让多模态大模型看清图像，通过动态调整中间层信息减少幻觉。
MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation
- 基于视觉识别早于生成的发现，动态融合中间层特征修正输出。
- 在多个基准上显著降低幻觉率，效果优于现有方法。
- 无需修改模型结构，适配各类解码策略和多模态模型。
多模态大语言模型（MLLMs）常出现幻觉现象，但其根本原因尚不明确。本文通过实证分析发现，尽管模型在最终输出中错误生成物体，但在前几层已能正确识别视觉物体。我们推测这可能源于语言模型强大的知识先验压制了视觉信息，导致幻觉。为此，提出一种新型动态修正解码方法DeCo，自适应选择合适前层，并按比例将知识融入最终层以调整输出概率。DeCo与模型无关，可无缝集成至多种经典解码策略，并适用于不同MLLM。在多个常用基准上评估显示，其相比基线大幅降低幻觉率，展现出显著缓解幻觉的潜力。代码已开源：https://github.com/zjunlp/DeCo。
原文摘要 · Abstract (English)
Multimodal Large Language Models (MLLMs) frequently exhibit hallucination phenomena, but the underlying reasons remain poorly understood. In this paper, we present an empirical analysis and find that, although MLLMs incorrectly generate the objects in the final output, they are actually able to recognize visual objects in the preceding layers. We speculate that this may be due to the strong knowledge priors of the language model suppressing the visual information, leading to hallucinations. Motivated by this, we propose a novel dynamic correction decoding method for MLLMs DeCo, which adaptively selects the appropriate preceding layers and proportionally integrates knowledge into the final layer to adjust the output logits. Note that DeCo is model agnostic and can be seamlessly incorporated with various classic decoding strategies and applied to different MLLMs. We evaluate DeCo on widely-used benchmarks, demonstrating that it can reduce hallucination rates by a large margin compared to baselines, highlighting its potential to mitigate hallucinations. Code is available at https://github.com/zjunlp/DeCo.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。