MedHEval：医疗大型视觉语言模型中的幻觉与缓解策略基准测试
MedHEval: Benchmarking Hallucinations and Mitigation Strategies in Medical Large Vision-Language Models
AI总结：
本文提出MedHEval基准，将医疗大型视觉语言模型幻觉分为视觉误解、知识缺陷和上下文不一致三类，系统评估了11个模型和7种缓解技术，揭示了现有方法局限性并推动更可信模型开发。
摘要：
大型视觉语言模型（LVLMs）在医疗领域日益重要，然而由于专业知识有限及医疗应用的复杂性，医疗大型视觉语言模型（Med-LVLMs）频繁产生幻觉。现有基准无法基于幻觉的根本原因有效评估，且缺乏对缓解策略的评估。为弥补此空白，我们引入了MedHEval，这是一种新型基准，通过将其分为三种根本原因：视觉误解、知识缺陷和上下文不一致，来系统评估Med-LVLMs中的幻觉与缓解策略。我们构建了包含综合评估指标的多样化封闭式与开放式医疗VQA数据集，以评估这些幻觉类型。我们在11种流行的(Med)-LVLMs上进行了广泛实验，并评估了7种最先进的幻觉缓解技术。结果显示，Med-LVLMs难以应对不同原因引发的幻觉，而现有缓解方法效果有限，特别是对于基于知识和上下文的错误。这些发现强调了对改进的匹配训练和专门缓解策略的需求，以增强Med-LVLMs的可靠性。MedHEval建立了一个用于评估和缓解医疗幻觉的标准化框架，指导更值得信赖的Med-LVLMs的开发。
英文摘要：
Large Vision Language Models (LVLMs) are becoming increasingly important in the medical domain, yet Medical LVLMs (Med-LVLMs) frequently generate hallucinations due to limited expertise and the complexity of medical applications. Existing benchmarks fail to effectively evaluate hallucinations based on their underlying causes and lack assessments of mitigation strategies. To address this gap, we introduce MedHEval, a novel benchmark that systematically evaluates hallucinations and mitigation strategies in Med-LVLMs by categorizing them into three underlying causes: visual misinterpretation, knowledge deficiency, and context misalignment. We construct a diverse set of close- and open-ended medical VQA datasets with comprehensive evaluation metrics to assess these hallucination types. We conduct extensive experiments across 11 popular (Med)-LVLMs and evaluate 7 state-of-the-art hallucination mitigation techniques. Results reveal that Med-LVLMs struggle with hallucinations arising from different causes while existing mitigation methods show limited effectiveness, especially for knowledge- and context-based errors. These findings underscore the need for improved alignment training and specialized mitigation strategies to enhance Med-LVLMs' reliability. MedHEval establishes a standardized framework for evaluating and mitigating medical hallucinations, guiding the development of more trustworthy Med-LVLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
