List your company

Factual Knowledge on MMLU-Pro

58.4EMGPT-4o-mini
49.66451.93254.256.468Jan 29, 2026

Evaluation results

MethodLinks
GPT-4o-minizero-shot-CoT=true, pass@1=true2026.0158.40.61
PIRzero-shot-CoT=true, pass@1=true, mode=interactive stronger US (gpt-4o-mini)2026.0152.871.32
Reasoning Basezero-shot-CoT=true, pass@1=true2026.0151.212.04
PIRzero-shot-CoT=true, pass@1=true, mode=interactive2026.0150.291.33
PIRzero-shot-CoT=true, pass@1=true, mode=self-play2026.01501.31