| arithmetic-reasoning-on-gsm8k | Qwen2.5-7B-Instruct (GRPO to ES) | – | Accuracy: 92.3 |
| code-generation-on-mbpp | Qwen2.5-7B-Instruct (ES to GRPO) | – | Accuracy: 75.2 |
| mathematical-reasoning-on-aime-2025 | DeepSeek-R1-Distill-Qwen-1.5B (GRPO to ES) | – | Accuracy: 24.5 |
| mathematical-reasoning-on-aime24 | DeepSeek-R1-Distill-Qwen-1.5B (ES to GRPO) | – | Acc: 29.0 |
| mathematical-reasoning-on-amc23 | DeepSeek-R1-Distill-Qwen-1.5B (ES to GRPO) | – | Acc: 72.3 |
| mathematical-reasoning-on-math-500 | DeepSeek-R1-Distill-Qwen-1.5B (GRPO to ES) | – | Accuracy: 84.5 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | DeepSeek-R1-Distill-Qwen-1.5B (GRPO to ES) | – | Accuracy: 30.6 |