List your company

Mathematical Reasoning on AIME 25 (Maj, Debate, Delta)

28.7Maj ScoreSDRL-freq
2.2849.1421622.858Jan 29, 2026

Evaluation results

MethodLinks
SDRL-freqBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.0128.7301.3
DAPOBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.0124.724-0.7
SDRL-randBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.0122.7241.3
SDRL-randBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.015.66.10.5
DAPOBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.0143.3-0.7
SDRL-freqBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.013.34.71.4