幻觉始于显著性下降
Hallucination Begins Where Saliency Drops
AI总结：
本文提出LVLMs-Saliency框架，通过融合注意力权重与输入梯度，利用显著性引导拒绝采样和局部连贯性强化机制，有效减少幻觉并提升模型可靠性。
摘要：
最近的研究已经考察了在大型视觉-语言模型（LVLMs）中注意力动态，以检测幻觉。然而，现有方法在可靠地区分幻觉输出与事实性输出方面仍然有限，因为它们仅依赖于前向传递的注意力模式，并忽视了通过网络传播的梯度信号，这些信号揭示了token影响力如何传播。为弥合这一差距，我们引入了LVLMs-Saliency，一种梯度感知的诊断框架，通过融合注意力权重与输入梯度来量化每个输出token的视觉基础强度。我们的分析揭示了一个决定性的模式：幻觉经常出现在前序输出token对下一个token的预测表现出低显著性时，这表明上下文记忆保留出现了断裂。利用这一见解，我们提出了一种双机制推理时间框架以减轻幻觉：（1）显著性引导拒绝采样（SGRS），在自回归解码过程中动态过滤候选token，通过拒绝那些显著性低于上下文自适应阈值的token，从而防止破坏连贯性的token进入输出序列；（2）局部连贯性强化（LocoRE），一个轻量级、即插即用的模块，通过加强当前token对最近前序token的注意力，积极对抗由LVLMs-Saliency识别出的上下文遗忘行为。在多个LVLMs上的广泛实验表明，我们的方法显著降低了幻觉率，同时保持流畅性和任务性能，提供了一种稳健且可解释的解决方案，以增强模型可靠性。代码可在：https://github.com/zhangbaijin/LVLMs-Saliency
英文摘要：
Recent studies have examined attention dynamics in large vision-language models (LVLMs) to detect hallucinations. However, existing approaches remain limited in reliably distinguishing hallucinated from factually grounded outputs, as they rely solely on forward-pass attention patterns and neglect gradient-based signals that reveal how token influence propagates through the network. To bridge this gap, we introduce LVLMs-Saliency, a gradient-aware diagnostic framework that quantifies the visual grounding strength of each output token by fusing attention weights with their input gradients. Our analysis uncovers a decisive pattern: hallucinations frequently arise when preceding output tokens exhibit low saliency toward the prediction of the next token, signaling a breakdown in contextual memory retention. Leveraging this insight, we propose a dual-mechanism inference-time framework to mitigate hallucinations: (1) Saliency-Guided Rejection Sampling (SGRS), which dynamically filters candidate tokens during autoregressive decoding by rejecting those whose saliency falls below a context-adaptive threshold, thereby preventing coherence-breaking tokens from entering the output sequence; and (2) Local Coherence Reinforcement (LocoRE), a lightweight, plug-and-play module that strengthens attention from the current token to its most recent predecessors, actively counteracting the contextual forgetting behavior identified by LVLMs-Saliency. Extensive experiments across multiple LVLMs demonstrate that our method significantly reduces hallucination rates while preserving fluency and task performance, offering a robust and interpretable solution for enhancing model reliability. Code is available at: https://github.com/zhangbaijin/LVLMs-Saliency
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。
