Video Sentence Grounding on ActivityNet Captions
54.35Recall@1 (IoU=0.5)BM-DETR + Ours
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| BM-DETR + OursSupervision=fully-supervised, Text Augmentation=false2026.05 | 54.35 | 66.85 | 92.13 | 85.13 | |
| G2L + OursSupervision=fully-supervised, Text Augmentation=false2026.05 | 54.26 | 66.34 | 91.77 | 84.29 | |
| G2LSupervision=fully-supervised, Text Augmentation=false2026.05 | 51.68 | — | — | 81.32 | |
| BM-DETRSupervision=fully-supervised, Text Augmentation=false2026.05 | 49.62 | — | — | — | |
| MMN + OursSupervision=fully-supervised, Text Augmentation=false2026.05 | 49.31 | 66.05 | 89.75 | 81.27 | |
| MMNSupervision=fully-supervised, Text Augmentation=false2026.05 | 48.59 | 65.05 | 87.25 | 79.5 | |
| BM-DETR + OursSupervision=fully-supervised, Text Augmentation=true2026.05 | 47.93 | 61.98 | 85.27 | 81.38 | |
| G2L + OursSupervision=fully-supervised, Text Augmentation=true2026.05 | 46.86 | 60.9 | 84.39 | 80.62 | |
| 2D-TAN + OursSupervision=fully-supervised, Text Augmentation=false2026.05 | 45.29 | 60.46 | 87.94 | 77.43 | |
| 2D-TANSupervision=fully-supervised, Text Augmentation=false2026.05 | 44.51 | 59.45 | 85.53 | 77.13 | |
| MCMT + OursSupervision=weakly-supervised, Text Augmentation=false2026.05 | 35.12 | 56.75 | 82.43 | 68.72 | |
| CNM + OursSupervision=weakly-supervised, Text Augmentation=false2026.05 | 34.08 | 56.11 | 81.09 | 67.34 | |
| MCMTSupervision=weakly-supervised, Text Augmentation=false2026.05 | 33.97 | 58.82 | — | — | |
| BM-DETRSupervision=fully-supervised, Text Augmentation=true2026.05 | 33.52 | 56.24 | 75.34 | 70.26 | |
| CNMSupervision=weakly-supervised, Text Augmentation=false2026.05 | 33.33 | 55.68 | — | — | |
| VCA + OursSupervision=weakly-supervised, Text Augmentation=false2026.05 | 33.19 | 51.72 | 72.85 | 55.11 | |
| MMN + OursSupervision=fully-supervised, Text Augmentation=true2026.05 | 33.08 | 58.76 | 75.33 | 73.59 | |
| G2LSupervision=fully-supervised, Text Augmentation=true2026.05 | 33.01 | 55.75 | 75.25 | 70.89 | |
| 2D-TAN + OursSupervision=fully-supervised, Text Augmentation=true2026.05 | 32.64 | 51.86 | 72.98 | 63.75 | |
| MCMT + OursSupervision=weakly-supervised, Text Augmentation=true2026.05 | 32.42 | 40.25 | 53.4 | 52.74 | |
| CNM + OursSupervision=weakly-supervised, Text Augmentation=true2026.05 | 31.77 | 39.56 | 52.88 | 51.99 | |
| MMNSupervision=fully-supervised, Text Augmentation=true2026.05 | 31.76 | 55.3 | 74.88 | 71.52 | |
| WSTAN + OursSupervision=weakly-supervised, Text Augmentation=false2026.05 | 31.56 | 53.1 | 80.24 | 65.77 | |
| VCASupervision=weakly-supervised, Text Augmentation=false2026.05 | 31 | 50.45 | 71.79 | 53.83 | |
| WSTANSupervision=weakly-supervised, Text Augmentation=false2026.05 | 30.01 | 52.45 | 79.38 | 63.42 | |
| MCMTSupervision=weakly-supervised, Text Augmentation=true2026.05 | 29.52 | 36.21 | 50.78 | 48.21 | |
| 2D-TANSupervision=fully-supervised, Text Augmentation=true2026.05 | 29.38 | 48.32 | 71.36 | 62.3 | |
| CNMSupervision=weakly-supervised, Text Augmentation=true2026.05 | 28.95 | 35.72 | 50.06 | 48.72 | |
| VCA + OursSupervision=weakly-supervised, Text Augmentation=true2026.05 | 28.56 | 32.99 | 48.31 | 44.07 | |
| WSTAN + OursSupervision=weakly-supervised, Text Augmentation=true2026.05 | 27.99 | 35.2 | 51.84 | 48.69 | |
| WSTANSupervision=weakly-supervised, Text Augmentation=true2026.05 | 25.74 | 33.72 | 49.3 | 45.88 | |
| VCASupervision=weakly-supervised, Text Augmentation=true2026.05 | 25.37 | 31.74 | 46.98 | 42.76 |