Vision-Language Navigation on GSA-R2R Basic Instructions Non-Residential v1 (test)
57.7SRMemoir (Ours)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Memoir (Ours)Pre-training=Full navigation graph, Retrieval=Enabled2025.10 | 57.7 | 52 | 61.9 | |
| GR-DUETPre-training=Full navigation graph2025.10 | 56.6 | 51.5 | 61 | |
| GR-DUET*Pre-training=Full navigation graph, Experimental conditions=Aligned2025.10 | 55.3 | 50.4 | 60.8 | |
| Memoir (w/o retrieval)Pre-training=Full navigation graph, Retrieval=Disabled2025.10 | 51.6 | 40.8 | 49.6 | |
| DUET + BT2025.10 | 49.5 | 46 | 59.4 | |
| DUET + MLM2025.10 | 48.3 | 38.8 | 48.4 | |
| DUET2025.10 | 48.1 | 37.3 | 45.9 | |
| DUET + MRC2025.10 | 48.1 | 37.3 | 45.9 | |
| DUET + TENT2025.10 | 46.5 | 33.7 | 42.6 | |
| DUET + SAR2025.10 | 44.6 | 31.5 | 40.6 | |
| OVER-NAV2025.10 | 16.6 | 13 | 35 | |
| TourHAMT2025.10 | 11 | 8.6 | 32.2 |