List your company

Collective Reasoning on HIDDENBENCH 18 tasks

80AccuracyGPT-4.1
0.64821.24941.8562.451May 15, 2025

Evaluation results

MethodLinks
GPT-4.1Communication Protocol=Exchange-then-Decide (Structured)2025.0580
Gemini-2.5-Flash-LiteCommunication Protocol=Exchange-then-Decide (Structured)2025.0574.3
Gemini-2.5-FlashCommunication Protocol=Exchange-then-Decide (Structured)2025.0572.7
Gemini-2.5-FlashCommunication Protocol=Baseline2025.0517.3
Gemini-2.5-Flash-LiteCommunication Protocol=Baseline2025.054.3
GPT-4.1Communication Protocol=Baseline2025.053.7