Med-HVL: Automatic Medical Domain Hallucination Evaluation for Large Vision-Language Models

Qianqi Yan, Xuehai He, Xin Eric WangAAAI 2024 · OpenReview

Abstract

Advancements in Large Vision-Language Models (LVLMs) have made significant progress in integrating visual and textual data. However, their deployment in the medical domain is impeded by critical issues of hallucinations, asking for reliable evaluation metrics and methods. We define two novel metrics: Object Hallucination and Domain Knowledge Hallucination, to quantify the hallucination of LVLMs in the medical domain. We propose a scalable, automated evaluation framework, Med-HVL, to assess and mitigate hallucinations at both object and domain knowledge levels. We reveal a significant presence of hallucinations in the LVLMs, emphasizing the need for domain-specific adaptations and finetuning to enhance their reliability for medical applications.
BibTeX
@inproceedings{yan2024med,
  author = {Qianqi Yan and Xuehai He and Xin Eric Wang},
  title = {{Med-HVL: Automatic Medical Domain Hallucination Evaluation for Large Vision-Language Models}},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year = {2024},
  url = {https://openreview.net/forum?id=rxx8leoPy0},
}

Loading related content...

Figure