# UniProbe：面向大型视觉语言模型的可学习 token 级幻觉检测器

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-08-11 20:01
- AIHOT 分数：51
- AIHOT 链接：https://aihot.virxact.com/items/cmsww2rj5067drox7lkd2s203
- 原文链接：https://arxiv.org/abs/2608.10835

## AI 摘要

UniProbe 提出一种轻量级、统一的可学习检测器，通过单次前向传播对冻结 LVLM 的异构计算轨迹建模，利用 GNN、ViT 和 GRU 交替处理空间、关系与序列证据，实现 token 级幻觉定位。在多种 LVLM 主干上达到 SOTA 检测性能，解码时可将物体幻觉降低最多 55%，延迟仅为标准生成的 1.06 倍。

## 正文

Large Vision-Language Models (LVLMs) achieve impressive visual reasoning and dialogue capabilities, yet frequently hallucinate content unsupported by the visual input. Effective mitigation requires token-level localization, enabling targeted intervention without discarding the entire response. Existing detectors require expensive full-model fine-tuning, rely on external verifiers that ignore the model's generation process, or reduce internal signals to isolated features and hand-crafted statistics, discarding spatial, sequential, and relational structure. We introduce UniProbe, a lightweight, unified, learnable detector that models a frozen LVLM's heterogeneous computational trace from a single forward pass. UniProbe constructs a directed graph over image patches, query tokens, and generated tokens, with attention weights encoding their relations. It processes this trace with alternating structure-aware modules: a GNN for relational evidence, a ViT for 2-D visual geometry, and a GRU for response order. Interleaving them allows spatial, relational, and sequential evidence to interact throughout the detector. We further develop a streaming variant for hallucination-aware decoding, which detects and resamples hallucinated tokens during generation, and a self-adaptation strategy aligning the detector with the LVLM's own generations. Across diverse LVLM backbones, UniProbe achieves state-of-the-art token-level and object-hallucination detection. During decoding, it reduces object hallucinations by up to 55% at 1.06times the latency of standard generation.