Flames Benchmark

What is Flames?

Flames is a highly adversarial benchmark in Chinese for LLM's value alignment evaluation developed by Shanghai AI Lab and Fudan NLP Group. Flames meticulously…

Released

2024-03-13

Evaluates

安全, Safety, NAACL 2024, 大语言模型, LLM, 安全对齐, Safety Alignment, 合作共建, Co-Built, 不支持, Unsupported

Openness

restricted

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

Flames paper and code

• Paper https://arxiv.org/abs/2311.06899
• Code https://github.com/AIFlames/Flames

Which sources cite Flames?

• Flames

Related 安全, Safety, NAACL 2024, 大语言模型, LLM, 安全对齐, Safety Alignment, 合作共建, Co-Built, 不支持, Unsupported benchmarks

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