List your company

Question Answering on CSQA (TSR/ASR)

82.25Task Success Rate (TSR)STAR
27.1341.4455.7570.06May 27, 2026

Evaluation results

MethodLinks
STARAgent Model=Qwen-Plus, Attack Type=Group Collusive Attack2026.0582.257.25
STARAgent Model=GPT-3.5-Turbo, Attack Type=Group Collusive Attack2026.0573.7513.5
G-SafeguardAgent Model=Qwen-Plus, Attack Type=Group Collusive Attack2026.0567.524.5
STARAgent Model=GPT-4-Turbo, Attack Type=Group Collusive Attack2026.0562.526.25
G-SafeguardAgent Model=GPT-4-Turbo, Attack Type=Group Collusive Attack2026.0543.548.25
G-SafeguardAgent Model=GPT-3.5-Turbo, Attack Type=Group Collusive Attack2026.0529.2565