Thinking Before Looking: Improving Multimodal LLM Reasoning via Mitigating Visual Hallucination
This addresses the issue of cross-modal biases and hallucinations in MLLMs, which is an incremental improvement over existing chain of thought methods.
The paper tackles the problem of visual hallucination in multimodal large language models (MLLMs) by proposing the Visual Inference Chain (VIC) framework, which constructs reasoning chains using textual context before visual input, resulting in significantly improved zero-shot performance across vision-related tasks.
Multimodal large language models (MLLMs) have advanced the integration of visual and linguistic modalities, establishing themselves as the dominant paradigm for visual-language tasks. Current approaches like chain of thought (CoT) reasoning have augmented the cognitive capabilities of large language models (LLMs), yet their adaptation to MLLMs is hindered by heightened risks of hallucination in cross-modality comprehension. In this paper, we find that the thinking while looking paradigm in current multimodal CoT approaches--where reasoning chains are generated alongside visual input--fails to mitigate hallucinations caused by misleading images. To address these limitations, we propose the Visual Inference Chain (VIC) framework, a novel approach that constructs reasoning chains using textual context alone before introducing visual input, effectively reducing cross-modal biases and enhancing multimodal reasoning accuracy. Comprehensive evaluations demonstrate that VIC significantly improves zero-shot performance across various vision-related tasks, mitigating hallucinations while refining the reasoning capabilities of MLLMs. Our code repository can be found at https://github.com/Terry-Xu-666/visual_inference_chain.