List your company

Question Answering on MMLU (TSR, ASR)

88.25Task Success Rate (TSR)STAR
57.3165.342573.37581.4075May 27, 2026

Evaluation results

MethodLinks
STARAgent Model=Qwen-Plus, Attack Type=Group Collusive Attack2026.0588.257.5
G-SafeguardAgent Model=Qwen-Plus, Attack Type=Group Collusive Attack2026.0584.510.25
STARAgent Model=GPT-4-Turbo, Attack Type=Group Collusive Attack2026.0580.512.75
STARAgent Model=GPT-3.5-Turbo, Attack Type=Group Collusive Attack2026.0577.512
G-SafeguardAgent Model=GPT-4-Turbo, Attack Type=Group Collusive Attack2026.0568.7521.75
G-SafeguardAgent Model=GPT-3.5-Turbo, Attack Type=Group Collusive Attack2026.0558.535.5