Vision-and-Language Navigation on REVERIE unseen (test)
81.5Success Rate (SR)Human
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Human2025.09 | 81.5 | 21.2 | 86.8 | 53.7 | 77.8 | — | — | |
| SIDTraining Environments=800 environments from HM3D2025.09 | 60.4 | 20.2 | 65.5 | 45.9 | 36.6 | 27.7 | — | |
| P3Nav2026.03 | 60.06 | — | — | 40.57 | 39.75 | 26.56 | — | |
| GOAT2024.04 | 57.72 | — | — | 40.53 | 38.32 | 26.7 | — | |
| GOAT2026.03 | 57.72 | — | — | 40.53 | 38.32 | 26.7 | — | |
| GOAT2025.09 | 57.7 | — | — | 40.5 | 38.3 | 26.7 | — | |
| VER2024.10 | 56.82 | 22.7 | 62.22 | 38.76 | 33.88 | 24.74 | — | |
| VER2026.03 | 56.82 | — | — | 38.76 | 33.88 | 23.19 | — | |
| VER2026.05 | 56.82 | 24.74 | 62.22 | 38.76 | 33.88 | 23.19 | — | |
| VER2025.09 | 56.8 | 24.7 | 62.2 | 38.8 | 33.9 | 23.2 | — | |
| BSG2025.09 | 56.5 | 22.9 | 62.8 | 38.7 | 33.2 | 22.3 | — | |
| BSG2024.10 | 56.45 | 22.9 | 62.83 | 38.7 | 33.15 | 22.34 | — | |
| BSG2026.03 | 56.45 | — | — | 38.7 | 33.15 | 22.34 | — | |
| BSG2026.05 | 56.45 | 22.9 | 62.83 | 38.7 | 33.15 | 22.34 | — | |
| ScaleVLNTraining Environments=800 environments from HM3D2025.09 | 56.1 | — | 62.7 | 39.5 | 32.5 | 22.8 | — | |
| 3D Value Map2026.05 | 55.9 | 20.14 | 60.12 | 38.77 | 35.68 | 25.5 | — | |
| AutoVLNTraining Environments=800 environments from HM3D2025.09 | 55.2 | — | 62.3 | 38.9 | 32.2 | 22.7 | — | |
| DUET + IDEAFeedback=No, Backbone=DUET2026.05 | 55.12 | — | 58.91 | 39.84 | — | 24.52 | 344.3 | |
| DUET + IDEABase Model=DUET, Time (ms)=344.32026.05 | 55.12 | — | 58.91 | 39.84 | — | 24.52 | — | |
| CONSOLE2024.12 | 55.1 | — | — | 37.1 | — | 22.2 | — | |
| DUET+SDB2026.04 | 54.92 | 22.35 | 58.63 | 37.74 | — | — | — | |
| SUSA2024.12 | 54.3 | — | — | 41.5 | — | 27.3 | — | |
| DUET + ATENAFeedback=Yes, Backbone=DUET2026.05 | 54.28 | — | 57.74 | 40.7 | — | 25.01 | 5.42 | |
| DSRG2024.04 | 54.04 | — | — | 37.09 | 32.49 | 22.18 | — | |
| DSRG2026.03 | 54.04 | — | — | 37.09 | 32.49 | 22.18 | — | |
| DUET + FeedTTAFeedback=Yes, Backbone=DUET2026.05 | 53.58 | — | 58.76 | 37.66 | — | 24.1 | 7.64 | |
| DUET + FSTTAFeedback=No, Backbone=DUET2026.05 | 53.4 | — | 58.44 | 36.43 | — | 22.4 | 833.6 | |
| DUET + FSTTABase Model=DUET, Time (ms)=833.62026.05 | 53.4 | — | 58.44 | 36.43 | — | 22.4 | — | |
| ENP-DUETBackbone=DUET2024.10 | 53.19 | 22.7 | 59.38 | 36.26 | 33.1 | 22.14 | — | |
| GridMM2024.04 | 53.13 | — | — | 36.6 | 34.87 | 23.45 | — | |
| GridMM2026.03 | 53.13 | — | — | 36.6 | 34.87 | 23.45 | — | |
| GridMM2026.05 | 53.13 | 19.97 | 59.55 | 36.6 | 34.87 | 23.45 | — | |
| GridMM2024.12 | 53.1 | — | — | 36.6 | — | 23.4 | — | |
| DUET + ReCAPFeedback=No, Backbone=DUET2026.05 | 53.07 | — | 57.72 | 36.52 | — | 22.47 | 359.4 | |
| DUET + ReCAPBase Model=DUET, Time (ms)=359.42026.05 | 53.07 | — | 57.72 | 36.52 | — | 22.47 | — | |
| DUET + SARBase Model=DUET, Time (ms)=267.52026.05 | 53.04 | — | 57.11 | 36.07 | — | 22.27 | — | |
| 3DGS-VLN2026.05 | 52.93 | 20.05 | 56.93 | 36.93 | 35.65 | 25.76 | — | |
| DUET + RLCFFeedback=Yes, Backbone=DUET2026.05 | 52.92 | — | 57.32 | 36.4 | — | 22.23 | 2.43 | |
| BEVBert2024.04 | 52.81 | — | — | 36.41 | 32.06 | 22.09 | — | |
| BEVBert2024.10 | 52.81 | — | 57.26 | 36.41 | 32.06 | 22.09 | — | |
| BEVBert2026.03 | 52.81 | — | — | 36.41 | 32.06 | 22.09 | — | |
| BEVBert2026.05 | 52.81 | — | 57.26 | 36.41 | 32.06 | 22.09 | — | |
| BEVBert2025.09 | 52.8 | — | 57.3 | 36.4 | 32.1 | 22.1 | — | |
| DUET + ViDABase Model=DUET, Time (ms)=7.31 × 10³2026.05 | 52.74 | — | 56.78 | 35.1 | — | 21.77 | — | |
| DUET + TentFeedback=No, Backbone=DUET2026.05 | 52.61 | — | 57.12 | 36.17 | — | 22.16 | 258.9 | |
| DUET + TentBase Model=DUET, Time (ms)=258.92026.05 | 52.61 | — | 57.12 | 36.17 | — | 22.16 | — | |
| COSMO2026.05 | 52.53 | — | 59.33 | 36.12 | — | — | — | |
| DUET2023.03 | 52.51 | 21.3 | 56.91 | 36.06 | — | — | — | |
| DUET2024.04 | 52.51 | — | — | 36.06 | 31.88 | 22.06 | — | |
| DUET2024.10 | 52.51 | 21.3 | 56.91 | 36.06 | 31.88 | 22.06 | — | |
| DUET2026.03 | 52.51 | — | — | 36.06 | 31.88 | 22.06 | — | |
| DUET2026.04 | 52.51 | 21.59 | 56.71 | 36.06 | — | — | — | |
| DUETFeedback=No2026.05 | 52.51 | — | 56.91 | 36.06 | — | 22.06 | 123.2 | |
| DUETBase Model=DUET, Time (ms)=123.22026.05 | 52.51 | — | 56.91 | 36.06 | — | 22.06 | — | |
| DUET2026.05 | 52.51 | 21.3 | 56.91 | 36.06 | 31.88 | 22.06 | — | |
| AZHP2024.12 | 52.5 | — | — | 36.1 | — | 22.5 | — | |
| DUET2024.12 | 52.5 | — | — | 36 | — | 22 | — | |
| DUET2025.09 | 52.5 | 21.3 | 56.9 | 36.1 | 31.9 | 22.1 | — | |
| KERMTraining Strategy=Pre-training and fine-tuning2023.03 | 52.43 | 17.32 | 57.58 | 39.21 | — | — | — | |
| KERM-ptTraining Strategy=Pre-training2023.03 | 52.26 | 18.38 | 57.44 | 37.46 | — | — | — | |
| KERM2024.12 | 52.2 | — | — | 37.4 | — | 23.1 | — | |
| ACME2024.12 | 51.9 | — | — | 34.7 | — | 23.6 | — | |
| LANA2024.10 | 51.72 | 18.83 | 57.2 | 36.45 | 32.95 | 22.85 | — | |
| LANA2026.05 | 51.72 | 18.83 | 57.2 | 36.45 | 32.95 | 22.85 | — | |
| FDA2024.12 | 49.6 | — | — | 36.4 | — | 22 | — | |
| NaviLLM2025.09 | 43.5 | — | 56.2 | 34.5 | — | — | — | |
| TD-STP2024.10 | 35.89 | — | 40.26 | 27.51 | 19.88 | 15.4 | — | |
| HAMT + FeedTTAFeedback=Yes, Backbone=HAMT2026.05 | 34.14 | — | 38.62 | 29.07 | — | 14.36 | 7.32 | |
| HOP+2024.04 | 33.82 | — | — | 28.24 | 20.2 | 16.86 | — | |
| HAMT + IDEAFeedback=No, Backbone=HAMT2026.05 | 32.81 | — | 38.14 | 28.52 | — | 14.45 | 245.8 | |
| HAMT + IDEABase Model=HAMT, Time (ms)=245.82026.05 | 32.81 | — | 38.14 | 28.52 | — | 14.45 | — | |
| HAMT + ATENAFeedback=Yes, Backbone=HAMT2026.05 | 32.55 | — | 38.19 | 28.38 | — | 14.32 | 5.13 | |
| ENP-VLNOBERTBackbone=VLNOBERT2024.10 | 30.95 | 15.9 | 34 | 24.46 | 17.14 | 14.12 | — | |
| HAMT + ReCAPFeedback=No, Backbone=HAMT2026.05 | 30.51 | — | 34.11 | 24.27 | — | 13.11 | 213.5 | |
| HAMT + ReCAPBase Model=HAMT, Time (ms)=213.52026.05 | 30.51 | — | 34.11 | 24.27 | — | 13.11 | — | |
| HAMT2023.03 | 30.4 | 13.62 | 33.41 | 26.67 | — | — | — | |
| HAMT2024.04 | 30.4 | — | — | 26.67 | 14.88 | 13.08 | — | |
| HAMT2024.10 | 30.4 | 13.62 | 33.41 | 26.67 | 14.88 | 13.08 | — | |
| HAMT2026.03 | 30.4 | — | — | 26.67 | 14.88 | 13.08 | — | |
| HAMT2026.04 | 30.4 | 13.62 | 33.41 | 26.67 | — | — | — | |
| HAMTFeedback=No2026.05 | 30.4 | — | 33.41 | 26.67 | — | 13.08 | 74.9 | |
| HAMTBase Model=HAMT, Time (ms)=74.92026.05 | 30.4 | — | 33.41 | 26.67 | — | 13.08 | — | |
| HAMT2026.05 | 30.4 | 13.62 | 33.41 | 26.67 | 14.88 | 13.08 | — | |
| HAMT + FSTTAFeedback=No, Backbone=HAMT2026.05 | 30.39 | — | 33.39 | 26.65 | — | 13.61 | 613.4 | |
| HAMT + FSTTABase Model=HAMT, Time (ms)=613.42026.05 | 30.39 | — | 33.39 | 26.65 | — | 13.61 | — | |
| AirBERT2023.03 | 30.28 | 17.91 | 34.2 | 23.61 | — | — | — | |
| Airbert2024.10 | 30.28 | 17.91 | 34.2 | 23.61 | 16.83 | 13.28 | — | |
| Airbert2026.05 | 30.28 | 17.91 | 34.2 | 23.61 | 16.83 | 13.28 | — | |
| HAMT + RLCFFeedback=Yes, Backbone=HAMT2026.05 | 30.21 | — | 33.11 | 26.13 | — | 12.97 | 2.32 | |
| HOP2023.03 | 30.17 | 16.38 | 33.06 | 24.34 | — | — | — | |
| HOP2024.10 | 30.17 | 16.38 | 33.06 | 24.34 | 17.69 | 14.34 | — | |
| HOP2026.05 | 30.17 | 16.38 | 33.06 | 24.34 | 17.69 | 14.34 | — | |
| VLNBERT2023.03 | 29.61 | 15.68 | 32.91 | 23.99 | — | — | — | |
| VLNOBERT2024.10 | 29.61 | 15.86 | 32.91 | 23.99 | 16.5 | 13.51 | — | |
| RecBERT2026.04 | 29.61 | 15.86 | 32.91 | 23.99 | — | — | — | |
| RecBERT2026.05 | 29.61 | 15.86 | 32.91 | 23.99 | 16.5 | 13.51 | — | |
| HAMT + SARBase Model=HAMT, Time (ms)=197.252026.05 | 25.86 | — | 27.94 | 23.08 | — | 11.58 | — | |
| HAMT + ViDABase Model=HAMT, Time (ms)=5.49 × 10³2026.05 | 24.81 | — | 27.03 | 22.45 | — | 11.23 | — | |
| HAMT + TentFeedback=No, Backbone=HAMT2026.05 | 23.73 | — | 25.06 | 21.78 | — | 10.82 | 147.7 | |
| HAMT + TentBase Model=HAMT, Time (ms)=147.72026.05 | 23.73 | — | 25.06 | 21.78 | — | 10.82 | — |