Mathematical Reasoning on AMC (Pass@1 Avg@32)
73.8Pass@1 (Avg@32)Fold-RL
Evaluation Results
| Method | Links | |
|---|---|---|
| Fold-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 73.8 | |
| Unfold-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Unfold2026.02 | 73.2 | |
| Mix-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 72.8 | |
| Fold-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 72.2 | |
| Mix-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 71.9 | |
| Unfold-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Unfold2026.02 | 71.2 | |
| Unfold-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 70.2 | |
| Unfold-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 69.7 | |
| Cold-StartModel Backbone=Qwen2.5-Math-7B, Gen Mode=Unfold2026.02 | 65.4 | |
| Zero-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Unfold2026.02 | 65.4 | |
| Cold-StartModel Backbone=Qwen3-4B-Base, Gen Mode=Unfold2026.02 | 64.1 | |
| Cold-StartModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 62.4 | |
| Zero-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Unfold2026.02 | 58.9 | |
| Cold-StartModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 57.3 |