Text-to-Video Retrieval on ACTNET-10
19.66R@1Upper Bound
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Upper Bound2026.01 | 19.66 | 44.57 | 59.18 | 7 | 37.16 | 0.27 | |
| StructAlign2026.01 | 18.26 | 40.31 | 54.68 | 9 | 48.82 | -0.68 | |
| StableFusion2026.01 | 18.21 | 40.45 | 53.94 | 9 | 56.14 | -0.01 | |
| VR-LwF2026.01 | 18.08 | 41.44 | 54.98 | 8.5 | 53.28 | 0.68 | |
| X-Pool2026.01 | 17.99 | 39.81 | 52.38 | 9.67 | 60.49 | 0.37 | |
| CLIP4Clip2026.01 | 17.85 | 41.05 | 54.88 | 8.67 | 54.97 | 0.75 | |
| ZSCL2026.01 | 17.67 | 41.05 | 54.05 | 9 | 55.74 | 0.35 | |
| LwF2026.01 | 17.56 | 40.18 | 53.67 | 9 | 55.33 | 0.63 | |
| CLIP-ViP2026.01 | 17.01 | 38.73 | 52.01 | 9.67 | 59.66 | 0.56 | |
| TVR+CL2026.01 | 16.88 | 38.87 | 51.82 | 9.67 | 61.16 | 0.44 | |
| MoE-Adapter2026.01 | 16.63 | 37.29 | 50.36 | 10.33 | 70.49 | -0.15 | |
| ZS CLIPMode=Zero-Shot2026.01 | 14.89 | 34.97 | 47.78 | 12 | 84.02 | 0 |