List your company

Reasoning on Reasoning Gym Word Sorting

84.33Success RateCBRL GRPO
-1.9920.4242.8365.24Mar 19, 2026

Evaluation results

MethodLinks
CBRL GRPOBackbone Model=Llama-3.2-3B-Instruct2026.0384.33
GRPOBackbone Model=Llama-3.2-3B-Instruct2026.0380
CBRL GRPOBackbone Model=Qwen2.5-3B-Instruct2026.0375.67
GRPOBackbone Model=Qwen2.5-3B-Instruct2026.0353.33
Baseline w/ Few ShotsBackbone Model=Llama-3.2-3B-Instruct2026.0313.33
Baseline w/ Few ShotsBackbone Model=Qwen2.5-3B-Instruct2026.038.67
BaselineBackbone Model=Llama-3.2-3B-Instruct2026.038.67
BaselineBackbone Model=Qwen2.5-3B-Instruct2026.031.33