ChainMPQ: Interleaved Text-Image Reasoning Chains for Mitigating Relation Hallucinations
This addresses reliability issues in multimodal AI systems, particularly for applications requiring accurate relational understanding, but it is incremental as it builds on existing LVLM frameworks without introducing new training paradigms.
The paper tackles the problem of relation hallucinations in Large Vision-Language Models, which are the most common type of hallucination but have been understudied, and proposes ChainMPQ, a training-free method that reduces these hallucinations by using interleaved text-image reasoning chains.
While Large Vision-Language Models (LVLMs) achieve strong performance in multimodal tasks, hallucinations continue to hinder their reliability. Among the three categories of hallucinations, which include object, attribute, and relation, relation hallucinations account for the largest proportion but have received the least attention. To address this issue, we propose ChainMPQ (Multi-Perspective Questions guided Interleaved Chain of Image and Text), a training-free method that improves relational inference in LVLMs by utilizing accumulated textual and visual memories. ChainMPQ first extracts subject and object keywords from the question to enhance the corresponding image regions. It then constructs multi-perspective questions that focus on the three core components of a relationship: the subject, the object, and the relation that links them. These questions are sequentially input to the model, with textual and visual memories from earlier steps providing supporting context for subsequent ones, thereby forming an interleaved chain of images and text that guides progressive relational reasoning. Experiments on multiple LVLMs and benchmarks show that ChainMPQ substantially reduces relation hallucinations, while ablation studies further validate the effectiveness of its three core modules.