Scientific Writing Evaluation on Scientific Writing Evaluation Tasks (including RevUtil) (test)
83RW ScoreSciRM
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SciRM2026.01 | 83 | 71 | 61 | 78 | 73 | |
| SciRM-RefAdditional Supervision=true2026.01 | 83 | 69 | 74 | 78 | 76 | |
| o3-miniType=Proprietary2026.01 | 79 | 60 | 78 | 84 | 75 | |
| Qwen32026.01 | 78 | 57 | 58 | 81 | 68 | |
| SkyworkModel Size=70B2026.01 | 78 | 53 | 42 | 80 | 63 | |
| Qwen2.52026.01 | 76 | 49 | 71 | 73 | 67 | |
| Granite2026.01 | 73 | 46 | 48 | 74 | 60 | |
| Prometheus2026.01 | 71 | 45 | 33 | 73 | 56 | |
| Selene2026.01 | 71 | 50 | 57 | 75 | 63 | |
| DS-Qwen2026.01 | 70 | 44 | 41 | 72 | 57 | |
| Llama2026.01 | 70 | 47 | 56 | 58 | 58 | |
| DS-GRM2026.01 | 68 | 45 | 35 | 66 | 54 |