TY  - RPRT
TI  - From Token Correction to Claim Admission: Training-Free Hallucination Mitigation for Large Vision-Language Models
AU  - Hyeonggeun Han
AU  - Taehyun Cho
AU  - Sangwoo Hong
PY  - 2026
UR  - https://arxiv.org/abs/2610.17706
ID  - 2610.17706
ER  -