Video-to-Text Retrieval on ActivityNet severity degree 5 (test)
18.26Performance (Gaussian Noise)HAT-VTR
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HAT-VTRBase Model=CLIP4Clip, Adaptation=HAT-VTR2026.02 | 18.26 | 18.81 | 32.54 | 23.9 | 27.58 | 36.47 | 26.87 | 12.75 | 20.42 | 8.42 | 17.94 | 23.45 | 22.28 | |
| HAT-VTRBase Model=Xpool, Adaptation=HAT-VTR2026.02 | 14.6 | 14.64 | 28.41 | 20.64 | 23.04 | 31.99 | 22.35 | 10.66 | 19.02 | 7.61 | 27.8 | 23.39 | 20.35 | |
| TCRBase Model=Xpool, Adaptation=TCR2026.02 | 9.21 | 9.92 | 23.65 | 16.19 | 14.54 | 27.21 | 15.76 | 6.1 | 10.88 | 4.05 | 18.79 | 17.65 | 14.5 | |
| SARBase Model=CLIP4Clip, Adaptation=SAR2026.02 | 8.87 | 1.26 | 18.97 | 12.53 | 11.06 | 33.94 | 18.45 | 1.36 | 6.24 | 1.32 | 10.01 | 24.3 | 12.36 | |
| SARBase Model=Xpool, Adaptation=SAR2026.02 | 8.48 | 1.53 | 19.52 | 10.45 | 10.84 | 29.43 | 14.85 | 1.99 | 14.64 | 2.79 | 22.84 | 23.35 | 13.39 | |
| TentBase Model=Xpool, Adaptation=Tent2026.02 | 7.77 | 1.14 | 16.37 | 6.28 | 9.27 | 29.45 | 14.97 | 1.08 | 6.45 | 2.07 | 22.86 | 23.23 | 11.75 | |
| EATABase Model=Xpool, Adaptation=EATA2026.02 | 7.75 | 0.39 | 10.09 | 3.68 | 10.72 | 28.66 | 16.7 | 0.43 | 6.1 | 1.5 | 22.55 | 22.98 | 10.96 | |
| XpoolBase Model=Xpool, Adaptation=Source2026.02 | 6.06 | 5.08 | 18.81 | 10.13 | 8.24 | 29.27 | 13 | 3.15 | 16.76 | 4.21 | 22.6 | 23.35 | 13.39 | |
| TCRBase Model=CLIP4Clip, Adaptation=TCR2026.02 | 5.45 | 11.61 | 27.17 | 17.51 | 18.67 | 18.22 | 11.73 | 7.46 | 7.44 | 4.27 | 7.2 | 17.61 | 12.86 | |
| CLIP4ClipBase Model=CLIP4Clip, Adaptation=Source2026.02 | 4.9 | 5.13 | 19.58 | 10.19 | 7.36 | 33.54 | 14.81 | 2.95 | 13.02 | 3.66 | 9.8 | 24.02 | 12.41 | |
| TentBase Model=CLIP4Clip, Adaptation=Tent2026.02 | 4.6 | 0.92 | 10.78 | 4.98 | 9.48 | 33.58 | 18.73 | 0.63 | 4.17 | 0.79 | 9.88 | 24.16 | 10.23 | |
| EATABase Model=CLIP4Clip, Adaptation=EATA2026.02 | 3.88 | 0.28 | 13.28 | 13.89 | 8.87 | 31.95 | 21.58 | 1.1 | 3.95 | 1.48 | 8.2 | 24.32 | 11.07 | |
| READBase Model=Xpool, Adaptation=READ2026.02 | 3.62 | 4.03 | 18.61 | 7.67 | 6.22 | 28.86 | 8.07 | 2.2 | 16.88 | 4.37 | 22.01 | 23.1 | 12.14 | |
| READBase Model=CLIP4Clip, Adaptation=READ2026.02 | 2.7 | 4.07 | 15.62 | 8.72 | 4.88 | 31.71 | 7.42 | 1.65 | 14.83 | 3.88 | 9.86 | 23.77 | 10.76 |