Hypothesis composition on TOMATO-STAR (test)
5.16TotalMS-HC-7B
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MS-HC-7BTraining Data=w/ 1× bounded composition data2026.03 | 5.16 | 2.24 | 1.6 | 1.31 | 203.84 | |
| MS-HC-7BTraining Data=w/ 2× bounded composition data2026.03 | 5.15 | 2.25 | 1.58 | 1.32 | 205.17 | |
| MS-HC-7B2026.03 | 5.08 | 2.21 | 1.58 | 1.29 | 204.12 | |
| R1-DISTILLED-QWEN-7B2026.03 | 4.34 | 1.96 | 1.4 | 0.97 | 231.02 |