HaluEval

What is HaluEval?

HaluEval evaluates the performance of LLMs in recognizing hallucination. It includes 5,000 general user queries with ChatGPT responses and 30,000 task-specific…

Released

2023-10-23

Evaluates

安全, Safety, 大语言模型, LLM, 安全对齐, Safety Alignment, 事实可靠性, Factual Reliability, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

HaluEval paper, code and dataset

• Paper https://arxiv.org/pdf/2305.11747
• Code https://github.com/RUCAIBox/HaluEval
• Dataset https://huggingface.co/datasets/pminervini/HaluEval

Which sources cite HaluEval?

• HaluEval

Related 安全, Safety, 大语言模型, LLM, 安全对齐, Safety Alignment, 事实可靠性, Factual Reliability, 不支持, Unsupported benchmarks

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