Vision-Language-Action Navigation on IndoorUAV-VLA (test)
37.72SRWorldVLN
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| WorldVLN2026.05 | 37.72 | — | — | — | — | — | — | — | — | 14.52 | |
| π02026.05 | 21.64 | — | — | — | — | — | — | — | — | 7.64 | |
| NaVid2026.05 | 18.21 | — | — | — | — | — | — | — | — | 3.21 | |
| π0-FAST2026.05 | 5.26 | — | — | — | — | — | — | — | — | 2.93 | |
| GPT-4o2026.05 | 4 | — | — | — | — | — | — | — | — | 6.57 | |
| Seq2Seq2026.05 | 1.2 | — | — | — | — | — | — | — | — | 2.74 | |
| OpenVLA2026.05 | 1.19 | — | — | — | — | — | — | — | — | 1.12 | |
| CMA2026.05 | 0.75 | — | — | — | — | — | — | — | — | 1.94 | |
| CMAFine-tuned=true2025.12 | — | 0.99 | 1.88 | 1.28 | 1.75 | 0.75 | 1.94 | 1.03 | 2.01 | — | |
| GPT-4oFine-tuned=false2025.12 | — | 11.69 | 9.2 | 30.3 | 12.3 | 4 | 6.57 | 1.96 | 4.84 | — | |
| NaVidFine-tuned=true2025.12 | — | 15.82 | 5.28 | 25.31 | 13.1 | 18.21 | 3.21 | 2.31 | 1.72 | — | |
| OpenVLAFine-tuned=true2025.12 | — | 7.81 | 2.42 | 22.52 | 2.89 | 1.19 | 1.12 | 0 | 0.12 | — | |
| Seq2SeqFine-tuned=true2025.12 | — | 1.33 | 2.74 | 1.6 | 2.63 | 1.2 | 2.74 | 1.03 | 3.03 | — | |
| π0Fine-tuned=true2025.12 | — | 27.16 | 9.44 | 46.58 | 14.52 | 21.64 | 7.64 | 7.55 | 4.27 | — | |
| π0-FASTFine-tuned=true2025.12 | — | 8.62 | 4.71 | 18.09 | 8.83 | 5.26 | 2.93 | 1.14 | 2.68 | — |