Motion-to-Text Retrieval on HumanML3D
68.86R@1CoAMD
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| CoAMD2026.04 | 68.86 | 82.77 | 88.09 | 92.36 | 97.21 | — | |
| LaMP2026.04 | 68.02 | 82.1 | 87.5 | 92.2 | 96.9 | — | |
| TMR2026.04 | 67.97 | 81.2 | 86.35 | 91.7 | 95.27 | — | |
| T2M2026.04 | 52 | 71.21 | 81.11 | 89.87 | 96.78 | — | |
| TEMOS2026.04 | 39.96 | 53.49 | 61.79 | 72.4 | 85.89 | — | |
| SGARVenue=NeurIPS’25, LLM Augmentation=true2026.03 | 13.82 | — | 23.38 | 30.09 | 41.83 | 16 | |
| Ours-LVenue=Proposed, Backbone=ViT-Large, RoBERTa-Large, LLM Augmentation=false2026.03 | 13.76 | 17.42 | 25.47 | 33.22 | 44.74 | 13 | |
| OursVenue=Proposed, LLM Augmentation=false2026.03 | 13.1 | 16.21 | 22.82 | 30.45 | 41.45 | 16 | |
| Lyu et al.Venue=ICLR’25, LLM Augmentation=false2026.03 | 12.39 | 15.55 | 22.17 | 29.25 | 40.34 | 17 | |
| MoPatch-LBackbone=Large, LLM Augmentation=false2026.03 | 12.29 | 15.65 | 22.51 | 29.36 | 40.1 | 17 | |
| PL-TMRVenue=AAAI’25, LLM Augmentation=true2026.03 | 12.25 | 14.95 | 21.45 | 28.34 | 39.11 | 19 | |
| MoPatchVenue=CVPR’24, LLM Augmentation=false2026.03 | 11.7 | 14.67 | 21.94 | 29.08 | 38.73 | 18 | |
| CARVenue=ECCV’24, LLM Augmentation=false2026.03 | 11.18 | 13.85 | 20.19 | 26.65 | 36.37 | 20.5 | |
| SGAR-LBackbone=Large, LLM Augmentation=true2026.03 | 11.07 | 14.1 | 20.84 | 27.38 | 39.37 | 17 | |
| TMRVenue=ICCV’23, LLM Augmentation=false2026.03 | 9.44 | — | 16.9 | 22.92 | 32.21 | 26 | |
| KinMoVenue=ICCV’25, LLM Augmentation=true2026.03 | 9.01 | 15.92 | 21.42 | 29.5 | 41.43 | 16 |