Skip to yearly menu bar Skip to main content


Workshop Fri, Jul 10, 2026 • 4:00 PM – 1:00 AM PDT AUDITORIUM

2nd Workshop on Compositional Learning: Safety, Interpretability, and Agents

Giacomo Camposampiero ⋅ Pietro Barbiero
Project Page [ OpenReview] 

Abstract

Video

Chat is not available.

Schedule

Timezone: America/Los_Angeles
4:25 PM
Opening Remarks
6:10 PM
Coffee Break
7:25 PM
Awards Ceremony
→ VASAE: Naming SAE Dictionary Directions with Vocabulary-Aligned Anchoring
Kairui Zhang ⋅ Ziwen Yu ⋅ Zahraa S Abdallah ⋅ Martha Lewis
→ Grad Detect: Gradient-Based Hallucination Detection in LLMs
Anand Kamat ⋅ Brent Werness ⋅ Daniel Blake
→ Visual Counterfactual Explanations with Compositional Generative Models
Daniil Kirilenko ⋅ dario fenoglio ⋅ Martin Gjoreski ⋅ Marc Langheinrich
→ CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents
Hanna Foerster ⋅ Tom Blanchard ⋅ Kristina Nikolić ⋅ Ilia Shumailov ⋅ Cheng Zhang ⋅ Robert Mullins ⋅ Nicolas Papernot ⋅ Florian Tramer ⋅ Aaron Zhao
→ Ask, Don’t Judge: Binary Questions for Interpretable LLM Evaluation and Self-Improvement
Sangwoo Cho ⋅ Kushal Chawla ⋅ Pengshan Cai ⋅ Zefang Liu ⋅ Chenyang Zhu ⋅ Austin Zhang ⋅ Sambit Sahu
→ Universality, Composition Generalization, and Algorithm Emulation All In-Context
Jerry Yao-Chieh Hu ⋅ Hong-Yu Chen ⋅ Po-Chiao Lin ⋅ Maojiang Su ⋅ Han Liu
→ COGITAO: A Procedural and Object-Centric Framework to Evaluate Compositional and Systematic Generalization
Yassine Taoudi-Benchekroun ⋅ Klim Troyan ⋅ Pascal J. Sager ⋅ Stefan Gerber ⋅ Lukas Tuggener ⋅ Thilo Stadelmann ⋅ Benjamin F. Grewe
→ FormalImG: Evaluating Structural Compositional Generalization for T2I Models
Hong-Jie You ⋅ Jie-Jing Shao ⋅ Xiao-Wen Yang ⋅ Zhi-Fan Wu ⋅ Lin-Han Jia ⋅ Lan-Zhe Guo ⋅ Yu-Feng Li
→ Struct-to-Reason: Enhancing Video Understanding of Vision-Language Models by Decoupling Perception and Reasoning via Structured Summary
Hengyu Liu ⋅ Chenxin LI ⋅ Wenbo Hu ⋅ Zhiqin Yang ⋅ Yuxin Chen ⋅ Ying Shan ⋅ Brandon Y Feng
→ CUA-Skill: Developing Computer Using Agents with a Skill Framework
Tianyi Chen ⋅ Yinheng Li ⋅ Michael Solodko ⋅ Sen Wang ⋅ Nan Jiang ⋅ Junheng Hao ⋅ Tingyuan Cui ⋅ Jongwoo Ko ⋅ Sara Abdali ⋅ Suzhen Zheng ⋅ Pashmina Cameron ⋅ Justin Wagle ⋅ Kazuhito Koishida
→ Causal Cartographer: From Mapping to Reasoning Over Counterfactual Worlds
Gael Gendron ⋅ Jože M Rožanec ⋅ Michael Witbrock ⋅ Gillian Dobbie
→ CB-Orchestrator: Adaptive Workflow Optimization for LLM Agents via Contextual Bandits
Jiahang Sun ⋅ Zhiwei Shang ⋅ Zhipiao Liu ⋅ Hongwei Yang ⋅ XIE GUOQING ⋅ Shuang Qiu ⋅ Zhongxiang Dai
→ Compositional Failure in Audio-Visual LLMs: Late-Layer Prior Dominance Under Cross-modal Conflict
Adarsh Sudheer ⋅ David Li ⋅ Omar El-Banna ⋅ Ishaan Kodarapu ⋅ Arjun Bahuguna ⋅ Vasu Sharma
→ The Theory and Practice of MAP Inference over Non-Convex Constraints
Leander Kurscheidt ⋅ Gabriele Masina ⋅ Roberto Sebastiani ⋅ Antonio Vergari
→ Which Way Did It Move? Diagnosing and Overcoming Directional Motion Blindness in Video LLMs
Jongseo Lee ⋅ Hyuntak Lee ⋅ Sunghun Kim ⋅ Sooa Kim ⋅ Jihoon Chung ⋅ Jinwoo Choi
→ Compositional Self-Improvement
Changho Shin ⋅ Daiwei Chen ⋅ John Cooper ⋅ Brenden Lake ⋅ Frederic Sala ⋅ Ramya Vinayak
→ Explaining is Harder Than Predicting Alone: Evaluating Concept-based Explanations of MLLMs as ICL Visual Classifiers
Carmen Quiles Ramírez ⋅ Leticia L Rodriguez ⋅ Nicolas Martorell ⋅ Natalia Díaz-Rodríguez
→ LGPro: Language-Guided Prototype Discovery for Compositional Zero-Shot Learning
Anna-Alina Bondarets ⋅ Taras Rumezhak ⋅ Volodymyr Karpiv
→ Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders
Shunchang Liu ⋅ Xin Chen ⋅ Belen Martin Urcelay ⋅ Francesco Croce
→ From Numbers to Narratives: Goal-Oriented Summarization of Machine Learning Model Differences
Nam Hyeon-Woo ⋅ Tae-Hyun Oh ⋅ Zeynep Akata ⋅ Stephan Alaniz
→ CMAG: Concept-Scaffolded Retrieval for Marketplace Avatar Generation
Rajeev Goel ⋅ Jason Ding ⋅ Phani H Wajjala ⋅ Pavan Turaga ⋅ Tejaswi Gowda ⋅ Krishna C Garikipati
→ Compositional Neuro-Symbolic Reasoning
Anugyan Das ⋅ Omkar Ghugarkar ⋅ Vishvesh Bhat ⋅ Asad Aali
→ Fusion is the New Mutation: Bandit-Guided Evolution on Workflow Graphs
Zhiwei Shang ⋅ Jiahang Sun ⋅ Mingrong Gong ⋅ Mingze Kong ⋅ Zikun Qu ⋅ Pingchen Lu ⋅ Junhao Dong ⋅ Zhipiao Liu ⋅ Hongwei Yang ⋅ XIE GUOQING ⋅ Yao Shu ⋅ Zhongxiang Dai
→ Attractor Inversion: A Geometric Account of Adversarial Manipulation in Human Decision-Making
Leo Lorence George ⋅ Anushri Iyer ⋅ Abhishek Bakshi ⋅ Pavan Kulkarni
→ Operads for compositional reasoning in LLMs
Nathaniel Bottman ⋅ Kyle Richardson
→ Spatially Stable GUI Grounding via Zoom Consistency Loss
Yebin Moon ⋅ Jiyeon Son ⋅ Tae-Hyun Oh
→ Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification
Patrick Knab ⋅ Sascha Marton ⋅ Philipp J Schubert ⋅ Drago A Guggiana Nilo ⋅ Christian Bartelt
→ Emergent Social Intelligence Risks in Generative Multi-Agent Systems
Yue Huang ⋅ Yu Jiang ⋅ Wenjie Wang ⋅ Haomin Zhuang ⋅ Xiaonan Luo ⋅ Yuchen Ma ⋅ Zhangchen Xu ⋅ Zichen Chen ⋅ Nuno Moniz ⋅ Zinan Lin ⋅ Pin-Yu Chen ⋅ Nitesh Chawla ⋅ Nouha Dziri ⋅ Huan Sun ⋅ Xiangliang Zhang
→ CompFlow: Composing Velocity Fields for Multi-Condition Generation
Luca Miglior ⋅ Vincenzo Gervasi ⋅ Davide Bacciu
→ Actionable Interpretability Must Be Defined in Terms of Symmetries: A Compositional Probabilistic Approach
Pietro Barbiero ⋅ Mateo Espinosa Zarlenga ⋅ Francesco Giannini ⋅ Alberto Termine ⋅ Filippo Bonchi ⋅ Mateja Jamnik ⋅ Giuseppe Marra
→ Chain-of-Thought Gradient Descent
Hong-Yu Chen ⋅ Venkat Ganti ⋅ Jerry Yao-Chieh Hu ⋅ Hude Liu ⋅ Han Liu
→ When Do Diffusion Models learn to Generate Multiple Objects?
Yujin Jeong ⋅ Arnas Uselis ⋅ Iro Laina ⋅ Seong Joon Oh ⋅ Anna Rohrbach
→ MultiVulnBench: A Large-Scale Benchmark for Count Bias in LLM-Based Multi-Vulnerability Detection
Manan Gupta ⋅ Chinmay Pushkar ⋅ Sanchit Kabra ⋅ Dhruv Kumar ⋅ Jagat S Challa
→ Adaptive Minds: Empowering Agents with LoRA-as-Tools
Pavan C Shekar ⋅ Aswanth Krishnan
→ MAVEN: Improving Generalization in Agentic Tool Calling
Omkar Ghugarkar ⋅ Vishvesh Bhat ⋅ Muhammad Ahmed Mohsin ⋅ Asad Aali
→ MKEvolve: A Modular Multi-Agent Framework for Kernel Code Generation
Jinsoo Yoo ⋅ Rajarshi Saha ⋅ Tao Yu ⋅ Shaowei Zhu ⋅ Wei Tang ⋅ Youngsuk Park
→ TAME the BALROG: Task-Adaptive Modular Evolution framework for Game Agents
Aleksandra Pasieka ⋅ Dominika Woszczyk ⋅ Antoine Cully ⋅ Borja G León
→ From Mechanistic to Compositional Interpretability
Ward Gauderis ⋅ Thomas Dooms ⋅ Steven T Homer ⋅ Kola Ayonrinde ⋅ Geraint Wiggins
→ Sparse Memory Finetuning as a Low-Forgetting Alternative to LoRA and Full Finetuning
Prakhar Gupta ⋅ Garv Shah ⋅ Satyam Goyal ⋅ Anirudh Kanchi
7:45 PM
Lunch Break
8:30 PM
Lunch Break
11:10 PM
Panel Discussion
Wenya Wang ⋅ Francesco Locatello ⋅ Noemi Elteto ⋅ Parisa Kordjamshidi ⋅ Noam Brown ⋅ Aya Ismail
12:00 AM
Poster Session 2
→ Causal-JEPA: Learning World Models through Object-Level Latent Masking
Heejeong Nam ⋅ Quentin Le Lidec ⋅ Lucas Maes ⋅ Yann LeCun ⋅ Randall Balestriero
→ Toward Compositional Latent Action Interfaces for Generalizable Agents
Heejeong Nam ⋅ Chandradithya S Jonnalagadda ⋅ Harshit Aggarwal ⋅ Eric Xu
→ Logit Grafting: The Post-Training Delta is Sparse, Portable, and Powerful
Apurv Verma ⋅ Binh-Nguyen Nguyen ⋅ Hai Phan ⋅ Lingxiao Wang
→ Policy Transfer for Hierarchical Goal-Conditioned Reinforcement Learning
Usman Islam ⋅ Zhixun Chen ⋅ Stefanos Leonardos ⋅ Matteo Leonetti ⋅ Yali Du
→ Tracking Training Phases in Compositional Learning with Task-Agnostic Measures
Niclas Dern ⋅ Selma Mazioud ⋅ Jakob Heiss ⋅ Avrajit Ghosh ⋅ Curtis J McDonald ⋅ Gabriel Clara ⋅ Bin Yu
→ Unsupervised Decomposition with Recombination-Consistent Diffusion Models
Archer Wang ⋅ Emile Anand ⋅ Yilun Du ⋅ Marin Soljačić
→ Improving the Compositionality of Triplet-Based Neural Algorithmic Reasoners
Stjepan Požgaj ⋅ Dobrik Georgiev ⋅ Marin Šilić ⋅ Goran Delac ⋅ Klemo Vladimir
→ Code-enabled language models can outperform reasoning models on diverse tasks
Cedegao Zhang ⋅ Cédric Colas ⋅ Gabriel Poesia ⋅ Josh Tenenbaum ⋅ Jacob Andreas
→ Do Thinking Tokens Help with Safety?
Narutatsu Ri ⋅ Abhishek Panigrahi ⋅ Sanjeev Arora
→ Reflection Anchors for Interpretable Compositional Visual Reasoning in Multimodal Reinforcement Learning
Xuan Gong ⋅ Hanbo Huang ⋅ Hao Zheng ⋅ Yiran Zhang ⋅ Wenbin Dai ⋅ Weishu Zhao ⋅ Shiyu Liang
→ Dual-Resolution Recursive Energy: Certified Contract–Expand Inference for Sequential Decision Making
Haozhou Gao ⋅ ZENG JIARUI ⋅ Wendi Ren ⋅ Yanwen Liu ⋅ Shuang Li
→ Compositional Adversarial Training for Robust Visual Watermarking
Anirudh Satheesh ⋅ Michael-Andrei Panaitescu-Liess ⋅ Andrew Y Xu ⋅ Georgios Milis ⋅ Heng Huang ⋅ Zikui Cai ⋅ Furong Huang
→ Installing and Obstructing Heuristics: Learning Dynamics in Nim
Leo Villani ⋅ Sultan Daniels ⋅ Ijin Yu ⋅ Anant Sahai
→ Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw
Zijun Wang ⋅ Haoqin Tu ⋅ Letian Zhang ⋅ Hardy Chen ⋅ Juncheng Wu ⋅ Xiangyan Liu ⋅ Zhenlong Yuan ⋅ Tianyu Pang ⋅ Michael Q Shieh ⋅ Fengze Liu ⋅ Zeyu Zheng ⋅ Huaxiu Yao ⋅ Yuyin Zhou ⋅ Cihang Xie
→ Don't Trust Stubborn Neighbors: A Security Framework for Agentic Networks
Samira Abedini ⋅ Sina Mavali ⋅ Lea Schönherr ⋅ Martin Pawelczyk ⋅ Rebekka Burkholz
→ A Theory of Atomic Features and Four Testable Predictions
Kenny Peng ⋅ Jon Kleinberg ⋅ Nikhil Garg
→ Meaning Representations as Variational Quantum Circuits
Tilen G Limbäck-Stokin ⋅ Tanishka Birdavade ⋅ Kin I Lo ⋅ Mehrnoosh Sadrzadeh
→ Learning to Theorize the World from Observation
Doojin Baek ⋅ Gyubin Lee ⋅ Junyeob Baek ⋅ Hosung Lee ⋅ Sungjin Ahn
→ RL Post-Training Builds Compositional Reasoning Strategies
Azwar Abdulsalam ⋅ Nishil Patel ⋅ Andrew Saxe
→ Compositional Consistency-Guided Decoding for Three-Way Logical Question Answering
Tianyi Huang ⋅ Ming Ren Hou ⋅ Jiaheng Su ⋅ Yutong Zhang ⋅ Ziling Zhang
→ Reasoning as State Transition: A Representational Analysis of Reasoning Evolution in Large Language Models
Siyuan Zhang ⋅ Jialian Li ⋅ Yichi Zhang ⋅ Xiao Yang ⋅ Yinpeng Dong ⋅ Hang Su
→ Direction-Conditioned Policies via Compositional Subgoal Scoring for Online Goal-Conditioned Reinforcement Learning
Swaminathan S K ⋅ Damiya Gondha ⋅ Theyanesh Eswaramoorthy Rajahkrishnan ⋅ Aritra Hazra
→ Additive Relational Bindings in Transformers: What Sparse Autoencoders Miss
Sebastian Hönig ⋅ Kushal Jain ⋅ Su Ji Park ⋅ Bart Bussmann ⋅ Patrick Leask
→ Separable Representations of Task Complexity and Deliberation in Reasoning Language Models.
Xuan-Quang Nguyen ⋅ Hieu Vu ⋅ Dung Viet Nguyen ⋅ Hai T Luu ⋅ Linh D Tran ⋅ Tan Nguyen
→ Playing Devil’s Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy
Ishaan Kelkar ⋅ Nebras Alam ⋅ Vikram Kakaria ⋅ Madhur Panwar ⋅ Vasu Sharma ⋅ Maheep Chaudhary
→ Measuring the Limits of Continual Learning for LLMs
Nimit Kalra ⋅ Narutatsu Ri ⋅ Zerzar Bukhari ⋅ Ang Li ⋅ Sanae Lotfi ⋅ Liam Fowl ⋅ Micah Goldblum
→ Learning What’s Missing: Failure-Driven Skill Discovery via Predicate Bridges
Yanwen Liu ⋅ Wendi Ren ⋅ Haozhou Gao ⋅ Shuang Li
→ Emergent Compositional Skills in Mixture-of-Experts VLAs
Shlok Shah ⋅ Rhiaan Jhaveri ⋅ Tharun Kumar Tiruppali Kalidoss ⋅ Chirayu Nimonkar ⋅ Ishaan Javali ⋅ Dhruv Ileshkumar Shah
→ Dimensionality Controls When Modularity Helps in Continual Learning
Kathrin Korte ⋅ Christian M Adriano ⋅ Joachim W Pedersen ⋅ Eleni Nisioti ⋅ Sebastian Risi
→ Mitigating Over-Personalization in Language Models via Structured Memory
Abdal Hakeem Hannoon ⋅ Andrew Zhao ⋅ Mihir Narayan ⋅ Sharvin Goyal ⋅ Ivaxi Sheth
→ Circuit Oracle: Automating Attribution Graph Analysis via Natural-Language Queries
Hong Kiat Tan ⋅ Shariar Kabir ⋅ Swastik Agrawal ⋅ Sai V Reddy Chereddy ⋅ Sriram Balasubramanian
→ Beyond Safe Data: Pretraining-Stage Alignment with Regular Safety Reflection
Jinhan Li ⋅ Kexian Tang ⋅ Yihan Xu ⋅ Zhuorui Ye ⋅ Kaifeng Lyu
→ Structure over Pixels: Learning Variable-Length Visual Programs
Piotr Wyrwiński ⋅ Kacper Dobek ⋅ Krzysztof Krawiec
→ ClinSeekAgent: Automating Multi-modal Evidence Seeking for Agentic Clinical Reasoning
Juncheng Wu ⋅ Letian Zhang ⋅ Yuhan Wang ⋅ Haoqin Tu ⋅ Hardy Chen ⋅ Zijun Wang ⋅ Cihang Xie ⋅ Yuyin Zhou
→ CLIP Models Generalize Less Than Compositional Benchmarks Suggest
Shuman Peng ⋅ Arnas Uselis ⋅ Darina Koishigarina ⋅ Martin Ester ⋅ Seong Joon Oh
→ HINT: Task Demonstrations for Hierarchical Inference in Abstract Reasoning
Nirlipta Pande ⋅ Georg Niess ⋅ Julian Gutheil ⋅ Roman Kern ⋅ Robert Legenstein ⋅ Robert Peharz
→ Fixed-Point Reasoning: Stable and Adaptive Deep Looped Models
Sajad Movahedi ⋅ Shlomo L Feigin ⋅ Vera Milovanović ⋅ Alexander Theus ⋅ Thomas Hofmann ⋅ Valentina Boeva ⋅ T. Konstantin Rusch ⋅ Antonio Orvieto
→ Sample Complexity of Scientific Discovery: PAC Learnability of Compositional Function Trees
Şuayp Talha Kocabay ⋅ Talha Rüzgar Akkuş ⋅ Kerem Yalçın
→ Introspective Coupling: LMs Explain Themselves Better Than Training Targets
Carl Guo ⋅ Laura Ruis ⋅ Jacob Andreas ⋅ Belinda Li
→ The Spurious Composition Problem: Conditional Independence as a Necessary and Sufficient Condition for Systematic Generalization
Siddharth Karuturi ⋅ Kaustubh Bukkapatnam ⋅ Soham Batra ⋅ Laksh Patel ⋅ Tanush A Shastry
→ MoTVLA: A Vision-Language-Action Model with Unified Fast-Slow Reasoning
Wenhui (Oscar) Huang ⋅ Changhe Chen ⋅ Han Qi ⋅ Chen Lv ⋅ Yilun Du ⋅ Heng Yang
→ Stop Probing, Start Coding: Why Linear Probes and Sparse Autoencoders Fail at Compositional Generalisation
Vitória Barin-Pacela ⋅ Shruti Joshi ⋅ Isabela Camacho ⋅ Simon Lacoste-Julien ⋅ David Klindt
→ Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer?
Franka Bause ⋅ Jonas Niederle ⋅ Martin Pawelczyk ⋅ Rebekka Burkholz