Reasoning on Reasoning Gym Word Sorting
84.33Success RateCBRL GRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| CBRL GRPOBackbone Model=Llama-3.2-3B-Instruct2026.03 | 84.33 | |
| GRPOBackbone Model=Llama-3.2-3B-Instruct2026.03 | 80 | |
| CBRL GRPOBackbone Model=Qwen2.5-3B-Instruct2026.03 | 75.67 | |
| GRPOBackbone Model=Qwen2.5-3B-Instruct2026.03 | 53.33 | |
| Baseline w/ Few ShotsBackbone Model=Llama-3.2-3B-Instruct2026.03 | 13.33 | |
| Baseline w/ Few ShotsBackbone Model=Qwen2.5-3B-Instruct2026.03 | 8.67 | |
| BaselineBackbone Model=Llama-3.2-3B-Instruct2026.03 | 8.67 | |
| BaselineBackbone Model=Qwen2.5-3B-Instruct2026.03 | 1.33 |