Mathematical Reasoning on AIME25 (Pass@1 (Avg@32))
28.3Pass@1 (Avg@32)Mix-RL
Evaluation Results
| Method | Links | |
|---|---|---|
| Mix-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 28.3 | |
| Mix-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 28 | |
| Unfold-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Unfold2026.02 | 27.8 | |
| Fold-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 27.8 | |
| Fold-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 26.9 | |
| Unfold-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Unfold2026.02 | 26.7 | |
| Cold-StartModel Backbone=Qwen3-4B-Base, Gen Mode=Unfold2026.02 | 25.4 | |
| Unfold-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 25.1 | |
| Unfold-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 25 | |
| Cold-StartModel Backbone=Qwen2.5-Math-7B, Gen Mode=Unfold2026.02 | 24.6 | |
| Cold-StartModel Backbone=Qwen2.5-Math-7B, Gen Mode=Fold2026.02 | 23.1 | |
| Zero-RLModel Backbone=Qwen3-4B-Base, Gen Mode=Unfold2026.02 | 22.5 | |
| Cold-StartModel Backbone=Qwen3-4B-Base, Gen Mode=Fold2026.02 | 22 | |
| Zero-RLModel Backbone=Qwen2.5-Math-7B, Gen Mode=Unfold2026.02 | 18.1 |