List your company

Mathematical Reasoning on MATH500, AMC23, AIME24, AIME25

54.2Majority ScoreSDRL-freq
31.94437.72243.549.278Jan 29, 2026

Evaluation results

MethodLinks
SDRL-freqBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.0154.257.73.5
SDRL-randBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.0153.755.61.9
DAPOBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.0152.852.90.1
SDRL-freqBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.0134.135.81.7
SDRL-randBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.0133.635.82.2
DAPOBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.0132.833.60.8