Object Goal Navigation on MP3D
58.9SRVLingNav
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VLingNavTraining-Free=false2026.06 | 58.9 | 26.5 | |
| Memory Over Maps (w/o VLM)Training Free=true, Open Vocab=true2026.03 | 54.5 | 30.5 | |
| Memory Over MapsTraining Free=true, Open Vocab=true2026.03 | 53.9 | 31.2 | |
| IntentNavLearning=Yes2026.06 | 53.8 | 23.1 | |
| STRIVELearning=No2026.06 | 52.3 | 23.1 | |
| MerNavTarget-Free=true, Zero-Shot=true2026.02 | 50.8 | 19.5 | |
| SysNavOpen-Set=true, Zero-Shot=true2026.03 | 50.7 | 18.1 | |
| SysNavLearning=No2026.06 | 50.7 | 18.1 | |
| ConsistNavUnsupervised=Yes, Zero-shot=Yes2026.05 | 50.6 | 25.7 | |
| RIMVenue=IROS’23, Zero-shot=false, Vision Interpolator=-, Language Interpolator=-2026.06 | 50.3 | 17 | |
| ReMemNavTraining-free=true, Zero-shot=true2026.03 | 49.8 | 24.3 | |
| Hydra-Nav-SFTLearning=Yes2026.06 | 49 | 16.5 | |
| EvolveNavVenue=-, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=Qwen3-8B [46]2026.06 | 49 | 19.1 | |
| TravExplorerZero-shot=true, Cross-Floor=true2026.05 | 48.8 | 21.1 | |
| FSUNav_CerebrumTraining-Free=Yes, Universal=Yes2026.04 | 48.75 | 26.93 | |
| HiCo-NavTraining Free=true2026.04 | 48.5 | 21.5 | |
| AERR-NavVenue=ours, Zero-shot=true2026.03 | 47.9 | 18.7 | |
| CogNavTraining-Free=Yes, Universal=No2026.04 | 46.6 | 16.1 | |
| Hierarchical 3D Scene Graph (Ours)Training-Free=true2026.06 | 45.9 | 19.6 | |
| TopoNavTarget-Free=true, Zero-Shot=true2026.02 | 45.5 | 16.8 | |
| WMNavTarget-Free=true, Zero-Shot=true2026.02 | 45.4 | 17.2 | |
| WMNavZero-shot=true2026.03 | 45.4 | 17.2 | |
| WMNav-GeminiVenue=IROS’25, Unsupervised=true, Zero-shot=true2026.03 | 45.4 | 17.2 | |
| WMNavTraining Free=true2026.04 | 45.4 | 17.2 | |
| WMNavLearning=No2026.06 | 45.4 | 17.2 | |
| ASCENTVenue=RAL’26, Zero-shot=true2026.03 | 44.5 | 15.5 | |
| ASCENTZero-shot=true, Cross-Floor=true2026.05 | 44.5 | 15.5 | |
| ASCENTVenue=RAL’26, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=Qwen2.5-7B [48]2026.06 | 44.5 | 15.5 | |
| SSMG-NavZero-shot=true2026.03 | 44.3 | 19.1 | |
| FBN-NavTraining-Free=true2026.06 | 42.1 | 18.1 | |
| InstructNavZero-shot=true2026.03 | 42 | 16.1 | |
| CompassNavVenue=ICLR’26, E2E=true, ME=false, Backbone Models=Qwen2.5-VL-7B2025.10 | 42 | 17.5 | |
| CompassNavLearning=Yes2026.06 | 42 | 17.5 | |
| PIGEONLearning=No2026.06 | 41.6 | 14.4 | |
| MFNPTarget-Free=true, Zero-Shot=true2026.02 | 41.1 | 15.4 | |
| MFNPTraining-free=true, Zero-shot=true2026.03 | 41.1 | 15.4 | |
| MFNPZero-shot=true, Cross-Floor=true2026.05 | 41.1 | 15.4 | |
| MFNPVenue=ICRA’25, Zero-shot=true, Vision Interpolator=Qwen-VLChat-Int4 [2], Language Interpolator=Qwen2-7B [47]2026.06 | 41.1 | 15.4 | |
| UniGoalTarget-Free=true, Zero-Shot=true2026.02 | 41 | 16.4 | |
| UniGoalZero-shot=true2026.03 | 41 | 16.4 | |
| UniGoalVenue=CVPR’25, E2E=false, ME=true, Vision Backbone=LLaVA-1.6-7B, Text Backbone=LLaMA-2-7B2025.10 | 41 | 16.4 | |
| Uni-NavVenue=CVPR’25, Unsupervised=true, Zero-shot=true2026.03 | 41 | 16.4 | |
| UniGoalTraining Free=true, Open Vocab=true2026.03 | 41 | 16.4 | |
| UniGoalTraining-free=true, Zero-shot=true2026.03 | 41 | 16.4 | |
| UniGoalTraining-Free=Yes, Universal=Yes2026.04 | 41 | 16.4 | |
| UniGoalTraining Free=true2026.04 | 41 | 16.4 | |
| UniGoalLearning=Yes2026.06 | 41 | 16.4 | |
| UniGoalTraining-Free=true2026.06 | 41 | 16.4 | |
| PEANUT+LOGTraining-Free=No, Universal=No2026.04 | 40.3 | 15.4 | |
| SG-NavTarget-Free=true, Zero-Shot=true2026.02 | 40.2 | 16 | |
| SG-NavZero-shot=true2026.03 | 40.2 | 16 | |
| SG-NavOpen-Set=true, Zero-Shot=true2026.03 | 40.2 | 16 | |
| SG-NavVenue=NeurIPS’24, E2E=false, ME=true, Vision Backbone=LLaVA-1.6-7B, Text Backbone=GPT-42025.10 | 40.2 | 16 | |
| SG-NAVVenue=NeurIPS’24, Zero-shot=true2026.03 | 40.2 | 16 | |
| SG-NavTraining Free=true, Open Vocab=true2026.03 | 40.2 | 16 | |
| SG-NavTraining-free=true, Zero-shot=true2026.03 | 40.2 | 16 | |
| SG-NavTraining-Free=Yes, Universal=No2026.04 | 40.2 | 16 | |
| SG-NavTraining Free=true2026.04 | 40.2 | 16 | |
| SG-NavUnsupervised=Yes, Zero-shot=Yes2026.05 | 40.2 | 16 | |
| SG-NavZero-shot=true, Cross-Floor=false2026.05 | 40.2 | 16 | |
| SG-Nav-GPTLearning=No2026.06 | 40.2 | 16 | |
| SG-NavVenue=NeurIPS’24, Zero-shot=true, Vision Interpolator=LLaVA-1.6-7B [24], Language Interpolator=GPT-4 [1]2026.06 | 40.2 | 16 | |
| SG-NavTraining-Free=true2026.06 | 40.2 | 16 | |
| SG-NavVenue=NeurIPS’24, Unsupervised=true, Zero-shot=true2026.03 | 40.1 | 16 | |
| ActPeptVenue=RAL’24, Zero-shot=true, Vision Interpolator=GraphSAGE [17], Language Interpolator=-2026.06 | 39.8 | 17.4 | |
| ApexNavOpen-Set=true, Zero-Shot=true2026.03 | 39.2 | 17.8 | |
| ApexNavTraining Free=true2026.04 | 39.2 | 17.8 | |
| ApexNavUnsupervised=Yes, Zero-shot=Yes2026.05 | 39.2 | 17.8 | |
| ApexNavZero-shot=true, Cross-Floor=false2026.05 | 39.2 | 17.8 | |
| ApexNavLearning=No2026.06 | 39.2 | 17.8 | |
| ApexNavVenue=RAL’25, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=DeepSeek-V3 [23]2026.06 | 39.2 | 17.8 | |
| ApexNavTraining-Free=true2026.06 | 39.2 | 17.8 | |
| SGMTarget-Free=false, Zero-Shot=true2026.02 | 37.7 | 14.7 | |
| SGMZero-shot=false2026.03 | 37.7 | 14.7 | |
| SGMVenue=CVPR’24, Unsupervised=false, Zero-shot=false2026.03 | 37.7 | 14.7 | |
| SR-NavVenue=Ours, Unsupervised=true, Zero-shot=true2026.03 | 37.7 | 17.1 | |
| SGMTraining-free=false, Zero-shot=true2026.03 | 37.7 | 14.7 | |
| SGMTraining Free=false2026.04 | 37.7 | 14.7 | |
| BeliefMapNavVenue=NeurIPS’25, Zero-shot=true2026.03 | 37.3 | 17.6 | |
| BeliefMapNavLearning=No2026.06 | 37.3 | 17.6 | |
| BeliefMapNavTraining-Free=true2026.06 | 37.3 | 17.6 | |
| OpenFMNavZero-shot=true2026.03 | 37.2 | 15.7 | |
| OpenFMNavOpen-Set=true, Zero-Shot=true2026.03 | 37.2 | 15.7 | |
| OpenFMNavVenue=NAACL’24, Zero-shot=true2026.03 | 37.2 | 15.7 | |
| OpenFMnavVenue=NAACL’24, Unsupervised=true, Zero-shot=true2026.03 | 37.2 | 15.7 | |
| OpenFMNavTraining Free=true, Open Vocab=true2026.03 | 37.2 | 15.7 | |
| OpenFMNavTraining-Free=Yes, Universal=No2026.04 | 37.2 | 15.7 | |
| OpenFMNavUnsupervised=Yes, Zero-shot=Yes2026.05 | 37.2 | 15.7 | |
| OpenFMNavZero-shot=true, Cross-Floor=false2026.05 | 37.2 | 15.7 | |
| OpenFMNavLearning=No2026.06 | 37.2 | 15.7 | |
| OpenFMNavVenue=NAACL-F’24, Zero-shot=true, Vision Interpolator=GPT-4V [1], Language Interpolator=GPT-4 [1]2026.06 | 37.2 | 15.7 | |
| VLFMTarget-Free=false, Zero-Shot=true2026.02 | 36.4 | 17.5 | |
| VLFMZero-shot=true2026.03 | 36.4 | 17.5 | |
| VLFMOpen-Set=true, Zero-Shot=true2026.03 | 36.4 | 17.5 | |
| VLFMVenue=ICRA’24, E2E=false, ME=true, Vision Backbone=BLIP-22025.10 | 36.4 | 17.5 | |
| VLFMVenue=ICRA’24, Zero-shot=true2026.03 | 36.4 | 17.5 | |
| VLFMTraining Free=true, Open Vocab=true2026.03 | 36.4 | 17.5 | |
| VLFMTraining-free=false, Zero-shot=true2026.03 | 36.4 | 17.5 | |
| VLFMTraining-Free=Yes, Universal=No2026.04 | 36.4 | 17.5 | |
| VLFMTraining Free=true2026.04 | 36.4 | 17.5 |