UHGEval

What is UHGEval?

UHGEval(Unconstrained Hallucination Generation Evaluation) benchmark contains hallucinations generated by LLMs with minimal restrictions.

Released

2024-05-24

Evaluates

知识, Knowledge, 大语言模型, LLM, 知识储备, 事实可靠性, Factual Reliability, 不支持, Unsupported

Openness

open

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

UHGEval paper, code and dataset

• Paper https://arxiv.org/pdf/2311.15296
• Code https://github.com/IAAR-Shanghai/UHGEval
• Dataset https://huggingface.co/datasets/Ki-Seki/UHGEvalDataset

Which sources cite UHGEval?

• UHGEval

Related 知识, Knowledge, 大语言模型, LLM, 知识储备, 事实可靠性, Factual Reliability, 不支持, Unsupported benchmarks

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