AutoAdvExBench

What is AutoAdvExBench?

AutoAdvExBench is a benchmark designed to evaluate large language models' (LLMs) ability to autonomously exploit adversarial example defenses, directly…

Released

2025-03-03

Evaluates

安全, Safety, 智能体, Agent, 任务执行, Task Execution, 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/AutoAdvExBench

AutoAdvExBench paper and code

• Paper https://arxiv.org/abs/2503.01811
• Code https://github.com/ethz-spylab/AutoAdvExBench

Which sources cite AutoAdvExBench?

• AutoAdvExBench

Related 安全, Safety, 智能体, Agent, 任务执行, Task Execution, 安全对齐, Safety Alignment, 不支持, Unsupported benchmarks

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