Attributing on CUVA instructions 1.0 (test)
0.51SB ScoreUPRM
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| UPRMfull_model=true2026.02 | 0.51 | 46 | 42.15 | 7.51 | 4.21 | |
| Holmes2026.02 | 0.49 | 38 | 25.43 | 6.75 | 3.63 | |
| UPRMVBE=removed2026.02 | 0.49 | 45 | 41.81 | 7.44 | 4.13 | |
| Omni-SILA2026.02 | 0.47 | 42 | 35.13 | 7.35 | 4.01 | |
| Hawkeye2026.02 | 0.47 | 43 | 35.64 | 7.32 | 4.06 | |
| UPRMPTR=removed2026.02 | 0.47 | 41 | 37.56 | 7.15 | 3.77 | |
| UPRMHPE=removed2026.02 | 0.46 | 40 | 30.18 | 7.01 | 4.01 | |
| UPRMORE=removed2026.02 | 0.46 | 44 | 40.51 | 7.37 | 3.98 | |
| UPRMPre-tuning=removed2026.02 | 0.45 | 40 | 30.21 | 6.74 | 3.61 | |
| PandaGPT2026.02 | 0.44 | 40 | 35.51 | 7.06 | 3.61 | |
| mPLUG-Owl2026.02 | 0.42 | 33 | 29.46 | 6.04 | 3.06 | |
| Chat-UniVi2026.02 | 0.41 | 34 | 31.33 | 6.23 | 3.18 | |
| UPRMUPE=removed2026.02 | 0.41 | 38 | 32.14 | 6.94 | 3.65 | |
| Video-LLaVA2026.02 | 0.4 | 37 | 28.51 | 6.69 | 3.49 | |
| Video Chat2026.02 | 0.39 | 36 | 27.94 | 6.54 | 3.23 | |
| Valley2026.02 | 0.3 | 31 | 22.1 | 5.93 | 2.97 |