@misc{indiciaee8f41d7d15e8,
  title = {From Token Correction to Claim Admission: Training-Free Hallucination Mitigation for Large Vision-Language Models},
  author = {Hyeonggeun Han and Taehyun Cho and Sangwoo Hong},
  year = {2026},
  url = {https://arxiv.org/abs/2610.17706},
  note = {Source identifier: 2610.17706}
}