AI RESEARCH
Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations
arXiv CS.CV
•
ArXi:2604.04863v1 Announce Type: new Large vision-language models (LVLMs) achieve strong performance on visual reasoning tasks but remain highly susceptible to hallucination. Existing detection methods predominantly rely on coarse, whole-image measures of how an object token relates to the input image. This global strategy is limited: hallucinated tokens may exhibit weak but widely scattered correlations across many local regions, which aggregate into deceptively high overall relevance, thus evading the current global hallucination detectors.