Dense Video Captioning on ViTT (test)
25SODA_cE2ESG
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| E2ESGBackbone=C3D2023.02 | 25 | — | 8.1 | — | |
| Sali4VidPretrained=true2026.03 | 15.08 | 53.87 | 10.05 | 0.91 | |
| HiCM2Pretrained=true2026.03 | 15.07 | 51.29 | 9.66 | 0.86 | |
| STaRCPretrained=true2026.03 | 15.01 | 56.04 | 10.49 | 0.89 | |
| Vid2SeqPretrained=true, reproduced_from_official=true2026.03 | 14.99 | 48.84 | 9.51 | 0.71 | |
| Vid2SeqBackbone=CLIP2023.02 | 13.5 | 43.5 | 8.5 | — | |
| Streaming V2SPretrained=true2026.03 | 10 | 25.2 | 5.8 | — |