimization Theory and Learning-Rate Scheduling for Large Model Training

Learning Survival Distributions with the Asymmetric Laplace Distribution

AssistanceZero: Scalably Solving Assistance Games

Dynamic Sparse Training of Diagonally Sparse Networks

L3A: Label-Augmented Analytic Adaptation for Multi-Label Class Incremental Learning

Language Models as Implicit Tree Search

Generalizing Causal Effects from Randomized Controlled Trials to Target Populations across Diverse Environments

Ladder-Residual: Parallelism-Aware Architecture for Accelerating Large Model Inference with Communication Overlapping

Beyond Message Passing: Neural Graph Pattern Machine

Language Models May Verbatim Complete Text They Were Not Explicitly Trained On

Function-to-Style Guidance of LLMs for Code Translation

In-Context Deep Learning via Transformer Models

TimePoint: Accelerated Time Series Alignment via Self-Supervised Keypoint and Descriptor Learning

Regularized Langevin Dynamics for Combinatorial Optimization

Preference-CFR: Beyond Nash Equilibrium for Better Game Strategies

Scaling Test-Time Compute Without Verification or RL is Suboptimal

Probing Visual Language Priors in VLMs

Discrete Markov Probabilistic Models: An Improved Discrete Score-Based Framework with sharp convergence bounds under minimal assumptions

Runtime Analysis of Evolutionary NAS for Multiclass Classification

Benchmarking Abstract and Reasoning Abilities Through A Theoretical Perspective

On the Dynamic Regret of Following the Regularized Leader: Optimism with History Pruning

What Do Learning Dynamics Reveal About Generalization in LLM Mathematical Reasoning?

On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents

Organize the Web: Constructing Domains Enhances Pre-Training Data Curation

Efficient Noise Calculation in Deep Learning-based MRI Reconstructions

Vision-Language Model Selection and Reuse for Downstream Adaptation

Tilted Sharpness-Aware Minimization

CoCoA-Mix: Confusion-and-Confidence-Aware Mixture Model for Context Optimization

A Meta-learner for Heterogeneous Effects in Difference-in-Differences

Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations

Right Now, Wrong Then: Non-Stationary Direct Preference Optimization under Preference Drift

Compression via Pre-trained Transformers: A Study on Byte-Level Multimodal Data

ReVISE: Learning to Refine at Test-Time via Intrinsic Self-Verification

Controlling Large Language Model with Latent Action

A Computationally Efficient Algorithm for Infinite-Horizon Average-Reward Linear MDPs

EvFocus: Learning to Reconstruct Sharp Images from Out-of-Focus Event Streams

Efficient and Separate Authentication Image Steganography Network

Cape: Context-Aware Prompt Perturbation Mechanism with Differential Privacy

Compressed Image Generation with Denoising Diffusion Codebook Models

Guided Zeroth-Order Methods for Stochastic Non-convex Problems with Decision-Dependent Distributions

The Limits of Tractable Marginalization

Comparing Few to Rank Many: Active Human Preference Learning Using Randomized Frank-Wolfe Method

Nonparametric Identification of Latent Concepts

Portable Reward Tuning: Towards Reusable Fine-Tuning across Different Pretrained Models

Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluation

VIP: Vision Instructed Pre-training for Robotic Manipulation

Balancing Efficiency and Expressiveness: Subgraph GNNs with Walk-Based Centrality

Momentum-Driven Adaptivity: Towards Tuning-Free Asynchronous Federated Learning

Peri-LN: Revisiting Normalization Layer in the Transformer Architecture

EVOLvE: Evaluating and Optimizing LLMs For In-Context Exploration

An Interpretable N-gram Perplexity Threat Model for Large Language Model Jailbreaks

NeuroTree: Hierarchical Functional Brain Pathway Decoding for Mental Health Disorders

On the Alignment between Fairness and Accuracy: from the Perspective of Adversarial Robustness

Bootstrapping Self-Improvement of Language Model Programs for Zero-Shot Schema Matching

Investigating Non-Transitivity in LLM-as-a-Judge

Optimizing Robustness and Accuracy in Mixture of Experts: A Dual-Model Approach

The Impact of On-Policy Parallelized Data Collection on Deep Reinforcement Learning Networks

Chip Placement with Diffusion Models

Reinforcement Learning with Random Time Horizons

On the Provable Separation of Scales in Maximal Update Parameterization

Gradient Descent Converges Arbitrarily Fast for Logistic Regression via Large and Adaptive Stepsizes

Tuning LLM Judge Design Decisions for 1/1000 of the Cost

Faster Approximation Algorithms for k-Center via Data Reduction

CERTAIN: Context Uncertainty-aware One-Shot Adaptation for Context-based Offline Meta Reinforcement Learning

Linear $Q$-Learning Does Not Diverge in $L^2$: Convergence Rates to a Bounded Set

Revisiting Non-Acyclic GFlowNets in Discrete Environments

Sanity Checking Causal Representation Learning on a Simple Real-World System

ELMO : Efficiency via Low-precision and Peak Memory Optimization in Large Output Spaces

SkipGPT: Each Token is One of a Kind

Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation

Learning to Trust Bellman Updates: Selective State-Adaptive Regularization for Offline RL

Towards flexible perception with visual memory

Solving Zero-Sum Convex Markov Games

Spatial Reasoning with Denoising Models

Interaction-Aware Gaussian Weighting for Clustered Federated Learning

FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification

Generalization Principles for Inference over Text-Attributed Graphs with Large Language Models

Discovering Symbolic Cognitive Models from Human and Animal Behavior

Identifiable Object Representations under Spatial Ambiguities

Neural Event-Triggered Control with Optimal Scheduling

StealthInk: A Multi-bit and Stealthy Watermark for Large Language Models

Physics-Informed Weakly Supervised Learning For Interatomic Potentials

Towards Global-level Mechanistic Interpretability: A Perspective of Modular Circuits of Large Language Models

RealRAG: Retrieval-augmented Realistic Image Generation via Self-reflective Contrastive Learning

AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs

Modularized Self-Reflected Video Reasoner for Multimodal LLM with Application to Video Question Answering

Addressing Imbalanced Domain-Incremental Learning through Dual-Balance Collaborative Experts

Temporal Query Network for Efficient Multivariate Time Series Forecasting

Neural Solver Selection for Combinatorial Optimization

Contextual Optimization Under Model Misspecification: A Tractable and Generalizable Approach

Learning with Exact Invariances in Polynomial Time

Safety Alignment Can Be Not Superficial With Explicit Safety Signals

Language Models over Canonical Byte-Pair Encodings

TS-SNN: Temporal Shift Module for Spiking Neural Networks

From Theory to Practice: Rethinking Green and Martin Kernels for Unleashing Graph Transformers

QMamba: On First Exploration of Vision Mamba for Image Quality Assessment

Binary Hypothesis Testing for Softmax Models and Leverage Score Models

Provably Efficient Exploration in Inverse Constrained Reinforcement Learning

Equivariant Polynomial Functional Networks

Policy-Regret Minimization in Markov Games with Function Approximation

In-Context Reinforcement Learning From Suboptimal Historical Data

GTR: A General, Multi-View, and Dynamic Framework for Trajectory Representation Learning

Online Curvature-Aware Replay: Leveraging $\mathbf{2^{nd}}$ Order Information for Online Continual Learning

Look Twice Before You Answer: Memory-Space Visual Retracing for Hallucination Mitigation in Multimodal Large Language Models

Deterministic Sparse Fourier Transform for Continuous Signals with Frequency Gap

Strategy Coopetition Explains the Emergence and Transience of In-Context Learning

General framework for online-to-nonconvex conversion: Schedule-free SGD is also effective for nonconvex optimization

Where is the Truth? The Risk of Getting Confounded in a Continual World

CMoS: Rethinking Time Series Prediction Through the Lens of Chunk-wise Spatial Correlations

Global curvature for second-order optimization of neural networks

The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text

Categorical Distributional Reinforcement Learning with Kullback-Leibler Divergence: Convergence and Asymptotics

Elucidating Flow Matching ODE Dynamics via Data Geometry and Denoisers

What can large language models do for sustainable food?

TreeLoRA: Efficient Continual Learning via Layer-Wise LoRAs Guided by a Hierarchical Gradient-Similarity Tree

Efficient Generative Modeling with Residual Vector Quantization-Based Tokens

DragSolver: A Multi-Scale Transformer for Real-World Automotive Drag Coefficient Estimation

Rethinking the Temperature for Federated Heterogeneous Distillation

Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models

MATS: An Audio Language Model under Text-only Supervision

Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks

A Bayesian Model Selection Criterion for Selecting Pretraining Checkpoints

Breaking the Curse of Multiagency in Robust Multi-Agent Reinforcement Learning

Active Learning for Efficient Discovery of Optimal Combinatorial Perturbations

Primphormer: Efficient Graph Transformers with Primal Representations

Sharp Generalization for Nonparametric Regression by Over-Parameterized Neural Networks: A Distribution-Free Analysis in Spherical Covariate

LEMoN: Label Error Detection using Multimodal Neighbors

Hierarchical Reinforcement Learning with Targeted Causal Interventions

Towards a Formal Theory of Representational Compositionality

Linear Contextual Bandits With Interference

Modeling All-Atom Glycan Structures via Hierarchical Message Passing and Multi-Scale Pre-training

High-Fidelity Simultaneous Speech-To-Speech Translation

Cover learning for large-scale topology representation

Falcon: Fast Visuomotor Policies via Partial Denoising

Double Machine Learning for Causal Inference under Shared-State Interference

Quadratic Upper Bound for Boosting Robustness

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM’s Reasoning Capability

From Low Rank Gradient Subspace Stabilization to Low-Rank Weights: Observations, Theories, and Applications

HEAP: Hyper Extended A-PDHG Operator for Constrained High-dim PDEs

MIPT: Multilevel Informed Prompt Tuning for Robust Molecular Property Prediction

Predicting High-precision Depth on Low-Precision Devices Using 2D Hilbert Curves

Differentiable Structure Learning with Ancestral Constraints

Learning Gaussian DAG Models without Condition Number Bounds

MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking

Enhancing Graph Contrastive Learning for Protein Graphs from Perspective of Invariance

QoS-Efficient Serving of Multiple Mixture-of-Expert LLMs Using Partial Runtime Reconfiguration

Stability and Generalization Analysis of Decentralized SGD: Sharper Bounds Beyond Lipschitzness and Smoothness

Lightweight Online Adaption for Time Series Foundation Model Forecasts

Open Materials Generation with Stochastic Interpolants

Pre-Training Graph Contrastive Masked Autoencoders are Strong Distillers for EEG

Visual Generation Without Guidance

Learning to Steer Learners in Games

Auditing Prompt Caching in Language Model APIs

Beyond Log-Concavity and Score Regularity: Improved Convergence Bounds for Score-Based Generative Models in W2-distance

Time-Aware World Model for Adaptive Prediction and Control

Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling

Stronger Neyman Regret Guarantees for Adaptive Experimental Design

Adaptive Self-improvement LLM Agentic System for ML Library Development

QuanONet: Quantum Neural Operator with Application to Differential Equation

Matrix Completion with Incomplete Side Information via Orthogonal Complement Projection

Functional Alignment Can Mislead: Examining Model Stitching

Enhancing Ligand Validity and Affinity in Structure-Based Drug Design with Multi-Reward Optimization

Algorithmic Recourse for Long-Term Improvement

Galileo: Learning Global & Local Features of Many Remote Sensing Modalities

ELEMENTAL: Interactive Learning from Demonstrations and Vision-Language Models for Reward Design in Robotics

Balancing the Scales: A Theoretical and Algorithmic Framework for Learning from Imbalanced Data

Learning from Sample Stability for Deep Clustering

Slimming the Fat-Tail: Morphing-Flow for Adaptive Time Series Modeling

Unisoma: A Unified Transformer-based Solver for Multi-Solid Systems

A Square Peg in a Square Hole: Meta-Expert for Long-Tailed Semi-Supervised Learning

ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation

RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models

Adversarial Robustness in Two-Stage Learning-to-Defer: Algorithms and Guarantees

Masked Generative Nested Transformers with Decode Time Scaling

NTK-DFL: Enhancing Decentralized Federated Learning in Heterogeneous Settings via Neural Tangent Kernel

From Kernels to Features: A Multi-Scale Adaptive Theory of Feature Learning

Contextual Online Decision Making with Infinite-Dimensional Functional Regression

InfAlign: Inference-aware language model alignment

HPS: Hard Preference Sampling for Human Preference Alignment

The Jailbreak Tax: How Useful are Your Jailbreak Outputs?

Stay Hungry, Keep Learning: Sustainable Plasticity for Deep Reinforcement Learning

Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning

Energy-Based Flow Matching for Generating 3D Molecular Structure

iDPA: Instance Decoupled Prompt Attention for Incremental Medical Object Detection

ExtPose: Robust and Coherent Pose Estimation by Extending ViTs

C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented Generation

Voronoi-grid-based Pareto Front Learning and Its Application to Collaborative Federated Learning

Retraining-free Merging of Sparse MoE via Hierarchical Clustering

PRIME: Deep Imbalanced Regression with Proxies

Weakly Supervised Anomaly Detection via Dual-Tailed Kernel

Graph Minimum Factor Distance and Its Application to Large-Scale Graph Data Clustering

Discovering Physics Laws of Dynamical Systems via Invariant Function Learning

Learning Joint Interventional Effects from Single-Variable Interventions in Additive Models

Counterfactual Voting Adjustment for Quality Assessment and Fairer Voting in Online Platforms with Helpfulness Evaluation

Deep Electromagnetic Structure Design Under Limited Evaluation Budgets

Strengthen Out-of-Distribution Detection Capability with Progressive Self-Knowledge Distillation

Byzantine-Resilient Federated Alternating Gradient Descent and Minimization for Partly-Decoupled Low Rank Matrix Learning

Primal-Dual Neural Algorithmic Reasoning

Test-Time Learning for Large Language Models

Extracting Rare Dependence Patterns via Adaptive Sample Reweighting

Annealing Flow Generative Models Towards Sampling High-Dimensional and Multi-Modal Distributions

Random Feature Representation Boosting

Optimal Decision Tree Pruning Revisited: Algorithms and Complexity

Task Generalization with Autoregressive Compositional Structure: Can Learning from $D$ Tasks Generalize to $D^T$ Tasks?

Synonymous Variational Inference for Perceptual Image Compression

Fraud-Proof Revenue Division on Subscription Platforms

On the Local Complexity of Linear Regions in Deep ReLU Networks

LASER: Attention with Exponential Transformation

Context is Key: A Benchmark for Forecasting with Essential Textual Information

Perceptual-GS: Scene-adaptive Perceptual Densification for Gaussian Splatting

Subobject-level Image Tokenization

Preference Learning for AI Alignment: a Causal Perspective

Meta Optimality for Demographic Parity Constrained Regression via Post-Processing

Unpaired Point Cloud Completion via Unbalanced Optimal Transport

Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search

Lean and Mean Adaptive Optimization via Subset-Norm and Subspace-Momentum with Convergence Guarantees

Gradient Inversion of Multimodal Models

Online Sparsification of Bipartite-Like Clusters in Graphs

Clustering Properties of Self-Supervised Learning

Causal Discovery from Conditionally Stationary Time Series

Distillation of Discrete Diffusion through Dimensional Correlations

Elucidating the Design Space of Multimodal Protein Language Models

Fair Clustering via Alignment

Geometric Hyena Networks for Large-scale Equivariant Learning

Black-Box Adversarial Attacks on LLM-Based Code Completion

EquivaMap: Leveraging LLMs for Automatic Equivalence Checking of Optimization Formulations

HGOT: Self-supervised Heterogeneous Graph Neural Network with Optimal Transport

Enforcing Idempotency in Neural Networks

scSSL-Bench: Benchmarking Self-Supervised Learning for Single-Cell Data

Dynamical Modeling of Behaviorally Relevant Spatiotemporal Patterns in Neural Imaging Data

A Reasoning-Based Approach to Cryptic Crossword Clue Solving

Learning With Multi-Group Guarantees For Clusterable Subpopulations

What makes an Ensemble (Un) Interpretable?

FastCAV: Efficient Computation of Concept Activation Vectors for Explaining Deep Neural Networks

Stray Intrusive Outliers-Based Feature Selection on Intra-Class Asymmetric Instance Distribution or Multiple High-Density Clusters

How Do Transformers Learn Variable Binding in Symbolic Programs?

Generalization Analysis for Supervised Contrastive Representation Learning under Non-IID Settings

Poly2Vec: Polymorphic Fourier-Based Encoding of Geospatial Objects for GeoAI Applications

A Unified Framework for Generalization Error Analysis of Learning with Arbitrary Discrete Weak Features

Online Detection of LLM-Generated Texts via Sequential Hypothesis Testing by Betting

LIMEFLDL: A Local Interpretable Model-Agnostic Explanations Approach for Label Distribution Learning

Generalized Venn and Venn-Abers Calibration with Applications in Conformal Prediction

SPRI: Aligning Large Language Models with Context-Situated Principles

LV-XAttn: Distributed Cross-Attention for Long Visual Inputs in Multimodal Large Language Models

Explicit Discovery of Nonlinear Symmetries from Dynamic Data

Correlated Errors in Large Language Models

Curvature Enhanced Data Augmentation for Regression

Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment

Does Low Rank Adaptation Lead to Lower Robustness against Training-Time Attacks?

Training High Performance Spiking Neural Network by Temporal Model Calibration

Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct Reasoning

Inverse Optimization via Learning Feasible Regions

Ranked Entropy Minimization for Continual Test-Time Adaptation

LGDM: Latent Guidance in Diffusion Models for Perceptual Evaluations

Learning to Match Unpaired Data with Minimum Entropy Coupling

Tensor Decomposition Based Memory-Efficient Incremental Learning

Directed Graph Grammars for Sequence-based Learning

FeatSharp: Your Vision Model Features, Sharper

Faster Rates for Private Adversarial Bandits

A Bregman Proximal Viewpoint on Neural Operators

An Instrumental Value for Data Production and its Application to Data Pricing

TokenSwift: Lossless Acceleration of Ultra Long Sequence Generation

GaussMarker: Robust Dual-Domain Watermark for Diffusion Models

SAND: One-Shot Feature Selection with Additive Noise Distortion

Chameleon: A Flexible Data-mixing Framework for Language Model Pretraining and Finetuning

The Lock-in Hypothesis: Stagnation by Algorithm

BOOD: Boundary-based Out-Of-Distribution Data Generation

Reliable Algorithm Selection for Machine Learning-Guided Design

Synthetic Text Generation for Training Large Language Models via Gradient Matching

Reasoning-as-Logic-Units: Scaling Test-Time Reasoning in Large Language Models Through Logic Unit Alignment

How Expressive are Knowledge Graph Foundation Models?

MMInference: Accelerating Pre-filling for Long-Context Visual Language Models via Modality-Aware Permutation Sparse Attention

HYGMA: Hypergraph Coordination Networks with Dynamic Grouping for Multi-Agent Reinforcement Learning

Beyond Topological Self-Explainable GNNs: A Formal Explainability Perspective

PIGDreamer: Privileged Information Guided World Models for Safe Partially Observable Reinforcement Learning

Pareto-Optimal Fronts for Benchmarking Symbolic Regression Algorithms

EasyInv: Toward Fast and Better DDIM Inversion

Hierarchical Equivariant Policy via Frame Transfer

SageAttention2: Efficient Attention with Thorough Outlier Smoothing and Per-thread INT4 Quantization

Flow Matching for Few-Trial Neural Adaptation with Stable Latent Dynamics

You Always Recognize Me (YARM): Robust Texture Synthesis Against Multi-View Corruption

Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations

Uncertainty Quantification for LLM-Based Survey Simulations

Concept Reachability in Diffusion Models: Beyond Dataset Constraints

An Efficient Pruner for Large Language Model with Theoretical Guarantee

One Wave To Explain Them All: A Unifying Perspective On Feature Attribution

Determinant Estimation under Memory Constraints and Neural Scaling Laws

Near Optimal Non-asymptotic Sample Complexity of 1-Identification

How to Move Your Dragon: Text-to-Motion Synthesis for Large-Vocabulary Objects

Heterogeneous Data Game: Characterizing the Model Competition Across Multiple Data Sources

Geometry-Informed Neural Networks

Dimension-Independent Rates for Structured Neural Density Estimation

Holistic Physics Solver: Learning PDEs in a Unified Spectral-Physical Space

Empirical Privacy Variance

Epsilon-VAE: Denoising as Visual Decoding

Learning Safe Strategies for Value Maximizing Buyers in Uniform Price Auctions

Training Flexible Models of Genetic Variant Effects from Functional Annotations using Accelerated Linear Algebra

Contrastive Private Data Synthesis via Weighted Multi-PLM Fusion

Differentiable Solver Search for Fast Diffusion Sampling

When do neural networks learn world models?

ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

Efficient Curvature-Aware Hypergradient Approximation for Bilevel Optimization

Potemkin Understanding in Large Language Models

Efficient Fine-Grained Guidance for Diffusion Model Based Symbolic Music Generation

Generation from Noisy Examples

Fast and Low-Cost Genomic Foundation Models via Outlier Removal

Patch-wise Structural Loss for Time Series Forecasting

Stealix: Model Stealing via Prompt Evolution

DataDecide: How to Predict Best Pretraining Data with Small Experiments

The Brain's Bitter Lesson: Scaling Speech Decoding With Self-Supervised Learning

Sample Complexity of Distributionally Robust Off-Dynamics Reinforcement Learning with Online Interaction

Reasoning Through Execution: Unifying Process and Outcome Rewards for Code Generation

Semantics-aware Test-time Adaptation for 3D Human Pose Estimation

Learning Time-Varying Multi-Region Brain Communications via Scalable Markovian Gaussian Processes

Reflect-then-Plan: Offline Model-Based Planning through a Doubly Bayesian Lens

Effective and Efficient Masked Image Generation Models

Is Noise Conditioning Necessary for Denoising Generative Models?

Behavior-Regularized Diffusion Policy Optimization for Offline Reinforcement Learning

One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional Discrepancy

Random Policy Evaluation Uncovers Policies of Generative Flow Networks

Divide and Conquer: Grounding LLMs as Efficient Decision-Making Agents via Offline Hierarchical Reinforcement Learning

Rethinking Time Encoding via Learnable Transformation Functions

Deep Reinforcement Learning from Hierarchical Preference Design

Geometric Median (GM) Matching for Robust k-Subset Selection from Noisy Data

TimeBridge: Non-Stationarity Matters for Long-term Time Series Forecasting

VideoRoPE: What Makes for Good Video Rotary Position Embedding?

Finite-Time Convergence Rates in Stochastic Stackelberg Games with Smooth Algorithmic Agents

Robust Spatio-Temporal Centralized Interaction for OOD Learning

NeuronTune: Towards Self-Guided Spurious Bias Mitigation

Self-cross Feature based Spiking Neural Networks for Efficient Few-shot Learning

OmniAudio: Generating Spatial Audio from 360-Degree Video

INRFlow: Flow Matching for INRs in Ambient Space

µnit Scaling: Simple and Scalable FP8 LLM Training

TIMING: Temporality-Aware Integrated Gradients for Time Series Explanation

Linear Mode Connectivity between Multiple Models modulo Permutation Symmetries

Telling Peer Direct Effects from Indirect Effects in Observational Network Data

Combinatorial Reinforcement Learning with Preference Feedback

EPIC: Efficient Position-Independent Caching for Serving Large Language Models

Neurosymbolic World Models for Sequential Decision Making

A Two-Stage Learning-to-Defer Approach for Multi-Task Learning

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing

Direct Motion Models for Assessing Generated Videos

Parallel Simulation for Log-concave Sampling and Score-based Diffusion Models

Flow of Reasoning: Training LLMs for Divergent Reasoning with Minimal Examples

Unisolver: PDE-Conditional Transformers Towards Universal Neural PDE Solvers

"Why Is There a Tumor?": Tell Me the Reason, Show Me the Evidence

Adaptive Partitioning Schemes for Optimistic Optimization

Securing Equal Share: A Principled Approach for Learning Multiplayer Symmetric Games

Lightspeed Geometric Dataset Distance via Sliced Optimal Transport

Large Displacement Motion Transfer with Unsupervised Anytime Interpolation

N2GON: Neural Networks for Graph-of-Net with Position Awareness

LAuReL: Learned Augmented Residual Layer

Focus On This, Not That! Steering LLMs with Adaptive Feature Specification

Symmetry-Robust 3D Orientation Estimation

Tight and Fast Bounds for Multi-Label Learning

MoRAgent: Parameter Efficient Agent Tuning with Mixture-of-Roles

Generalizing from SIMPLE to HARD Visual Reasoning: Can We Mitigate Modality Imbalance in VLMs?

Compositional Generalization via Forced Rendering of Disentangled Latents

Learning Attribute-Aware Hash Codes for Fine-Grained Image Retrieval via Query Optimization

Enhancing Visual Localization with Cross-Domain Image Generation

GRAM: A Generative Foundation Reward Model for Reward Generalization

Efficient Long Context Fine-tuning with Chunk Flow

Efficient Diffusion Models for Symmetric Manifolds

Learning Adversarial MDPs with Stochastic Hard Constraints

Variational Phylogenetic Inference with Products over Bipartitions

TimePro: Efficient Multivariate Long-term Time Series Forecasting with Variable- and Time-Aware Hyper-state

Multi-Objective Causal Bayesian Optimization

Multiaccuracy and Multicalibration via Proxy Groups

Are High-Quality AI-Generated Images More Difficult for Models to Detect?

From Language Models over Tokens to Language Models over Characters

MIB: A Mechanistic Interpretability Benchmark

ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning

A Multi-Region Brain Model to Elucidate the Role of Hippocampus in Spatially Embedded Decision-Making

Aggregation Buffer: Revisiting DropEdge with a New Parameter Block

SNS-Bench: Defining, Building, and Assessing Capabilities of Large Language Models in Social Networking Services

Breaking Silos: Adaptive Model Fusion Unlocks Better Time Series Forecasting

TraceGrad: a Framework Learning Expressive SO(3)-equivariant Non-linear Representations for Electronic-Structure Hamiltonian Prediction

Evaluating LLMs Across Multi-Cognitive Levels: From Medical Knowledge Mastery to Scenario-Based Problem Solving

Delay-DSGN: A Dynamic Spiking Graph Neural Network with Delay Mechanisms for Evolving Graph

Policy Optimization for CMDPs with Bandit Feedback: Learning Stochastic and Adversarial Constraints

Solving Probabilistic Verification Problems of Neural Networks using Branch and Bound

On the Power of Learning-Augmented Search Trees

The Ripple Effect: On Unforeseen Complications of Backdoor Attacks

Deep Principal Support Vector Machines for Nonlinear Sufficient Dimension Reduction

Distributionally Robust Active Learning for Gaussian Process Regression

Token Cleaning: Fine-Grained Data Selection for LLM Supervised Fine-Tuning

M3-JEPA: Multimodal Alignment via Multi-gate MoE based on the Joint-Embedding Predictive Architecture

Skrr: Skip and Re-use Text Encoder Layers for Memory Efficient Text-to-Image Generation

The Power of Random Features and the Limits of Distribution-Free Gradient Descent

Large Language Model-driven Large Neighborhood Search for Large-Scale MILP Problems

RUN: Reversible Unfolding Network for Concealed Object Segmentation

Machine Learning meets Algebraic Combinatorics: A Suite of Datasets Capturing Research-level Conjecturing Ability in Pure Mathematics

Visual and Domain Knowledge for Professional-level Graph-of-Thought Medical Reasoning

World Model Implanting for Test-time Adaptation of Embodied Agents

Discovering a Zero (Zero-Vector Class of Machine Learning)

Provable Zero-Shot Generalization in Offline Reinforcement Learning

Optimal Information Retention for Time-Series Explanations

Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity

Neural Genetic Search in Discrete Spaces

Progressive Tempering Sampler with Diffusion

When to Forget? Complexity Trade-offs in Machine Unlearning

Optimizing Large Language Model Training Using FP4 Quantization

REG: Rectified Gradient Guidance for Conditional Diffusion Models

Counting atoms faster: policy-based nuclear magnetic resonance pulse sequencing for atomic abundance measurement

Model Immunization from a Condition Number Perspective

UniDB: A Unified Diffusion Bridge Framework via Stochastic Optimal Control

Global-Local Dirichlet Processes for Clustering Grouped Data in the Presence of Group-Specific Idiosyncratic Variables

MOGIC: Metadata-infused Oracle Guidance for Improved Extreme Classification

Quantum Optimization via Gradient-Based Hamiltonian Descent

Can MLLMs Reason in Multimodality? EMMA: An Enhanced MultiModal ReAsoning Benchmark

Are LLMs Prescient? A Continuous Evaluation using Daily News as the Oracle

Knowledge-Guided Wasserstein Distributionally Robust Optimization

CUPS: Improving Human Pose-Shape Estimators with Conformalized Deep Uncertainty

A Chaotic Dynamics Framework Inspired by Dorsal Stream for Event Signal Processing

DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT Space

S2-Track: A Simple yet Strong Approach for End-to-End 3D Multi-Object Tracking

Reidentify: Context-Aware Identity Generation for Contextual Multi-Agent Reinforcement Learning

Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection

Avoiding spurious sharpness minimization broadens applicability of SAM

Active Learning of Deep Neural Networks via Gradient-Free Cutting Planes

Recommendations with Sparse Comparison Data: Provably Fast Convergence for Nonconvex Matrix Factorization

Canonical Rank Adaptation: An Efficient Fine-Tuning Strategy for Vision Transformers

How to Train Your Multi-Exit Model? Analyzing the Impact of Training Strategies

VTGaussian-SLAM: RGBD SLAM for Large Scale Scenes with Splatting View-Tied 3D Gaussians

Sliding Puzzles Gym: A Scalable Benchmark for State Representation in Visual Reinforcement Learning

Universal Approximation of Mean-Field Models via Transformers

Mahalanobis++: Improving OOD Detection via Feature Normalization

Feature out! Let Raw Image as Your Condition for Blind Face Restoration

Adaptive Sensitivity Analysis for Robust Augmentation against Natural Corruptions in Image Segmentation

Outlier-Aware Post-Training Quantization for Discrete Graph Diffusion Models

A Cross Modal Knowledge Distillation & Data Augmentation Recipe for Improving Transcriptomics Representations through Morphological Features

Optimizing Noise Distributions for Differential Privacy

Dissecting Submission Limit in Desk-Rejections: A Mathematical Analysis of Fairness in AI Conference Policies

TtBA: Two-third Bridge Approach for Decision-Based Adversarial Attack

GrokFormer: Graph Fourier Kolmogorov-Arnold Transformers

EnsLoss: Stochastic Calibrated Loss Ensembles for Preventing Overfitting in Classification

A Theoretical Study of (Hyper) Self-Attention through the Lens of Interactions: Representation, Training, Generalization

How Contaminated Is Your Benchmark? Measuring Dataset Leakage in Large Language Models with Kernel Divergence

Automatic Differentiation of Optimization Algorithms with Time-Varying Updates

Ensemble Distribution Distillation via Flow Matching

Nonparametric Teaching for Graph Property Learners

Bayesian Active Learning for Bivariate Causal Discovery

Rethinking Addressing in Language Models via Contextualized Equivariant Positional Encoding

MixMin: Finding Data Mixtures via Convex Minimization

Sparse Autoencoders, Again?

Phase transitions for the existence of unregularized M-estimators in single index models

On the Adversarial Robustness of Multi-Kernel Clustering

PaperBench: Evaluating AI’s Ability to Replicate AI Research

Unconstrained Robust Online Convex Optimization

Relational Invariant Learning for Robust Solvation Free Energy Prediction

WMAdapter: Adding WaterMark Control to Latent Diffusion Models

Vector Grimoire: Codebook-based Shape Generation under Raster Image Supervision

Interpolating Neural Network-Tensor Decomposition (INN-TD): a scalable and interpretable approach for large-scale physics-based problems

Nonparametric Modern Hopfield Models

TypyBench: Evaluating LLM Type Inference for Untyped Python Repositories

AffectGPT: A New Dataset, Model, and Benchmark for Emotion Understanding with Multimodal Large Language Models

ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization

video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model

Modulated Diffusion: Accelerating Generative Modeling with Modulated Quantization

Reinforcement Learning Control of a Physical Robot Device for Assisted Human Walking without a Simulator

A Non-isotropic Time Series Diffusion Model with Moving Average Transitions

Learning with Expected Signatures: Theory and Applications

Principal-Agent Bandit Games with Self-Interested and Exploratory Learning Agents

VideoJAM: Joint Appearance-Motion Representations for Enhanced Motion Generation in Video Models

Rethinking the Stability-Plasticity Trade-off in Continual Learning from an Architectural Perspective

Diffusion Sampling Correction via Approximately 10 Parameters

How Transformers Learn Regular Language Recognition: A Theoretical Study on Training Dynamics and Implicit Bias

Representation Preserving Multiclass Agnostic to Realizable Reduction

BiMaCoSR: Binary One-Step Diffusion Model Leveraging Flexible Matrix Compression for Real Super-Resolution

Dynamical phases of short-term memory mechanisms in RNNs

AdaPTS: Adapting Univariate Foundation Models to Probabilistic Multivariate Time Series Forecasting

Learning to Quantize for Training Vector-Quantized Networks

TANGO: Clustering with Typicality-Aware Nonlocal Mode-Seeking and Graph-Cut Optimization

SGD Jittering: A Training Strategy for Robust and Accurate Model-Based Architectures

On the Importance of Gaussianizing Representations

Unified Screening for Multiple Diseases

CtrlSynth: Controllable Image Text Synthesis for Data-Efficient Multimodal Learning

SparseLoRA: Accelerating LLM Fine-Tuning with Contextual Sparsity

Survival Analysis via Density Estimation

Let LLM Tell What to Prune and How Much to Prune

Alberta Wells Dataset: Pinpointing Oil and Gas Wells from Satellite Imagery

MIRROR: Make Your Object-Level Multi-View Generation More Consistent with Training-Free Rectification

QuRe: Query-Relevant Retrieval through Hard Negative Sampling in Composed Image Retrieval

STP: Self-play LLM Theorem Provers with Iterative Conjecturing and Proving

Sketch to Adapt: Fine-Tunable Sketches for Efficient LLM Adaptation

Adversarial Combinatorial Semi-bandits with Graph Feedback

DynaMind: Reasoning over Abstract Video Dynamics for Embodied Decision-Making

PROXSPARSE: REGULARIZED LEARNING OF SEMI-STRUCTURED SPARSITY MASKS FOR PRETRAINED LLMS

Aligning Multimodal Representations through an Information Bottleneck

PARM: Multi-Objective Test-Time Alignment via Preference-Aware Autoregressive Reward Model

XAttnMark: Learning Robust Audio Watermarking with Cross-Attention

DAMA: Data- and Model-aware Alignment of Multi-modal LLMs

MP-Nav: Enhancing Data Poisoning Attacks against Multimodal Learning

Continuous Bayesian Model Selection for Multivariate Causal Discovery

Position: Evaluating Generative AI Systems Is a Social Science Measurement Challenge

Position: Constants are Critical in Regret Bounds for Reinforcement Learning

Position: The Future of Bayesian Prediction Is Prior-Fitted

Position: AI Agents Need Authenticated Delegation

Position: Explainable AI Cannot Advance Without Better User Studies

Position: All Current Generative Fidelity and Diversity Metrics are Flawed

Position: Graph Matching Systems Deserve Better Benchmarks

Position: Certified Robustness Does Not (Yet) Imply Model Security

Position: Spectral GNNs Rely Less on Graph Fourier Basis than Conceived

Position: LLMs Need a Bayesian Meta-Reasoning Framework for More Robust and Generalizable Reasoning

Position: AI Scaling: From Up to Down and Out

Position: Probabilistic Modelling is Sufficient for Causal Inference

Position: Don't Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints

Position: Contextual Integrity is Inadequately Applied to Language Models

Position: Challenges and Future Directions of Data-Centric AI Alignment

Position: A Critical Perspective on The Value in Studying Deep Learning Phenomena

Position: We Need An Algorithmic Understanding of Generative AI

Position: You Can't Manufacture a NeRF

Position: Solve Layerwise Linear Models First to Understand Neural Dynamical Phenomena (Neural Collapse, Emergence, Lazy/Rich Regime, and Grokking)

Position: Democratic AI is Possible. The Democracy Levels Framework Shows How It Might Work.

Generative Audio Language Modeling with Continuous-valued Tokens and Masked Next-Token Prediction

Attention-Only Transformers via Unrolled Subspace Denoising

Learning Policy Committees for Effective Personalization in MDPs with Diverse Tasks

AdvAgent: Controllable Blackbox Red-teaming on Web Agents

GuardAgent: Safeguard LLM Agents via Knowledge-Enabled Reasoning

Simplifying DINO via Coding Rate Regularization

Diverse Prototypical Ensembles Improve Robustness to Subpopulation Shift

Putnam-AXIOM: A Functional & Static Benchmark for Measuring Higher Level Mathematical Reasoning in LLMs

AtlasD: Automatic Local Symmetry Discovery

Text-to-LoRA: Instant Transformer Adaption

Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search

A Unified View on Learning Unnormalized Distributions via Noise-Contrastive Estimation

WAVE: Weighted Autoregressive Varying Gate for Time Series Forecasting

Linear Transformers as VAR Models: Aligning Autoregressive Attention Mechanisms with Autoregressive Forecasting

Human-Aligned Image Models Improve Visual Decoding from the Brain

Policy-labeled Preference Learning: Is Preference Enough for RLHF?

BaWA: Automatic Optimizing Pruning Metric for Large Language Models with Balanced Weight and Activation

Q-Supervised Contrastive Representation: A State Decoupling Framework for Safe Offline Reinforcement Learning

Iterative Vectors: In-Context Gradient Steering without Backpropagation

Decision Mixer: Integrating Long-term and Local Dependencies via Dynamic Token Selection for Decision-Making

Heads up! Large Language Models Can Perform Tasks Without Your Instruction via Selective Attention Head Masking

Divide and Conquer: Learning Label Distribution with Subtasks

Approximately Correct Label Distribution Learning

BRiTE: Bootstrapping Reinforced Thinking Process to Enhance Language Model Reasoning

The Sample Complexity of Online Strategic Decision Making with Information Asymmetry and Knowledge Transportability

DPO Meets PPO: Reinforced Token Optimization for RLHF

Learning to Incentivize in Repeated Principal-Agent Problems with Adversarial Agent Arrivals

Oscillation-Reduced MXFP4 Training for Vision Transformers

AutoCATE: End-to-End, Automated Treatment Effect Estimation

DEALing with Image Reconstruction: Deep Attentive Least Squares

LAION-C: An Out-of-Distribution Benchmark for Web-Scale Vision Models

Do Vision-Language Models Really Understand Visual Language?

NeuralCohort: Cohort-aware Neural Representation Learning for Healthcare Analytics

Differentially Private Federated $k$-Means Clustering with Server-Side Data

One Leaf Reveals the Season: Occlusion-Based Contrastive Learning with Semantic-Aware Views for Efficient Visual Representation

Improving LLMs for Recommendation with Out-Of-Vocabulary Tokens

Efficient Online Reinforcement Learning for Diffusion Policy

Understanding Overadaptation in Supervised Fine-Tuning: The Role of Ensemble Methods

Handling Imbalanced Pseudolabels for Vision-Language Models with Concept Alignment and Confusion-Aware Calibrated Margin

Power Mean Estimation in Stochastic Continuous Monte-Carlo Tree Search

FACTER: Fairness-Aware Conformal Thresholding and Prompt Engineering for Enabling Fair LLM-Based Recommender Systems

Offline Opponent Modeling with Truncated Q-driven Instant Policy Refinement

Deep Bayesian Filter for Bayes-Faithful Data Assimilation

Empirical Design in Reinforcement Learning

Linear convergence of Sinkhorn's algorithm for generalized static Schrödinger bridge

Widening the Network Mitigates the Impact of Data Heterogeneity on FedAvg

Learning Configurations for Data-Driven Multi-Objective Optimization

A Geometric Approach to Personalized Recommendation with Set-Theoretic Constraints Using Box Embeddings

Looking Beyond the Top-1: Transformers Determine Top Tokens in Order

Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation

Score-Based Diffusion Policy Compatible with Reinforcement Learning via Optimal Transport

Latent Action Learning Requires Supervision in the Presence of Distractors

Understanding the Kronecker Matrix-Vector Complexity of Linear Algebra

Optimal Survey Design for Private Mean Estimation

Near Optimal Best Arm Identification for Clustered Bandits

Generalized additive models via direct optimization of regularized decision stump forests

HaploVL: A Single-Transformer Baseline for Multi-Modal Understanding

Deep Neural Cellular Potts Models

Multi-Domain Graph Foundation Models: Robust Knowledge Transfer via Topology Alignment

Low-distortion and GPU-compatible Tree Embeddings in Hyperbolic Space

Sampling Binary Data by Denoising through Score Functions

An Analysis for Reasoning Bias of Language Models with Small Initialization

Understanding the Emergence of Multimodal Representation Alignment

GAPrompt: Geometry-Aware Point Cloud Prompt for 3D Vision Model

Banyan: Improved Representation Learning with Explicit Structure

FedClean: A General Robust Label Noise Correction for Federated Learning

Hierarchical Overlapping Clustering on Graphs: Cost Function, Algorithm and Scalability

Cowpox: Towards the Immunity of VLM-based Multi-Agent Systems

AsymRnR: Video Diffusion Transformers Acceleration with Asymmetric Reduction and Restoration

Hide & Seek: Transformer Symmetries Obscure Sharpness & Riemannian Geometry Finds It

Improved Approximations for Hard Graph Problems using Predictions

TimeStacker: A Novel Framework with Multilevel Observation for Capturing Nonstationary Patterns in Time Series Forecasting

ROPO: Robust Preference Optimization for Large Language Models

Graph Inverse Style Transfer for Counterfactual Explainability

Differentially Private Analysis for Binary Response Models: Optimality, Estimation, and Inference

Mixture of Experts Made Intrinsically Interpretable

A Lens into Interpretable Transformer Mistakes via Semantic Dependency

The Energy Loss Phenomenon in RLHF: A New Perspective on Mitigating Reward Hacking

MPO: An Efficient Post-Processing Framework for Mixing Diverse Preference Alignment

NBDI: A Simple and Effective Termination Condition for Skill Extraction from Task-Agnostic Demonstrations

GRADEO: Towards Human-Like Evaluation for Text-to-Video Generation via Multi-Step Reasoning

Evaluating Neuron Explanations: A Unified Framework with Sanity Checks

S4S: Solving for a Fast Diffusion Model Solver

Generalization Analysis for Controllable Learning

Field Matching: an Electrostatic Paradigm to Generate and Transfer Data

TimeStep Master: Asymmetrical Mixture of Timestep LoRA Experts for Versatile and Efficient Diffusion Models in Vision

One Example Shown, Many Concepts Known! Counterexample-Driven Conceptual Reasoning in Mathematical LLMs

Provable Length Generalization in Sequence Prediction via Spectral Filtering

Generalization of noisy SGD in unbounded non-convex settings

Explicit Exploration for High-Welfare Equilibria in Game-Theoretic Multiagent Reinforcement Learning

Enhancing Foundation Models for Time Series Forecasting via Wavelet-based Tokenization

Quantum Speedups in Regret Analysis of Infinite Horizon Average-Reward Markov Decision Processes

X-Hacking: The Threat of Misguided AutoML

Representative Language Generation

Demystifying the Paradox of Importance Sampling with an Estimated History-Dependent Behavior Policy in Off-Policy Evaluation

From Black Boxes to Transparent Minds: Evaluating and Enhancing the Theory of Mind in Multimodal Large Language Models

Learning Utilities from Demonstrations in Markov Decision Processes

Knowledge Retention in Continual Model-Based Reinforcement Learning

Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Design

Instance-Optimal Pure Exploration for Linear Bandits on Continuous Arms

Algorithms and Hardness for Active Learning on Graphs

Local Manifold Approximation and Projection for Manifold-Aware Diffusion Planning

Can Diffusion Models Learn Hidden Inter-Feature Rules Behind Images?

Discrete and Continuous Difference of Submodular Minimization

Learnable Spatial-Temporal Positional Encoding for Link Prediction

Value-Based Deep RL Scales Predictably

Generative Intervention Models for Causal Perturbation Modeling

Cross-regularization: Adaptive Model Complexity through Validation Gradients

Beyond Cropped Regions: New Benchmark and Corresponding Baseline for Chinese Scene Text Retrieval in Diverse Layouts

Guarantees of a Preconditioned Subgradient Algorithm for Overparameterized Asymmetric Low-rank Matrix Recovery

Learning Parametric Distributions from Samples and Preferences

AuPair: Golden Example Pairs for Code Repair

Fixed-Confidence Multiple Change Point Identification under Bandit Feedback

Positive-unlabeled AUC Maximization under Covariate Shift

Learngene Tells You How to Customize: Task-Aware Parameter Initialization at Flexible Scales

Deliberation in Latent Space via Differentiable Cache Augmentation

AdvI2I: Adversarial Image Attack on Image-to-Image Diffusion Models

Causal Abstraction Learning based on the Semantic Embedding Principle

Accelerating Spectral Clustering under Fairness Constraints

polybasic Speculative Decoding Through a Theoretical Perspective

Provable Benefit of Random Permutations over Uniform Sampling in Stochastic Coordinate Descent

Graph-Supported Dynamic Algorithm Configuration for Multi-Objective Combinatorial Optimization

COExpander: Adaptive Solution Expansion for Combinatorial Optimization

Flow Q-Learning

VerbalTS: Generating Time Series from Texts

Self-Supervised Learning of Intertwined Content and Positional Features for Object Detection

Measuring Representational Shifts in Continual Learning: A Linear Transformation Perspective

LLMScan: Causal Scan for LLM Misbehavior Detection

Closed-Loop Long-Horizon Robotic Planning via Equilibrium Sequence Modeling

Retrieval Augmented Zero-Shot Enzyme Generation for Specified Substrate

Wasserstein Flow Matching: Generative Modeling Over Families of Distributions

Projection Pursuit Density Ratio Estimation

Implicit Riemannian Optimism with Applications to Min-Max Problems

Agent Workflow Memory

Refining Adaptive Zeroth-Order Optimization at Ease

Improving Value Estimation Critically Enhances Vanilla Policy Gradient

Nested Expectations with Kernel Quadrature

OW-VAP: Visual Attribute Parsing for Open World Object Detection

From Local Details to Global Context: Advancing Vision-Language Models with Attention-Based Selection

The Harder Path: Last Iterate Convergence for Uncoupled Learning in Zero-Sum Games with Bandit Feedback

Correlation Clustering Beyond the Pivot Algorithm

Gradient-based Explanations for Deep Learning Survival Models

From Complex to Atomic: Enhancing Augmented Generation via Knowledge-Aware Dual Rewriting and Reasoning

Demystifying Catastrophic Forgetting in Two-Stage Incremental Object Detector

SSHR: More Secure Generative Steganography with High-Quality Revealed Secret Images

Improving Multi-Class Calibration through Normalization-Aware Isotonic Techniques

Nonconvex Theory of $M$-estimators with Decomposable Regularizers

Generalization Bounds via Meta-Learned Model Representations: PAC-Bayes and Sample Compression Hypernetworks

Set Valued Predictions For Robust Domain Generalization

Sharp Optimality of Simple, Plug-in Estimation of the Fisher Information of a Smoothed Density

AAAR-1.0: Assessing AI’s Potential to Assist Research

A Market for Accuracy: Classification Under Competition

Efficient Network Automatic Relevance Determination

Dual Feature Reduction for the Sparse-group Lasso and its Adaptive Variant

SAE-V: Interpreting Multimodal Models for Enhanced Alignment

AutoEval Done Right: Using Synthetic Data for Model Evaluation

Deep Linear Network Training Dynamics from Random Initialization: Data, Width, Depth, and Hyperparameter Transfer

Tightening Causal Bounds via Covariate-Aware Optimal Transport

Optimal Fair Learning Robust to Adversarial Distribution Shift

Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations

Exploiting Similarity for Computation and Communication-Efficient Decentralized Optimization

ARS: Adaptive Reward Scaling for Multi-Task Reinforcement Learning

Can DBNNs Robust to Environmental Noise for Resource-constrained Scenarios?

LLM Enhancers for GNNs: An Analysis from the Perspective of Causal Mechanism Identification

Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents

Point Cloud Dataset Distillation

MODULI: Unlocking Preference Generalization via Diffusion Models for Offline Multi-Objective Reinforcement Learning

Beyond The Rainbow: High Performance Deep Reinforcement Learning on a Desktop PC

UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent

Enhancing Performance of Explainable AI Models with Constrained Concept Refinement

The Courage to Stop: Overcoming Sunk Cost Fallacy in Deep Reinforcement Learning

Extractive Structures Learned in Pretraining Enable Generalization on Finetuned Facts

Consensus Based Stochastic Optimal Control

Latent Variable Causal Discovery under Selection Bias

Causal Abstraction Inference under Lossy Representations

Identification of Latent Confounders via Investigating the Tensor Ranks of the Nonlinear Observations

Fast Incomplete Multi-view Clustering by Flexible Anchor Learning

Beyond CVaR: Leveraging Static Spectral Risk Measures for Enhanced Decision-Making in Distributional Reinforcement Learning

Reducing Tool Hallucination via Reliability Alignment

Towards Understanding Catastrophic Forgetting in Two-layer Convolutional Neural Networks

Towards Escaping from Class Dependency Modeling for Multi-Dimensional Classification

Learning Initial Basis Selection for Linear Programming via Duality-Inspired Tripartite Graph Representation and Comprehensive Supervision

Measuring In-Context Computation Complexity via Hidden State Prediction

When Dynamic Data Selection Meets Data Augmentation: Achieving Enhanced Training Acceleration

Non-Stationary Predictions May Be More Informative: Exploring Pseudo-Labels with a Two-Phase Pattern of Training Dynamics

Understanding Sharpness Dynamics in NN Training with a Minimalist Example: The Effects of Dataset Difficulty, Depth, Stochasticity, and More

Adversarial Cooperative Rationalization: The Risk of Spurious Correlations in Even Clean Datasets

From Jack of All Trades to Master of One: Specializing LLM-based Autoraters to a Test Set

BlockDialect: Block-wise Fine-grained Mixed Format Quantization for Energy-Efficient LLM Inference

Learning Mean Field Control on Sparse Graphs

Theoretical Performance Guarantees for Partial Domain Adaptation via Partial Optimal Transport

CursorCore: Assist Programming through Aligning Anything

Counterfactual Graphical Models: Constraints and Inference

A Near-Optimal Single-Loop Stochastic Algorithm for Convex Finite-Sum Coupled Compositional Optimization

SENSEI: Semantic Exploration Guided by Foundation Models to Learn Versatile World Models

ERICT: Enhancing Robustness by Identifying Concept Tokens in Zero-Shot Vision Language Models

CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization

Refined generalization analysis of the Deep Ritz Method and Physics-Informed Neural Networks

MARGE: Improving Math Reasoning with Guided Exploration

Don't Restart, Just Reuse: Reoptimizing MILPs with Dynamic Parameters

Low-Dimension-to-High-Dimension Generalization and Its Implications for Length Generalization

Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes

CABS: Conflict-Aware and Balanced Sparsification for Enhancing Model Merging

Finite-Time Global Optimality Convergence in Deep Neural Actor-Critic Methods for Decentralized Multi-Agent Reinforcement Learning

You Get What You Give: Reciprocally Fair Federated Learning

Contour Integration Underlies Human-Like Vision

On the Out-of-Distribution Generalization of Self-Supervised Learning

Deep Fuzzy Multi-view Learning for Reliable Classification

Geometric Resampling in Nearly Linear Time for Follow-the-Perturbed-Leader with Best-of-Both-Worlds Guarantee in Bandit Problems

Automated Red Teaming with GOAT: the Generative Offensive Agent Tester

Is Your Model Fairly Certain? Uncertainty-Aware Fairness Evaluation for LLMs

DSP: Dynamic Sequence Parallelism for Multi-Dimensional Transformers

Test-Time Training Provably Improves Transformers as In-context Learners

Robust Multi-Agent Reinforcement Learning with Stochastic Adversary

No Soundness in the Real World: On the Challenges of the Verification of Deployed Neural Networks

Unsupervised Learning for Class Distribution Mismatch

Faster Stochastic Optimization with Arbitrary Delays via Adaptive Asynchronous Mini-Batching

Ranking with Multiple Oracles: From Weak to Strong Stochastic Transitivity

TabFlex: Scaling Tabular Learning to Millions with Linear Attention

BAnG: Bidirectional Anchored Generation for Conditional RNA Design

Online Episodic Convex Reinforcement Learning

Understanding Synthetic Context Extension via Retrieval Heads

Learning Bayesian Nash Equilibrium in Auction Games via Approximate Best Response

CAD-Editor: A Locate-then-Infill Framework with Automated Training Data Synthesis for Text-Based CAD Editing

MoH: Multi-Head Attention as Mixture-of-Head Attention

Offline Model-based Optimization for Real-World Molecular Discovery

On Volume Minimization in Conformal Regression

Provable Policy Gradient for Robust Average-Reward MDPs Beyond Rectangularity

Reducing Confounding Bias without Data Splitting for Causal Inference via Optimal Transport

Autoencoder-Based Hybrid Replay for Class-Incremental Learning

Are Large Language Models Ready for Multi-Turn Tabular Data Analysis?

Multivariate Conformal Selection

How does Labeling Error Impact Contrastive Learning? A Perspective from Data Dimensionality Reduction

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms

Neutral residues: revisiting adapters for model extension

Modular Duality in Deep Learning

Fully Dynamic Embedding into $\ell_p$ Spaces

On the Benefits of Active Data Collection in Operator Learning

Active Fine-Tuning of Multi-Task Policies

LMRL Gym: Benchmarks for Multi-Turn Reinforcement Learning with Language Models

Safely Learning Optimal Auctions: A Testable Learning Framework for Mechanism Design

DeepCrossAttention: Supercharging Transformer Residual Connections

SecEmb: Sparsity-Aware Secure Federated Learning of On-Device Recommender System with Large Embedding

Direct Density Ratio Optimization: A Statistically Consistent Approach to Aligning Large Language Models

Behavior-agnostic Task Inference for Robust Offline In-context Reinforcement Learning

Global Context-aware Representation Learning for Spatially Resolved Transcriptomics

Enabling Optimal Decisions in Rehearsal Learning under CARE Condition

Domain-Adapted Diffusion Model for PROTAC Linker Design Through the Lens of Density Ratio in Chemical Space

Instruction-Following Pruning for Large Language Models

On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization

Optimal transport-based conformal prediction

Controllable Data Generation with Hierarchical Neural Representations

Zero Shot Generalization of Vision-Based RL Without Data Augmentation

Hardware and Software Platform Inference

On the Role of Label Noise in the Feature Learning Process

Aligning Spoken Dialogue Models from User Interactions

Limitations of measure-first protocols in quantum machine learning

Focal-SAM: Focal Sharpness-Aware Minimization for Long-Tailed Classification

Low-Rank Adapting Models for Sparse Autoencoders

Cooperation of Experts: Fusing Heterogeneous Information with Large Margin

On the Clean Generalization and Robust Overfitting in Adversarial Training from Two Theoretical Views: Representation Complexity and Training Dynamics

Grokking in the Wild: Data Augmentation for Real-World Multi-Hop Reasoning with Transformers

Continual Reinforcement Learning by Planning with Online World Models

Toward a Unified Theory of Gradient Descent under Generalized Smoothness

Diffusion Models are Secretly Exchangeable: Parallelizing DDPMs via Auto Speculation

STAR: Learning Diverse Robot Skill Abstractions through Rotation-Augmented Vector Quantization

The Best of Both Worlds: Bridging Quality and Diversity in Data Selection with Bipartite Graph

TextCenGen: Attention-Guided Text-Centric Background Adaptation for Text-to-Image Generation

Diversity By Design: Leveraging Distribution Matching for Offline Model-Based Optimization

Minimalist Concept Erasure in Generative Models

Volume Optimality in Conformal Prediction with Structured Prediction Sets

FedBEns: One-Shot Federated Learning based on Bayesian Ensemble

Unifying Specialized Visual Encoders for Video Language Models

DMOSpeech: Direct Metric Optimization via Distilled Diffusion Model in Zero-Shot Speech Synthesis

Accelerating Large Language Model Reasoning via Speculative Search

Reward-free World Models for Online Imitation Learning

Unlocking the Power of Rehearsal in Continual Learning: A Theoretical Perspective

SPEX: Scaling Feature Interaction Explanations for LLMs

Compact Matrix Quantum Group Equivariant Neural Networks

When Data-Free Knowledge Distillation Meets Non-Transferable Teacher: Escaping Out-of-Distribution Trap is All You Need

Learning Representations of Instruments for Partial Identification of Treatment Effects

DIS-CO: Discovering Copyrighted Content in VLMs Training Data

Validating Mechanistic Interpretations: An Axiomatic Approach

Robust Sparsification via Sensitivity

Distributed Event-Based Learning via ADMM

Mastering Massive Multi-Task Reinforcement Learning via Mixture-of-Expert Decision Transformer

On Teacher Hacking in Language Model Distillation

Approximate Forest Completion and Learning-Augmented Algorithms for Metric Minimum Spanning Trees

RATE: Causal Explainability of Reward Models with Imperfect Counterfactuals

Breaking the Quadratic Barrier: Robust Cardinality Sketches for Adaptive Queries

Prediction-Powered E-Values

Optimizing Temperature for Language Models with Multi-Sample Inference

Vision Graph Prompting via Semantic Low-Rank Decomposition

Robust Conformal Outlier Detection under Contaminated Reference Data

Contrastive Localized Language-Image Pre-Training

Calibrated Language Models and How to Find Them with Label Smoothing

Distributed Parallel Gradient Stacking(DPGS): Solving Whole Slide Image Stacking Challenge in Multi-Instance Learning

SEFE: Superficial and Essential Forgetting Eliminator for Multimodal Continual Instruction Tuning

Noisy SIGNSGD Is More Differentially Private Than You (Might) Think

PieClam: A Universal Graph Autoencoder Based on Overlapping Inclusive and Exclusive Communities

CateKV: On Sequential Consistency for Long-Context LLM Inference Acceleration

Proposer-Agent-Evaluator (PAE): Autonomous Skill Discovery For Foundation Model Internet Agents

Joint Learning of Energy-based Models and their Partition Function

High-Dimensional Tensor Regression With Oracle Properties

In-Context Fine-Tuning for Time-Series Foundation Models

Understanding the Unfairness in Network Quantization

An Error Analysis of Flow Matching for Deep Generative Modeling

Adversarial Robustness via Deformable Convolution with Stochasticity

CLARIFY: Contrastive Preference Reinforcement Learning for Untangling Ambiguous Queries

Sampling from Binary Quadratic Distributions via Stochastic Localization

LOGO --- Long cOntext aliGnment via efficient preference Optimization

On Linear Convergence in Smooth Convex-Concave Bilinearly-Coupled Saddle-Point Optimization: Lower Bounds and Optimal Algorithms

Clone-Robust AI Alignment

Occult: Optimizing Collaborative Communications across Experts for Accelerated Parallel MoE Training and Inference

Certification for Differentially Private Prediction in Gradient-Based Training

No-Regret is not enough! Bandits with General Constraints through Adaptive Regret Minimization

ReverB-SNN: Reversing Bit of the Weight and Activation for Spiking Neural Networks

The Hidden Dimensions of LLM Alignment: A Multi-Dimensional Analysis of Orthogonal Safety Directions

Relating Misfit to Gain in Weak-to-Strong Generalization Beyond the Squared Loss

Collapse-Proof Non-Contrastive Self-Supervised Learning

Bayesian Optimization from Human Feedback: Near-Optimal Regret Bounds

Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification

Safety-Polarized and Prioritized Reinforcement Learning

Competitively Consistent Clustering

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models

Projection Optimization: A General Framework for Multi-Objective and Multi-Group RLHF

Improving the Continuity of Goal-Achievement Ability via Policy Self-Regularization for Goal-Conditioned Reinforcement Learning

Scalable Model Merging with Progressive Layer-wise Distillation

Fluctuations of the largest eigenvalues of transformed spiked Wigner matrices

Divide and Conquer: Exploring Language-centric Tree Reasoning for Video Question-Answering

Hyper-Transforming Latent Diffusion Models

Policy Gradient with Tree Expansion

Learnware Specification via Dual Alignment

Editable Noise Map Inversion: Encoding Target-image into Noise For High-Fidelity Image Manipulation

Thinking LLMs: General Instruction Following with Thought Generation

Quantum Speedup for Hypergraph Sparsification

Enhancing Parallelism in Decentralized Stochastic Convex Optimization

Point-Level Topological Representation Learning on Point Clouds

Visual Attention Never Fades: Selective Progressive Attention ReCalibration for Detailed Image Captioning in Multimodal Large Language Models

Position: It Is Time We Test Neural Computation In Vitro

Position: The Right to AI

Position: AI Competitions Provide the Gold Standard for Empirical Rigor in GenAI Evaluation

Position: Iterative Online-Offline Joint Optimization is Needed to Manage Complex LLM Copyright Risks

Positional Attention: Expressivity and Learnability of Algorithmic Computation