Video-to-Text Retrieval on MSRVTT-1kA severity degree 3
41Gaussian ScoreHAT-VTR
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HAT-VTRBase Architecture=Xpool, Test-Time Adaptation (TTA) Method=HAT-VTR2026.02 | 41 | 35.8 | 44.2 | 36.2 | 44.7 | 46 | 40.9 | 30.6 | 39.5 | 30.3 | 39.1 | 38.1 | 38.9 | |
| EATABase Architecture=Xpool, Test-Time Adaptation (TTA) Method=EATA2026.02 | 37.7 | 27.2 | 42 | 30.2 | 41.7 | 43.4 | 37.2 | 26.5 | 35.8 | 26.1 | 36 | 36.7 | 35 | |
| HAT-VTRBase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=HAT-VTR2026.02 | 36 | 34 | 40.3 | 33.5 | 40.3 | 42.6 | 37.4 | 28.8 | 36.8 | 27 | 35 | 36.9 | 35.7 | |
| TCRBase Architecture=Xpool, Test-Time Adaptation (TTA) Method=TCR2026.02 | 35.2 | 29.3 | 41 | 30.5 | 42.6 | 43.9 | 36.1 | 23.7 | 36.2 | 25.7 | 35.5 | 36.5 | 34.7 | |
| EATABase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=EATA2026.02 | 34.5 | 22.7 | 37.4 | 23.9 | 36.9 | 41.9 | 34.8 | 23.6 | 31.7 | 21.5 | 31.4 | 36.1 | 31.4 | |
| TentBase Architecture=Xpool, Test-Time Adaptation (TTA) Method=Tent2026.02 | 34.3 | 15.3 | 39.5 | 26 | 41 | 44.1 | 36 | 23.7 | 35.6 | 21.4 | 34.9 | 36.9 | 32.4 | |
| SARBase Architecture=Xpool, Test-Time Adaptation (TTA) Method=SAR2026.02 | 34.3 | 18.1 | 39.6 | 25.8 | 40.9 | 43.8 | 35.7 | 24 | 35.5 | 22.4 | 35 | 37.2 | 32.7 | |
| TCRBase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=TCR2026.02 | 32.7 | 26.6 | 37.1 | 28.2 | 37.7 | 39.8 | 36 | 22 | 31.6 | 21.6 | 29.9 | 34.9 | 31.5 | |
| XpoolBase Architecture=Xpool, Test-Time Adaptation (TTA) Method=None2026.02 | 32 | 17.7 | 38.6 | 24.9 | 40.5 | 43.6 | 34.1 | 21.7 | 35.6 | 19.8 | 35.3 | 37 | 31.7 | |
| SARBase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=SAR2026.02 | 29.8 | 14.3 | 36.6 | 22 | 36.5 | 41 | 35.5 | 20.6 | 30.3 | 19 | 30.6 | 36.7 | 29.4 | |
| READBase Architecture=Xpool, Test-Time Adaptation (TTA) Method=READ2026.02 | 29.8 | 18.8 | 37.6 | 23.8 | 40.6 | 43.8 | 32.8 | 18.9 | 35 | 19.4 | 35.5 | 37 | 31.1 | |
| TentBase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=Tent2026.02 | 29.5 | 10.8 | 36.4 | 21.1 | 36.5 | 41 | 35.4 | 20.1 | 30.3 | 17.8 | 30.6 | 36.7 | 28.9 | |
| CLIP4ClipBase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=None2026.02 | 27.1 | 14.8 | 35.2 | 20.4 | 36.5 | 40.7 | 34 | 18.9 | 29.2 | 18 | 30.1 | 36.4 | 28.4 | |
| READBase Architecture=CLIP4Clip, Test-Time Adaptation (TTA) Method=READ2026.02 | 25.8 | 16.8 | 34.3 | 20.3 | 35.5 | 40.5 | 33.2 | 17.3 | 29.5 | 17.7 | 30.4 | 36 | 28.1 |