Mathematical Reasoning on AIME 2025 1 (Accuracy)
25AccuracyTA-OPD+Ent.
Evaluation Results
| Method | Links | |
|---|---|---|
| TA-OPD+Ent.Setting=Qwen3-8B (GRPO) -> Qwen3-4B, Supervised-token budget=10%2026.05 | 25 | |
| TIPSetting=Qwen3-8B (GRPO) -> Qwen3-4B, Supervised-token budget=10%2026.05 | 23.67 | |
| EntropySetting=Qwen3-8B (GRPO) -> Qwen3-4B, Supervised-token budget=10%2026.05 | 23.33 | |
| Pure OPDSetting=Qwen3-8B (GRPO) -> Qwen3-4B, Supervised-token budget=10%2026.05 | 21.67 | |
| TA-OPD+Ent.Setting=Qwen3-14B -> Qwen3-4B, Supervised-token budget=10%2026.05 | 21.67 | |
| Pure OPDSetting=Qwen3-14B -> Qwen3-4B, Supervised-token budget=10%2026.05 | 21.66 | |
| TA-OPDSetting=Qwen3-8B (GRPO) -> Qwen3-4B, Supervised-token budget=10%2026.05 | 21.31 | |
| TA-OPDSetting=Qwen3-14B -> Qwen3-4B, Supervised-token budget=10%2026.05 | 18.34 | |
| TIPSetting=Qwen3-14B -> Qwen3-4B, Supervised-token budget=10%2026.05 | 16.66 | |
| Pure OPDSetting=Qwen3-4B -> Qwen3-1.7B, Supervised-token budget=10%2026.05 | 15 | |
| TA-OPDSetting=Qwen3-4B -> Qwen3-1.7B, Supervised-token budget=10%2026.05 | 15 | |
| EntropySetting=Qwen3-14B -> Qwen3-4B, Supervised-token budget=10%2026.05 | 15 | |
| BaseSetting=Qwen3-8B (GRPO) -> Qwen3-4B, Supervised-token budget=10%2026.05 | 13.33 | |
| BaseSetting=Qwen3-14B -> Qwen3-4B, Supervised-token budget=10%2026.05 | 13.33 | |
| TIPSetting=Qwen3-4B -> Qwen3-1.7B, Supervised-token budget=10%2026.05 | 11.67 | |
| BaseSetting=Qwen3-4B -> Qwen3-1.7B, Supervised-token budget=10%2026.05 | 10 | |
| TA-OPD+Ent.Setting=Qwen3-4B -> Qwen3-1.7B, Supervised-token budget=10%2026.05 | 8.34 | |
| EntropySetting=Qwen3-4B -> Qwen3-1.7B, Supervised-token budget=10%2026.05 | 8.33 | |
| TA-OPD+Ent.Setting=DeepSeek-R1-Distill-Qwen-14B -> Qwen2.5-3B, Supervised-token budget=10%2026.05 | 3.33 | |
| BaseSetting=DeepSeek-R1-Distill-Qwen-14B -> Qwen2.5-3B, Supervised-token budget=10%2026.05 | 1.67 | |
| Pure OPDSetting=DeepSeek-R1-Distill-Qwen-14B -> Qwen2.5-3B, Supervised-token budget=10%2026.05 | 1.67 | |
| EntropySetting=DeepSeek-R1-Distill-Qwen-14B -> Qwen2.5-3B, Supervised-token budget=10%2026.05 | 1.67 | |
| TIPSetting=DeepSeek-R1-Distill-Qwen-14B -> Qwen2.5-3B, Supervised-token budget=10%2026.05 | 1.67 | |
| TA-OPDSetting=DeepSeek-R1-Distill-Qwen-14B -> Qwen2.5-3B, Supervised-token budget=10%2026.05 | 1.67 |