CHARM Benchmark
Explore score history ↗CHARM is the first benchmark for comprehensively and in-depth evaluating the commonsense reasoning ability of large language models (LLMs) in Chinese, which covers both globally known and Chinese-specific commonsense.
- Publisher
- Shanghai AI Laboratory (published the hub card)
- Released
- 2024-08-11
- Evaluates
- 推理, Reasoning, ACL 2024, 大语言模型, LLM, 逻辑推理
- Openness
- unknown
- Importer
- OpenCompass Hub
Source provenance
CHARM paper and code
- Paper · https://arxiv.org/abs/1809.05053
- Paper · https://arxiv.org/abs/2004.05986
- Paper · https://arxiv.org/abs/2005.00333
- Paper · https://arxiv.org/abs/2007.08124
- Paper · https://arxiv.org/abs/2112.10668
- Paper · https://arxiv.org/abs/2302.04752
- Paper · https://arxiv.org/abs/2306.09212
- Paper · https://arxiv.org/abs/2312.12853
- Paper · https://arxiv.org/abs/2403.14112
- Code
No reported scores are on record for this benchmark yet.