Vision-and-Language Navigation on REVERIE seen (val)
84.33SRDUET + ATENA
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| DUET + ATENAFeedback=Yes, Backbone=DUET2026.05 | 84.33 | 74.31 | 85.52 | — | — | — | 59.99 | 5.42 | |
| DUET + FeedTTAFeedback=Yes, Backbone=DUET2026.05 | 84.19 | 75.54 | 86.16 | — | — | — | 60.32 | 7.64 | |
| GOAT2024.04 | 78.64 | 71.4 | — | — | — | 63.74 | 57.85 | — | |
| DUET + IDEAFeedback=No, Backbone=DUET2026.05 | 78.24 | 67.74 | 78.45 | — | — | — | 55.07 | 344.3 | |
| DUET + IDEABase Model=DUET, Time (ms)=344.32026.05 | 78.24 | 67.74 | 78.45 | — | — | — | 55.07 | — | |
| KERMTraining Strategy=Pre-training and fine-tuning2023.03 | 76.88 | 70.45 | 79.2 | 12.84 | — | — | — | — | |
| DSRG2024.04 | 75.69 | 68.09 | — | — | — | 61.07 | 54.72 | — | |
| DUET + FSTTAFeedback=No, Backbone=DUET2026.05 | 75.48 | 65.84 | 75.59 | — | — | — | 52.23 | 833.6 | |
| DUET + FSTTABase Model=DUET, Time (ms)=833.62026.05 | 75.48 | 65.84 | 75.59 | — | — | — | 52.23 | — | |
| DUET + ReCAPFeedback=No, Backbone=DUET2026.05 | 74.72 | 65.87 | 75.06 | — | — | — | 53.42 | 359.4 | |
| DUET + ReCAPBase Model=DUET, Time (ms)=359.42026.05 | 74.72 | 65.87 | 75.06 | — | — | — | 53.42 | — | |
| DUET + RLCFFeedback=Yes, Backbone=DUET2026.05 | 73.84 | 64.97 | 74.67 | — | — | — | 52.47 | 2.43 | |
| BEVBert2024.04 | 73.72 | 65.32 | — | — | — | 57.7 | 51.73 | — | |
| DUET + ViDABase Model=DUET, Time (ms)=7.31 × 10³2026.05 | 72.49 | 63.49 | 73.99 | — | — | — | 50.89 | — | |
| Meta-Explore*Memory=Top. Map, Exploit=local goal, dataset=modified (partially fixed object categories)2023.03 | 71.89 | 65.71 | 73.44 | 13.03 | — | — | — | — | |
| KERM-ptTraining Strategy=Pre-training2023.03 | 71.89 | 64.04 | 74.49 | 14.25 | — | — | — | — | |
| DUET + TentFeedback=No, Backbone=DUET2026.05 | 71.89 | 64.06 | 73.72 | — | — | — | 50.41 | 258.9 | |
| DUET + TentBase Model=DUET, Time (ms)=258.92026.05 | 71.89 | 64.06 | 73.72 | — | — | — | 50.41 | — | |
| DUETMemory=Top. Map, Exploit=X2023.03 | 71.75 | 63.94 | 73.86 | 13.86 | — | — | — | — | |
| DUET2023.03 | 71.75 | 63.94 | 73.68 | 13.86 | — | — | — | — | |
| DUET2024.04 | 71.75 | 63.94 | — | — | — | 57.41 | 51.14 | — | |
| DUET2026.04 | 71.75 | 63.94 | 73.86 | 13.86 | — | — | — | — | |
| DUETFeedback=No2026.05 | 71.75 | 63.94 | 73.86 | — | — | — | 51.14 | 123.2 | |
| DUETBase Model=DUET, Time (ms)=123.22026.05 | 71.75 | 63.94 | 73.86 | — | — | — | 51.14 | — | |
| DUET + SARBase Model=DUET, Time (ms)=267.52026.05 | 71.75 | 64.43 | 74.84 | — | — | — | 51.7 | — | |
| Meta-ExploreMemory=Top. Map, Exploit=local goal2023.03 | 71.68 | 63.9 | 73.79 | 13.84 | — | — | — | — | |
| DUET+SDB2026.04 | 69.5 | 62.07 | 70.56 | 13.83 | — | — | — | — | |
| SIAMemory=Seq, Exploit=X2023.03 | 61.91 | 57.08 | 65.85 | 13.61 | — | — | — | — | |
| HAMT + ATENAFeedback=Yes, Backbone=HAMT2026.05 | 57.34 | 48.08 | 52.92 | — | — | — | 29.6 | 5.13 | |
| HOP+2024.04 | 55.87 | 49.55 | — | — | — | 40.76 | 36.22 | — | |
| HAMT + FeedTTAFeedback=Yes, Backbone=HAMT2026.05 | 55.8 | 49.7 | 62.97 | — | — | — | 31.8 | 7.32 | |
| HOP2023.03 | 53.76 | 47.19 | 54.88 | 13.8 | — | — | — | — | |
| VLN BERTMemory=Rec, Exploit=X2023.03 | 51.79 | 47.96 | 53.9 | 13.44 | — | — | — | — | |
| VLNBERT2023.03 | 51.79 | 47.96 | 53.9 | 13.44 | — | — | — | — | |
| RecBERT2026.04 | 51.79 | 47.96 | 53.9 | 13.44 | — | — | — | — | |
| FAST-MATTNMemory=Rec, Exploit=X2023.03 | 50.53 | 45.5 | 55.17 | 16.35 | — | — | — | — | |
| HAMT + IDEAFeedback=No, Backbone=HAMT2026.05 | 47.33 | 42.13 | 50.67 | — | — | — | 26.82 | 245.8 | |
| HAMT + IDEABase Model=HAMT, Time (ms)=245.82026.05 | 47.33 | 42.13 | 50.67 | — | — | — | 26.82 | — | |
| AirbertMemory=Seq, Exploit=X2023.03 | 47.01 | 42.34 | 48.98 | 15.16 | — | — | — | — | |
| AirBERT2023.03 | 47.01 | 42.34 | 49.98 | 15.16 | — | — | — | — | |
| HAMT + RLCFFeedback=Yes, Backbone=HAMT2026.05 | 44.21 | 41.13 | 47.92 | — | — | — | 25.97 | 2.32 | |
| HAMT + ReCAPFeedback=No, Backbone=HAMT2026.05 | 44.06 | 40.69 | 48.49 | — | — | — | 25.46 | 213.5 | |
| HAMT + ReCAPBase Model=HAMT, Time (ms)=213.52026.05 | 44.06 | 40.69 | 48.49 | — | — | — | 25.46 | — | |
| HAMT + SARBase Model=HAMT, Time (ms)=197.252026.05 | 43.85 | 39.97 | 47.12 | — | — | — | 25.44 | — | |
| HAMT + ViDABase Model=HAMT, Time (ms)=5.49 × 10³2026.05 | 43.55 | 41.23 | 47.67 | — | — | — | 25.27 | — | |
| HAMT + TentFeedback=No, Backbone=HAMT2026.05 | 43.43 | 40.78 | 46.03 | — | — | — | 25.81 | 147.7 | |
| HAMT + TentBase Model=HAMT, Time (ms)=147.72026.05 | 43.43 | 40.78 | 46.03 | — | — | — | 25.81 | — | |
| HAMTMemory=Seq, Exploit=X2023.03 | 43.29 | 40.19 | 47.65 | 12.79 | — | — | — | — | |
| HAMT2023.03 | 43.29 | 40.19 | 47.65 | 12.79 | — | — | — | — | |
| HAMT2024.04 | 43.29 | 40.19 | — | — | — | 27.2 | 25.18 | — | |
| HAMT2026.04 | 43.29 | 40.19 | 47.65 | 12.79 | — | — | — | — | |
| HAMTFeedback=No2026.05 | 43.29 | 40.19 | 47.65 | — | — | — | 25.18 | 74.9 | |
| HAMTBase Model=HAMT, Time (ms)=74.92026.05 | 43.29 | 40.19 | 47.65 | — | — | — | 25.18 | — | |
| HAMT + FSTTAFeedback=No, Backbone=HAMT2026.05 | 42.87 | 39.56 | 48.21 | — | — | — | 24.58 | 613.4 | |
| HAMT + FSTTABase Model=HAMT, Time (ms)=613.42026.05 | 42.87 | 39.56 | 48.21 | — | — | — | 24.58 | — | |
| SMNAMemory=Rec, Exploit=homing2023.03 | 41.25 | 39.61 | 43.29 | 7.54 | — | — | — | — | |
| SMNA2026.04 | 41.25 | 39.61 | 43.29 | 7.54 | — | — | — | — | |
| Omnigraph + HAMT2024.03 | 40 | 35 | 37 | 9.8 | 44 | — | — | — | |
| Seq2SeqMemory=Rec, Exploit=X2023.03 | 29.59 | 24.01 | 35.7 | 12.88 | — | — | — | — | |
| Seq2Seq2023.03 | 29.59 | 24.01 | 35.7 | 12.88 | — | — | — | — | |
| Seq2Seq2026.04 | 29.59 | 24.01 | 35.7 | 12.88 | — | — | — | — | |
| RCMMemory=Rec, Exploit=X2023.03 | 23.33 | 21.82 | 29.44 | 10.7 | — | — | — | — | |
| RCM2023.03 | 23.33 | 21.82 | 29.44 | 10.7 | — | — | — | — | |
| HAMT2024.03 | 23 | 20 | 22 | 9.8 | 38 | — | — | — |