通过潜在空间引导减少视觉-语言模型中的幻觉
Reducing Hallucinations in Vision-Language Models via Latent Space Steering
AI总结：
本文提出视觉与文本干预（VTI）方法，通过在推理时引导潜在空间增强视觉特征稳定性，以任务无关且零额外成本的方式有效减少大视觉-语言模型中的幻觉。
摘要：
幻觉对大视觉-语言模型（LVLMs）在应用中的部署构成了挑战。与大语言模型（LLMs）不同，LVLMs 中的幻觉通常源于视觉输入与文本输出之间的错位。本文研究了幻觉的底层机制，重点关注 LVLMs 区别于大语言模型（LLMs）的独特结构。我们发现，幻觉往往源于文本解码器对视觉输入的敏感性，这是图像编码器和文本解码器分别预训练时的一种自然现象。受此启发，我们提出了视觉与文本干预（Visual and Textual Intervention，VTI），这是一种新技术，旨在通过在推理过程中引导潜在空间表示来增强视觉特征的稳定性，从而减少幻觉。作为一种与任务无关的测试时干预方法，VTI 可以轻松应用于任何问题，且无需额外成本。大量实验表明，VTI 能够有效减少幻觉，并在多个指标上优于基线方法，凸显了视觉特征稳定性在 LVLMs 中的关键作用。
英文摘要：
Hallucination poses a challenge to the deployment of large vision-language models (LVLMs) in applications. Unlike in large language models (LLMs), hallucination in LVLMs often arises from misalignments between visual inputs and textual outputs. This paper investigates the underlying mechanisms of hallucination, focusing on the unique structure of LVLMs that distinguishes them from large language models (LLMs). We identify that hallucinations often arise from the sensitivity of text decoders to vision inputs, a natural phenomenon when image encoders and text decoders are pre-trained separately. Inspired by this, we introduce Visual and Textual Intervention (VTI), a novel technique designed to reduce hallucinations by steering latent space representations during inference to enhance the stability of vision features. As a task-agnostic test-time intervention, VTI can be easily applied to any problem without additional cost. Extensive experiments demonstrate that it can effectively reduce hallucinations and outperform baseline methods across multiple metrics, highlighting the critical role of vision feature stability in LVLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
