List your company

Logical Reasoning on LogiQA (TSR, ASR)

48.75Task Success Rate (TSR)Qwen-Plus
2.9914.8726.7538.63May 27, 2026

Evaluation results

MethodLinks
Qwen-PlusAttack Scenario=Independent2026.0548.7547
Qwen-PlusAttack Scenario=Group Collusive2026.054551.25
Qwen-PlusAttack Scenario=Cooperative2026.054353.25
GPT-5-NanoAttack Scenario=Independent2026.0518.578.5
GPT-5-NanoAttack Scenario=Group Collusive2026.0511.2586.75
GPT-5-NanoAttack Scenario=Cooperative2026.056.591.25
GPT-3.5-TurboAttack Scenario=Group Collusive2026.055.593.75
GPT-3.5-TurboAttack Scenario=Cooperative2026.05594.75
GPT-3.5-TurboAttack Scenario=Independent2026.054.7594.5