MedHallu Benchmark

What is MedHallu?

MedHallu is a comprehensive benchmark dataset designed to evaluate the ability of large language models to detect hallucinations in medical question-answering…

Released

2025-02-20

Evaluates

其他, Other, Medical, 科学智能, AI for Science, 科学推理, Scientific Reasoning, 不支持, Unsupported

Openness

restricted

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/MedHallu

MedHallu paper, code and dataset

• Paper https://arxiv.org/abs/2502.14302
• Code https://github.com/MedHallu/MedHalu
• Dataset https://huggingface.co/datasets/UTAustin-AIHealth/MedHallu

Which sources cite MedHallu?

• MedHallu

Related 其他, Other, Medical, 科学智能, AI for Science, 科学推理, Scientific Reasoning, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• Ada-LEval
• AECBench
• AFQMC
• AgentBoard