Embodied Visual Tracking on EVT-Bench
92.1ST Success Rate (SR)CoMaTrack
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| CoMaTrackVLM Model Size (Parameters)=3B2026.03 | 92.1 | 90.3 | 0.9 | 74.2 | 80.5 | 2.1 | 57.5 | 73.4 | 12 | |
| TrackVLA++VLM Model Size (Parameters)=7B, Fine-tuning=true2026.03 | 90.9 | 82.7 | 1.5 | 74 | 73.7 | 3.5 | 55.9 | 63.8 | 15.1 | |
| VLingNavVLM Model Size (Parameters)=7B, Fine-tuning=true2026.03 | 88.4 | 81.2 | 2.1 | 67.7 | 73.5 | 5.5 | — | — | — | |
| NavFoMVLM Model Size (Parameters)=7B, Fine-tuning=true2026.03 | 88.4 | 80.7 | — | 62 | 67.9 | — | — | — | — | |
| TrackVLAVLM Model Size (Parameters)=7B, Fine-tuning=true2026.03 | 85.1 | 78.6 | 1.7 | 57.6 | 63.2 | 5.8 | 50.2 | 63.7 | 17.1 | |
| Uni-NaVidVLM Model Size (Parameters)=7B, Fine-tuning=true2026.03 | 53.3 | 67.2 | 12.6 | 31.9 | 50.1 | 21.3 | 15.8 | 41.5 | 26.5 | |
| IBVSDetector=GroundingDINO [36]2026.03 | 42.9 | 56.2 | 3.8 | 10.6 | 28.4 | 6.1 | 15.2 | 39.5 | 4.9 | |
| EVTVisual Foundation Model=SoM [57] and GPT-4o [22]2026.03 | 32.5 | 49.9 | 40.5 | 15.7 | 35.7 | 53.3 | 18.3 | 21 | 44.9 | |
| EVT2026.03 | 24.4 | 39.1 | 42.5 | 3.2 | 11.2 | 47.9 | 17.4 | 21.1 | 45.6 | |
| PoliFormerDetector=GroundingDINO [36]2026.03 | 4.7 | 15.5 | 40.1 | 2.6 | 13.2 | 44.5 | 3 | 15.4 | 41.5 |