List your company

Data Contamination Detection on AIME 2025

76F1 ScoreSelf-Critique
9.4426.724461.28Oct 10, 2025

Evaluation results

MethodLinks
Self-CritiqueTarget Model=Qwen2.5-7B-Instruct2025.10760.72
Entropy-NoiseTarget Model=DeepSeek-Math-7B-Instruct2025.10720.69
Self-CritiqueTarget Model=DeepSeek-Math-7B-Instruct2025.10710.61
Entropy-TempTarget Model=DeepSeek-Math-7B-Instruct2025.10700.54
CDDTarget Model=Qwen2.5-7B-Instruct2025.10670.52
Entropy-NoiseTarget Model=Qwen2.5-7B-Instruct2025.10670.63
Min-K%++Target Model=DeepSeek-Math-7B-Instruct2025.10670.56
CDDTarget Model=DeepSeek-Math-7B-Instruct2025.10650.51
PPLTarget Model=DeepSeek-Math-7B-Instruct2025.10620.41
RecallTarget Model=Qwen2.5-7B-Instruct2025.10550.65
Min-K%Target Model=Qwen2.5-7B-Instruct2025.10520.44
RecallTarget Model=DeepSeek-Math-7B-Instruct2025.10520.56
Min-K%++Target Model=Qwen2.5-7B-Instruct2025.10460.45
PPLTarget Model=Qwen2.5-7B-Instruct2025.10420.56
Entropy-TempTarget Model=Qwen2.5-7B-Instruct2025.10120.42
Min-K%Target Model=DeepSeek-Math-7B-Instruct2025.10120.4