List your company

Question Answering on LogiQA (TSR/ASR Metrics)

74.5Task Success Rate (TSR)STAR
43.351.459.567.6May 27, 2026

Evaluation results

MethodLinks
STARAgent Model=Qwen-Plus, Attack Type=Group Collusive Attack2026.0574.518.25
G-SafeguardAgent Model=Qwen-Plus, Attack Type=Group Collusive Attack2026.0571.7522
STARAgent Model=GPT-3.5-Turbo, Attack Type=Group Collusive Attack2026.055235.5
STARAgent Model=GPT-4-Turbo, Attack Type=Group Collusive Attack2026.0549.541.75
G-SafeguardAgent Model=GPT-4-Turbo, Attack Type=Group Collusive Attack2026.0548.7542.5
G-SafeguardAgent Model=GPT-3.5-Turbo, Attack Type=Group Collusive Attack2026.0544.546.25