SG-Bench

What is SG-Bench?

SG-Bench assess LLM safety across various tasks and prompt types. It integrates both generative and discriminative evaluation tasks and includes extended data…

Released

2024-10-29

Evaluates

安全, Safety, NeurIPS 2024, 大语言模型, LLM, 安全对齐, Safety Alignment, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

SG-Bench paper and code

• Paper https://arxiv.org/abs/2410.21965
• Code https://github.com/MurrayTom/SG-Bench

Which sources cite SG-Bench?

• SG-Bench

Related 安全, Safety, NeurIPS 2024, 大语言模型, LLM, 安全对齐, Safety Alignment, 不支持, Unsupported benchmarks

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