Preprint
Jul 2026
Role-Break in Attention Heads: Understanding and Detecting Hallucinations in VLMs
A lightweight linear detector is built on top of Role-Break that requires no fine-tuning of the VLM, whose feature dimension stays below 5,000 and reaches an average AUROC of 93.23 across six VLMs and four benchmarks.
Mingyu Wang, Weilin Jin, Wenbo Li et al.
· 0 citations