Reasoning on Reasoning Gym Puzzle24
60.67Success RateCBRL GRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| CBRL GRPOBackbone Model=Qwen2.5-3B-Instruct2026.03 | 60.67 | |
| GRPOBackbone Model=Qwen2.5-3B-Instruct2026.03 | 48 | |
| CBRL GRPOBackbone Model=Llama-3.2-3B-Instruct2026.03 | 22 | |
| GRPOBackbone Model=Llama-3.2-3B-Instruct2026.03 | 20.67 | |
| Baseline w/ Few ShotsBackbone Model=Qwen2.5-3B-Instruct2026.03 | 6.67 | |
| BaselineBackbone Model=Qwen2.5-3B-Instruct2026.03 | 4.33 | |
| Baseline w/ Few ShotsBackbone Model=Llama-3.2-3B-Instruct2026.03 | 4 | |
| BaselineBackbone Model=Llama-3.2-3B-Instruct2026.03 | 2.67 |