Legal Moves Prediction on Chess Reasoning Dataset Legal Moves
96IoUgpt-oss-120b (Medium)
Evaluation Results
| Method | Links | |
|---|---|---|
| gpt-oss-120b (Medium)Training Stage=Baseline2026.04 | 96 | |
| SFT8 + SFT9 XLTraining Stage=RL2026.04 | 87 | |
| SFT8 XLTraining Stage=RL2026.04 | 79 | |
| SFT9 XLTraining Stage=RL2026.04 | 75 | |
| SFT8 [BM - All]Training Stage=RL2026.04 | 67 | |
| SFT6 [BM]Training Stage=RL2026.04 | 66 | |
| SFT7 [BL]Training Stage=RL2026.04 | 63 | |
| SFT9 [BL - All]Training Stage=RL2026.04 | 59 | |
| SFT3 [VABP]Training Stage=RL2026.04 | 58 | |
| SFT4 [FBA]Training Stage=RL2026.04 | 58 | |
| SFT5 [GS]Training Stage=RL2026.04 | 57 | |
| SFT8 + SFT9 XLTraining Stage=SFT2026.04 | 52 | |
| SFT4 [FBA]Training Stage=SFT2026.04 | 49 | |
| SFT9 XLTraining Stage=SFT2026.04 | 47 | |
| SFT7 [BL]Training Stage=SFT2026.04 | 46 | |
| SFT8 XLTraining Stage=SFT2026.04 | 46 | |
| SFT2 [RSA]Training Stage=RL2026.04 | 44 | |
| SFT8 [BM - All]Training Stage=SFT2026.04 | 44 | |
| Llama 4 MaverickTraining Stage=Baseline2026.04 | 43 | |
| SFT6 [BM]Training Stage=SFT2026.04 | 42 | |
| SFT3 [VABP]Training Stage=SFT2026.04 | 41 | |
| SFT9 [BL - All]Training Stage=SFT2026.04 | 40 | |
| SFT2 [RSA]Training Stage=SFT2026.04 | 37 | |
| SFT5 [GS]Training Stage=SFT2026.04 | 37 | |
| Qwen2.5 7B-InstructTraining Stage=Baseline2026.04 | 26 | |
| SFT1 [RSPM]Training Stage=SFT2026.04 | 26 | |
| SFT1 [RSPM]Training Stage=RL2026.04 | 17 |