General Reasoning on AGIEval (en)
2.132Speedup RatioMTP-D ensemble
Evaluation results
| Method | Links | |
|---|---|---|
| MTP-D ensembleLoop strategy=4 to 8, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 2.132 | |
| MTP-DLoop strategy=4 to 8, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 2.071 | |
| MTP-DLoop strategy=4 to 8, Training protocol=Continued pre-training, Training data size=350B tokens2026.03 | 2.068 | |
| MTP-DLoop strategy=1 to 8, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.964 | |
| MTP-DLoop strategy=4, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.94 | |
| MTPLoop strategy=4 to 8, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.87 | |
| MTP-DLoop strategy=4 to 8, Training protocol=Training Free, Training data size=70B tokens2026.03 | 1.851 | |
| MTP-DLoop strategy=4 to 16, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.766 | |
| MTPLoop strategy=1 to 8, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.764 | |
| MTP-DLoop strategy=4 to 8 to 16, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.735 | |
| MTP-DLoop strategy=1 to 16, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.535 | |
| MTP-DLoop strategy=4 to 16, Training protocol=Training Free, Training data size=70B tokens2026.03 | 1.51 | |
| MTP-DLoop strategy=1 to 8, Training protocol=Training Free, Training data size=70B tokens2026.03 | 1.337 | |
| MTP-DLoop strategy=1, Training protocol=Continued pre-training, Training data size=70B tokens2026.03 | 1.128 | |
| MTP-DLoop strategy=1 to 16, Training protocol=Training Free, Training data size=70B tokens2026.03 | 1.062 |