V-DPO: Mitigating Hallucination in Large Vision Language Models via Vision-Guided Direct Preference Optimization
Yuxi Xie, Guanzhen Li, Xu Xiao, Min‐Yen Kan · 2024
Large vision-language models (LVLMs) suffer from hallucination, resulting in misalignment between the output textual response and the input visual content.Recent research indicates that the over-reliance on the Large Language Model (LLM) backbone, as one cause of the LVLM hallucination, inherently introduces bias from language priors, leading to insufficient context attention to the visual inputs.We tackle this issue of hallucination by mitigating such over-reliance through preference learning.We propose Vision-guided Direct Preference Optimization (V-DPO) to enhance visual context learning at training time.To interpret the effectiveness and generalizability of V-DPO on different types of training data, we construct a synthetic dataset containing both response-and image-contrast preference pairs, compared against existing humanannotated hallucination samples.Our approach achieves significant improvements compared with baseline methods across various hallucination benchmarks.Our analysis indicates that V-DPO excels in learning from imagecontrast preference data, demonstrating its superior ability to elicit and understand nuances of visual context.