Mathematical Reasoning on AIME decontaminated 24 (Accuracy)
27.9AccuracyDAPO-Math-17k
Evaluation Results
| Method | Links | |
|---|---|---|
| DAPO-Math-17kBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 27.9 | |
| DAPO++Backbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 25.9 | |
| DeepScaleRBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 23.4 | |
| Skywork-OR1-RL-DataBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 17.3 | |
| DeepMath-103KBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 17.1 | |
| OpenR1-Math-220kBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 16.5 | |
| Qwen3-8B-BaseBackbone=Qwen3-8B, Sampling Strategy=Mean@322026.05 | 10.4 | |
| DAPO++Backbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 9.4 | |
| DeepMath-103KBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 8.4 | |
| DAPO-Math-17kBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 7 | |
| OpenR1-Math-220kBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 6.9 | |
| DeepScaleRBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 6.3 | |
| Skywork-OR1-RL-DataBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 5.8 | |
| Qwen3-1.7B-BaseBackbone=Qwen3-1.7B, Sampling Strategy=Mean@322026.05 | 4.7 |