Reducing Object Hallucination in LVLMs via Emphasizing Image-negative Tokens
For practitioners deploying LVLMs, this work offers a simple training-time fix to reduce object hallucination, though the gains are incremental over existing methods.
The paper addresses object hallucination in large vision-language models (LVLMs) by identifying that most generated tokens are minimally influenced by image information, and proposes adjusting training weights based on visual dependence and filtering hallucination-prone data, achieving reduced hallucination without extra inference cost across three LVLM variants.
Object hallucination is a significant challenge that hinders the application of large vision-language models (LVLMs) in practice. We hypothesize that one possible origin of hallucination is the model's tendency to prioritize text generation over meaningful interaction with images. To explore this, we examine the generation process and categorize text tokens into three groups: image-positive, invariant, and negative, based on their visual dependence on input image tokens. Our analysis reveals that most generated tokens are minimally influenced by the image information. This suggests that during the model's training stage, more emphasis is placed on learning how to follow textual instructions, rather than extracting information from images. Based on this finding, we propose adjusting the training weights of different tokens depending on their visual dependence to control hallucination. Additionally, we remove a portion of the training data that potentially contains more hallucinations as a data filtering strategy. Both methods achieve a reduction in hallucination without compromising response length or introducing additional computational costs during inference. We validate our methods across three LVLM variants, demonstrating the effectiveness and general applicability.