Temporal Sentence Grounding on Charades STA
69.93Rank-1 Recall @ IoU=0.5MKTN
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| MKTNType=Multi-Pair (MP)2024.12 | 69.93 | 46.27 | 97.16 | 69.13 | — | |
| SCADABackbone=I3D2026.04 | 64.59 | 43.65 | 92.69 | 72.31 | 55.94 | |
| MRTNetType=Fully-supervised (FS)2024.12 | 62.5 | 43.63 | — | — | — | |
| DPQFBackbone=I3D2026.04 | 62.2 | 42.52 | — | — | 55.03 | |
| MGPNBackbone=I3D2026.04 | 60.82 | 41.16 | 89.77 | 64.73 | — | |
| MESMType=Fully-supervised (FS)2024.12 | 56.69 | 35.99 | — | — | — | |
| MS-2D-TANBackbone=I3D2026.04 | 56.64 | 36.21 | 89.14 | 61.13 | — | |
| MMDistType=Weakly-supervised (WS)2024.12 | 54.72 | 26 | — | — | — | |
| DRNType=Fully-supervised (FS)2024.12 | 53.09 | 31.75 | 89.06 | 60.05 | — | |
| VDIType=Fully-supervised (FS)2024.12 | 52.32 | 31.37 | 87.03 | 62.3 | — | |
| SCADABackbone=C3D2026.04 | 50.44 | 31.19 | 89.27 | 61.32 | 45.49 | |
| D2VLMBackbone=LLM-based2026.04 | 50.3 | 26 | — | — | — | |
| CPLType=Weakly-supervised (WS)2024.12 | 49.24 | 22.39 | 84.71 | 52.37 | — | |
| APGNBackbone=C3D2026.04 | 48.2 | 29.37 | 89.05 | 58.49 | — | |
| VSLNetBackbone=I3D2026.04 | 47.31 | 30.19 | — | — | 45.15 | |
| CPNBackbone=C3D2026.04 | 46.08 | 26.05 | — | — | 43.9 | |
| MS-2D-TANBackbone=C3D2026.04 | 41.1 | 23.25 | 81.53 | 48.55 | — | |
| VideoExpertBackbone=LLM-based2026.04 | 40.3 | 20.9 | — | — | 41.1 | |
| TRACEBackbone=LLM-based2026.04 | 40.3 | 19.4 | — | — | — | |
| LCNetType=Weakly-supervised (WS)2024.12 | 39.19 | 18.17 | 80.56 | 45.24 | — | |
| VCAType=Weakly-supervised (WS)2024.12 | 38.13 | 19.57 | 78.75 | 37.75 | — | |
| HawkEyeBackbone=LLM-based2026.04 | 31.4 | 14.5 | — | — | 33.7 | |
| ICVCType=Weakly-supervised (WS)2024.12 | 31.02 | 16.53 | 77.53 | 41.91 | — | |
| MomentorBackbone=LLM-based2026.04 | 26.6 | 11.6 | — | — | 28.5 |