UMUI Judgment Calibration on UNLI (full)
0.0573MSECLUE-D
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CLUE-DParameters=3B2026.04 | 0.0573 | 0.509 | 0.0592 | |
| Q3-VLParameters=32B2026.04 | 0.0753 | 0.666 | 0.0298 | |
| TMTOParameters=14B2026.04 | 0.0757 | 0.555 | 0.147 | |
| CLUE-TParameters=3B2026.04 | 0.081 | 0.955 | 0.097 | |
| Q3Parameters=32B2026.04 | 0.106 | 1.09 | 0.133 | |
| Q2.5-OParameters=7B2026.04 | 0.114 | 1.51 | 0.142 | |
| Q2.5-VLParameters=32B2026.04 | 0.126 | 3.1 | 0.155 | |
| AFParameters=7B2026.04 | 0.272 | 8.87 | 0.385 | |
| Q2-AParameters=7B2026.04 | 0.778 | 8.85 | 0.379 |