Video Grounding on Charades-STA
75.3R@1 IoU=0.5UniTime
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| UniTimeModel Scale=7B, Task-specific design for TVG=true2026.01 | 75.3 | 56.9 | — | — | — | — | — | |
| VideoLoomModel Scale=8B2026.01 | 70 | 48.3 | — | — | — | — | — | |
| DeCaf-Grounder2025.05 | 68.79 | 47.55 | 91.53 | 72.96 | — | 70.21 | — | |
| EaTRTraining Setting=Fine-tuning2026.01 | 68.4 | 44.9 | — | — | — | — | — | |
| TimeSuiteModel Scale=7B2026.01 | 67.1 | 43 | — | — | — | — | — | |
| TimeSuiteTraining Setting=Fine-tuning2026.01 | 67.1 | 43 | — | — | 79.4 | — | — | |
| MSTG2026.01 | 66.64 | 47.03 | 93.66 | 72.53 | — | — | — | |
| SSRNFeatures=I3D2024.04 | 65.59 | 42.65 | 94.76 | 65.48 | — | — | — | |
| SSRN2026.01 | 65.59 | 42.65 | 94.76 | 65.48 | — | — | — | |
| SSRNTraining Setting=Fine-tuning2026.01 | 65.5 | 42.6 | — | — | — | — | — | |
| SnAGInference Scheme=query-centric, Features=I3D2024.04 | 65.13 | 45.91 | 92.8 | 71.75 | — | — | — | |
| SnAGInference Scheme=video-centric, Features=I3D2024.04 | 64.62 | 46.26 | 92.55 | 71.94 | — | — | — | |
| SnAG2025.05 | 64.62 | 46.26 | 92.55 | 71.94 | — | 68.84 | — | |
| SnAG2026.01 | 64.62 | 46.26 | 92.55 | 71.94 | — | — | — | |
| SnAGTraining Setting=Fine-tuning2026.01 | 64.6 | 46.2 | — | — | — | — | — | |
| SMINFeatures=I3D2024.04 | 64.06 | 40.75 | 89.49 | 68.09 | — | — | — | |
| SMIN2025.05 | 64.06 | 40.75 | 89.49 | 68.09 | — | 65.6 | — | |
| MGSL-NetFeatures=I3D2024.04 | 63.98 | 41.03 | 93.21 | 63.85 | — | — | — | |
| MGSL-Net2026.01 | 63.98 | 41.03 | 93.21 | 63.85 | — | — | — | |
| APGNFeatures=I3D2024.04 | 62.58 | 38.86 | 91.24 | 62.11 | — | — | — | |
| TRACEModel Scale=7B2026.01 | 61.7 | 41.4 | — | — | — | — | — | |
| ARROWGEV-7BTraining Setting=Zero-shot2026.01 | 61.6 | 37.2 | — | — | 78 | 54.1 | — | |
| IA-NetFeatures=I3D2024.04 | 61.29 | 37.91 | 89.78 | 62.04 | — | — | — | |
| MESM2026.01 | 61.24 | 38.04 | — | — | — | — | — | |
| CBLNFeatures=I3D2024.04 | 61.13 | 38.22 | 90.33 | 61.69 | — | — | — | |
| RaNetFeatures=I3D2024.04 | 60.4 | 39.65 | 89.57 | 64.54 | — | — | — | |
| CPNetFeatures=I3D2024.04 | 60.27 | 38.74 | — | — | — | — | — | |
| UniVTGTraining Setting=Fine-tuning2026.01 | 60.2 | 38.6 | — | — | 72.6 | — | — | |
| R2-TuningSize=-, FT=true2025.03 | 59.8 | 37 | — | — | 70.9 | 50.9 | — | |
| DeNetFeatures=I3D2024.04 | 59.7 | 38.52 | 91.24 | 66.83 | — | — | — | |
| UVCOM2026.01 | 59.25 | 36.64 | — | — | — | — | — | |
| VideoMindSize=7B, FT=false2025.03 | 59.1 | 31.2 | — | — | 73.5 | 50.2 | — | |
| Time-R1Training Setting=Zero-shot, Official weights reproduction=true2026.01 | 59 | 32.4 | — | — | 77.6 | 52.4 | — | |
| HawkEyeModel Scale=7B, Task-specific design for TVG=true2026.01 | 58.3 | 28.8 | — | — | — | — | — | |
| HawkEyeTraining Setting=Fine-tuning2026.01 | 58.3 | 28.8 | — | — | 72.5 | — | — | |
| UniVTGSize=-, FT=true2025.03 | 58.1 | 35.6 | — | — | 70.8 | 50.1 | — | |
| VTG-LLMModel Scale=7B2026.01 | 57.2 | 33.4 | — | — | — | — | — | |
| CBPFeatures=I3D2024.04 | 56.64 | 36.21 | 89.14 | 61.13 | — | — | — | |
| 2D-TANFeatures=I3D2024.04 | 54.92 | 34.26 | 76.5 | 60.02 | — | — | — | |
| LGI*+BSPfeature=RGB-only Kinetics pre-trained TSM + BSP, fine-tuning=false2020.11 | 53.63 | 29.27 | — | — | 68.76 | 50.55 | — | |
| VideoChat-FlashTraining Setting=Zero-shot2026.01 | 53.1 | 27.6 | — | — | 74.5 | — | — | |
| DRN2020.11 | 53.09 | 31.75 | — | — | — | — | — | |
| DRNFeatures=I3D2024.04 | 53.09 | 31.75 | 89.06 | 60.05 | — | — | — | |
| Moment-DETRSize=-, FT=true2025.03 | 52.1 | 30.6 | — | — | 65.8 | 45.5 | — | |
| SnAGInference Scheme=video-centric, Features=C3D2024.04 | 51.72 | 33.52 | 92.55 | 64.11 | — | — | — | |
| 2D-TAN*+BSPfeature=RGB-only Kinetics pre-trained TSM + BSP, fine-tuning=false2020.11 | 51.64 | 29.57 | — | — | — | — | — | |
| SnAGInference Scheme=query-centric, Features=C3D2024.04 | 51.1 | 31.77 | 90.22 | 62.15 | — | — | — | |
| VideoMindSize=2B, FT=false2025.03 | 51.1 | 26 | — | — | 67.6 | 45.2 | — | |
| CPNFeatures=I3D2024.04 | 51.07 | 31.54 | — | — | — | — | — | |
| BPNetFeatures=I3D2024.04 | 50.75 | 31.64 | — | — | — | — | — | |
| SSRNFeatures=C3D2024.04 | 50.39 | 31.42 | 90.68 | 59.94 | — | — | — | |
| D2VLM-3.8BModel Size=3.8B, Conference=ICCV'252025.12 | 50.3 | 26 | — | — | — | — | — | |
| TimeSuiteTraining Setting=Zero-shot2026.01 | 48.7 | 24 | — | — | 69.9 | — | — | |
| VideoChat-T-7BModel Size=7B, Conference=ICLR'252025.12 | 48.7 | 24 | — | — | — | — | — | |
| 2D-TAN*feature=RGB-only Kinetics pre-trained TSM, fine-tuning=false2020.11 | 48.36 | 27.88 | — | — | — | — | — | |
| APGNFeatures=C3D2024.04 | 48.2 | 29.37 | 89.05 | 58.49 | — | — | — | |
| CBLNFeatures=C3D2024.04 | 47.94 | 28.22 | 88.2 | 57.47 | — | — | — | |
| TimeChatModel Scale=7B2026.01 | 46.7 | 23.7 | — | — | — | — | — | |
| MAN2020.11 | 46.53 | 22.72 | — | — | — | — | — | |
| E.T. ChatSize=4B, FT=false2025.03 | 45.9 | 20 | — | — | 65.7 | 42.3 | — | |
| E.T.Chat-3.8BModel Size=3.8B, Conference=NeurIPS'242025.12 | 45.9 | 20 | — | — | — | — | — | |
| 2D-TANTraining Setting=Fine-tuning2026.01 | 45.8 | 27.9 | — | — | 57.3 | — | — | |
| LGI*feature=RGB-only Kinetics pre-trained TSM, fine-tuning=false2020.11 | 45.65 | 23.87 | — | — | 60.67 | 43.4 | — | |
| 2D-TAN2020.11 | 42.8 | 23.25 | — | — | — | — | — | |
| 2D-TANFeatures=C3D2024.04 | 41.1 | 23.25 | 81.53 | 48.55 | — | — | — | |
| CPNetFeatures=C3D2024.04 | 40.32 | 22.47 | — | — | — | — | — | |
| TRACETraining Setting=Zero-shot2026.01 | 40.3 | 19.4 | — | — | — | — | — | |
| TRACE-7BModel Size=7B, Conference=ICLR'252025.12 | 40.3 | 19.4 | — | — | — | — | — | |
| VideoChat-TPOSize=7B, FT=false2025.03 | 40.2 | 18.4 | — | — | 58.3 | 38.1 | — | |
| Qwen-2.5-VL-7BTraining Setting=Zero-shot, Official weights reproduction=true2026.01 | 38.7 | 16.5 | — | — | 59.6 | 38.3 | — | |
| BPNetFeatures=C3D2024.04 | 38.25 | 20.51 | — | — | — | — | — | |
| DRNFeatures=C3D2024.04 | 36.9 | 20.28 | 66.84 | 42.92 | — | — | — | |
| CBPFeatures=C3D2024.04 | 36.8 | 18.87 | 70.94 | 50.19 | — | — | — | |
| TVG-R1Training Setting=Zero-shot, Official weights reproduction=true2026.01 | 36.1 | 13.8 | — | — | 60.7 | 39.3 | — | |
| MLVI2020.11 | 35.6 | 15.8 | — | — | 54.7 | — | — | |
| QSPNFeatures=C3D2024.04 | 35.6 | 15.8 | 79.4 | 45.4 | — | — | — | |
| VTimeLLMSize=13B, FT=false2025.03 | 34.3 | 14.7 | — | — | 55.3 | 34.6 | — | |
| VTimeLLM-13BModel Size=13B, Conference=CVPR'242025.12 | 34.3 | 14.7 | — | — | — | — | — | |
| VTG-LLMTraining Setting=Zero-shot2026.01 | 33.8 | 15.7 | — | — | 52 | — | — | |
| VTG-LLM-7BModel Size=7B, Conference=AAAI'252025.12 | 33.8 | 15.7 | — | — | — | — | — | |
| ChatVTGSize=7B, FT=false2025.03 | 33 | 15.9 | — | — | 52.7 | 34.9 | — | |
| ChatVTGTraining Setting=Zero-shot2026.01 | 33 | 15.9 | — | — | 52.7 | — | — | |
| TimeChatSize=7B, FT=false2025.03 | 32.2 | 13.4 | — | — | 51.5 | — | — | |
| TimeChatTraining Setting=Zero-shot2026.01 | 32.2 | 15.7 | — | — | 46.7 | 32.2 | — | |
| TimeChat-7BModel Size=7B, Conference=CVPR'242025.12 | 32.2 | 13.4 | — | — | — | — | — | |
| HawkEyeSize=7B, FT=false2025.03 | 31.4 | 14.5 | — | — | 50.6 | 33.7 | — | |
| HawkEyeTraining Setting=Zero-shot2026.01 | 31.4 | 14.5 | — | — | 50.6 | — | — | |
| HawkEye-7BModel Size=7B, Conference=arXiv'242025.12 | 31.4 | 14.5 | — | — | — | — | — | |
| TimeChat + Seq2TimeRelative Position Token (RPT)=true2024.11 | 31.2 | 13.7 | — | — | — | — | — | |
| TimeChat + Seq2TimeRelative Position Token (RPT)=false2024.11 | 29.3 | 12.8 | — | — | — | — | — | |
| VTimeLLM [12]2024.11 | 27.5 | 11.4 | — | — | — | — | — | |
| VTimeLLMTraining Setting=Zero-shot2026.01 | 27.5 | 11.4 | — | — | 51 | 31.2 | — | |
| VTimeLLM-7BModel Size=7B, Conference=CVPR'242025.12 | 27.5 | 11.4 | — | — | — | — | — | |
| SAP2020.11 | 27.42 | 13.36 | — | — | — | — | — | |
| TimeChat [30]*2024.11 | 27.2 | 11.7 | — | — | — | — | — | |
| Monmentor [29]2024.11 | 26.6 | 11.6 | — | — | — | — | — | |
| MomentorSize=7B, FT=false2025.03 | 26.6 | 11.6 | — | — | 42.6 | 28.5 | — | |
| Momentor-7BModel Size=7B, Conference=ICML'242025.12 | 26.6 | 11.6 | — | — | — | — | — | |
| SMRL2020.11 | 24.36 | 9.01 | — | — | — | — | — | |
| CTRLFeatures=C3D2024.04 | 23.63 | 8.89 | 58.92 | 29.52 | — | — | — |