Verb Prediction on VidSitu (val)
56.15Top-1 Verb AccuracyHostSG
Evaluation results
| Method | Links | ||
|---|---|---|---|
| HostSGBackbone=M-TDE2026.04 | 56.15 | 86.33 | |
| OME+OIEBackbone=I3D2026.04 | 53.36 | 83.94 | |
| CineMEC (old backbone)Backbone=FR + SF2026.04 | 49.86 | 78.36 | |
| CineMECBackbone=YS + SF2026.04 | 49.32 | 79.83 | |
| TypesDev-ucofiaBackbone=BLIP22026.04 | 47.23 | — | |
| VideoWhispererBackbone=YS + SF2026.04 | 46.02 | 76.49 | |
| VideoWhispererBackbone=FR + SF2026.04 | 45.06 | 75.59 | |
| VidSitu-SlowFastBackbone=SF2026.04 | 32.64 | 69.2 |