Self-Augmented Visual Contrastive Decoding
By: Eun Woo Im, Muhammad Kashif Ali, Vivek Gupta
Potential Business Impact:
Makes AI less likely to make up answers.
Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal capabilities, but they inherit the tendency to hallucinate from their underlying language models. While visual contrastive decoding has been proposed to mitigate this issue, existing methods often apply generic visual augmentations that disregard the specific context provided by the text query, limiting their effectiveness. This study introduces a novel training-free decoding strategy that addresses these limitations, featuring two key contributions. First, a self-augmentation prompting strategy that leverages the intrinsic knowledge of the model to dynamically align semantics between the query and the visual augmentation. Second, an adaptive thresholding algorithm that adaptively adjusts next token candidate size based on the output sparsity, utilizing full information from the logit distribution. Extensive experiments across four LVLMs and seven benchmarks demonstrate that the proposed decoding significantly enhances factual consistency compared to state-of-the-art decoding methods. This work highlights the importance of integrating query-dependent augmentation and entropy-aware decoding for improving effective generation of LVLMs.
Similar Papers
Med-VCD: Mitigating Hallucination for Medical Large Vision Language Models through Visual Contrastive Decoding
CV and Pattern Recognition
Makes AI doctors give more accurate answers.
Improving Alignment in LVLMs with Debiased Self-Judgment
CV and Pattern Recognition
Model learns to check its own answers better.
Improving Alignment in LVLMs with Debiased Self-Judgment
CV and Pattern Recognition
Model judges its own answers to be safer.