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

Defense Robustness on GCG Attack

100Keyword Match Rate

ES2

17.8439.1760.581.83Mar 1, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.03
10010092
2026.03
10010098
2026.03
999988
2026.03
989894
2026.03
979593
2026.03
969484
2026.03
969493
2026.03
908873
2026.03
898878
2026.03
898884
2026.03
888787
2026.03
797473
2026.03
565044
2026.03
282422
2026.03
252424
2026.03
211919