Score: 2

Self-Augmented Visual Contrastive Decoding

Published: October 15, 2025 | arXiv ID: 2510.13315v1

By: Eun Woo Im, Muhammad Kashif Ali, Vivek Gupta

Potential Business Impact:

Makes AI less likely to make up answers.

Business Areas:
Image Recognition Data and Analytics, Software

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.

Country of Origin
πŸ‡¨πŸ‡³ πŸ‡ΊπŸ‡Έ China, United States

Page Count
23 pages

Category
Computer Science:
CV and Pattern Recognition