Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Adversarial Robustness on HarmBench

56.25DR

Base

4.2517.7531.2544.75May 30, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
56.2546.2543.7567.5
2025.05
33.7527.54053.75
2025.05
32.53333.7558.75
2025.05
31.2530.537.555
2025.05
22.523.2522.548.75
2025.05
11.251623.7541.25
2025.05
7.51417.516.25
2025.05
7.5141523.19
2025.05
7.51310.6715
2025.05
6.2511.751516.25
2025.05
6.2512.7513.7513.75
2025.05
6.25116.2513.75