CriticBench

What is CriticBench?

CriticBench, a novel benchmark designed to comprehensively and reliably evaluate four key critique ability dimensions of LLMs: feedback, comparison, refinement…

Released

2024-02-22

Evaluates

理解, Understanding, 大语言模型, LLM, 逻辑推理, Reasoning, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

CriticBench paper, code and dataset

• Paper https://arxiv.org/abs/2402.13764
• Code https://github.com/open-compass/CriticBench
• Dataset https://huggingface.co/datasets/opencompass/CriticBench

Which sources cite CriticBench?

• CriticBench

Related 理解, Understanding, 大语言模型, LLM, 逻辑推理, Reasoning, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• ActiveFly-Bench
• Ada-LEval
• AdvancedMathBench
• AECBench