List your company

Mathematical Problem Solving on John O'Bryan Competition (Overall)

651Total ProblemsDeepSeek-R1 (Teacher)
618.45634.725651667.275Jun 30, 2026

Evaluation results

MethodLinks
DeepSeek-R1 (Teacher)2026.0665159491.4———
Qwen2.5-7B Base2026.0665142164.67———
Qwen2.5-7BIterations=1000, Status=overfit2026.0665144067.59———
Qwen2.5-7BIterations=200, Aggregate=mean2026.06651—69.43707164.3
Qwen2.5-7BIterations=200, Seed=422026.0665145269.43———
Qwen2.5-7BIterations=200, Seed=1232026.0665145169.28———
Qwen2.5-7BIterations=200, Seed=2562026.0665145269.43———
Qwen2.5-7BIterations=200, Seed=5122026.0665145369.58———
Qwen2.5-7BIterations=200, Seed=9992026.0665145269.43———