IRIS: Implicit Reward-Guided Internal Sifting for Mitigating Multimodal Hallucination

Hallucination remains a fundamental challenge for Multimodal Large Language Models (MLLMs). While Direct Preference Optimization (DPO) is a key alignment framework, existing approaches often rely heavily on costly external evaluators for scoring or rewriting, incurring off-policy learnability gaps and discretization loss. Due to the lack of access to internal states, such feedback overlooks the fine-grained conflicts between different modalities that lead to hallucinations during generation. To address this issue, we propose IRIS (Implicit Reward-Guided Internal Sifting), which leverages continuous implicit rewards in the native log-probability space to preserve fine-grained preference information and capture internal modal competition. After an SFT warm-up, IRIS performs on-policy preference alignment by sifting self-generated responses sampled from the current policy. These responses are then ranked with multimodal implicit rewards to form preference pairs that drive optimization toward resolving modal conflicts. Extensive experiments demonstrate that IRIS achieves highly competitive performance on key hallucination benchmarks using only 5.7k samples, without requiring any external feedback during preference alignment. These results confirm that IRIS provides an efficient and principled paradigm for mitigating MLLM hallucinations. Code is available \href{https://github.com/ShawnLee0910/IRIS}{here}.

When an AI assistant answers questions about an image, it may invent details that are not actually there. IRIS addresses this by having the model produce multiple answers and then use its own internal signals to choose which answers are more supported by the image. This makes preference training cheaper and helps the model give more faithful image-based responses.