Adversarial Attack
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
81.6SIN tSuc
3
Apr 6, 2026
81.6Target Success Rate (tSuc)
3
Apr 6, 2026
100Attack Success Rate (ASR)
3
Mar 12, 2026
15.3Attack Success Rate
3
Mar 12, 2026
21.7Attack Success Rate (ASR)
3
Mar 12, 2026
48.88GPT-4o Attack Success Rate
3
Mar 6, 2026
63.42Success Rate (GPT-4o)
3
Mar 6, 2026
26.8Attack Success Rate (ASR)
3
Mar 5, 2026
80ASR
3
Feb 26, 2026
75.94Accuracy (%)
3
Feb 26, 2026
0.044Attack Success Rate
3
Feb 26, 2026
46.3Attack Success Rate
3
Feb 26, 2026
0.888Attack Success Rate
3
Feb 26, 2026
25.39Avg Execution Time (ms)
3
Feb 26, 2026
26.26Avg Query Time (ms)
3
Feb 26, 2026
4.23Avg Execution Time (ms/query)
3
Feb 26, 2026
39.7Drop
2
May 25, 2026
53.9Drop
2
May 25, 2026
3.93Tool Calls
2
Mar 27, 2026
57.9T-ASR
2
Mar 6, 2026
34.2T-ASR
2
Mar 6, 2026
40T-ASR
2
Mar 6, 2026
51T-ASR
2
Mar 6, 2026
60T-ASR
2
Mar 6, 2026
64T-ASR
2
Mar 6, 2026