Mathematical Reasoning on MATH 500 (Maj, Debate, Delta)
88.8Maj ScoreSDRL-rand
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SDRL-randBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.01 | 88.8 | 86.6 | -2.2 | |
| SDRL-freqBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.01 | 87 | 85.9 | -1.1 | |
| DAPOBackbone=Qwen3-4B-Base, Debate system agents (N)=5, Debate rounds=12026.01 | 86.9 | 83.1 | -3.8 | |
| SDRL-freqBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.01 | 71.4 | 71.7 | 0.3 | |
| DAPOBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.01 | 70.9 | 71.1 | 0.2 | |
| SDRL-randBackbone=Qwen2.5-3B, Debate system agents (N)=5, Debate rounds=12026.01 | 70.7 | 71.4 | 0.7 |