Object Goal Navigation on HM3D
100Success RateLoTIS + Obstacle Avoidance
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LoTIS + Obstacle AvoidanceInitialization=On-Trajectory, Camera Setup=Matched, Obstacle Avoidance=true2026.02 | 100 | 99.8 | — | |
| LoTISInitialization=On-Trajectory, Camera Setup=Matched, Obstacle Avoidance=false2026.02 | 98.5 | 98.4 | — | |
| LoTIS + Obstacle AvoidanceInitialization=On-Trajectory, Camera Setup=Cross, Obstacle Avoidance=true2026.02 | 97.5 | 96.7 | — | |
| LoTIS + Obstacle AvoidanceInitialization=Off-Trajectory, Camera Setup=Matched, Obstacle Avoidance=true2026.02 | 95.1 | 89.6 | — | |
| LoTIS + Obstacle AvoidanceInitialization=Off-Trajectory, Camera Setup=Cross, Obstacle Avoidance=true2026.02 | 92.2 | 85.1 | — | |
| LoTISInitialization=On-Trajectory, Camera Setup=Cross, Obstacle Avoidance=false2026.02 | 90.2 | 90.1 | — | |
| ApexNavTraining Free=true2025.12 | 76.2 | 38 | — | |
| LoTISInitialization=Off-Trajectory, Camera Setup=Cross, Obstacle Avoidance=false2026.02 | 74.5 | 73 | — | |
| LoTISInitialization=Off-Trajectory, Camera Setup=Matched, Obstacle Avoidance=false2026.02 | 74 | 72.3 | — | |
| XGXVenue=ICRA’24, Zero-shot=false, Vision Interpolator=-, Language Interpolator=-2026.06 | 72.9 | 35.7 | — | |
| CogNavTraining Free=true2025.12 | 72.5 | 26.2 | — | |
| EvolveNavVenue=-, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=Qwen3-8B [46]2026.06 | 67.3 | 33.9 | — | |
| ViNTInitialization=On-Trajectory, Camera Setup=Matched2026.02 | 65.7 | 64.4 | — | |
| ASCENTVenue=RAL’26, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=Qwen2.5-7B [48]2026.06 | 65.4 | 33.5 | — | |
| PIRLNavVenue=CVPR’23, Zero-shot=false, Vision Interpolator=-, Language Interpolator=-2026.06 | 64.1 | 27.1 | — | |
| MSGNavVenue=CVPR’26, Zero-shot=true, Vision Interpolator=GPT-4o [19], Language Interpolator=GPT-4o [19]2026.06 | 63 | 31.4 | — | |
| ImagineNav++(Oracle)Open-Vocabulary=true, Mapless=true2025.12 | 62.5 | 32.8 | — | |
| OVRLOpen-Vocabulary=false, Mapless=true2025.12 | 62 | 26.8 | — | |
| Schrödinger’s NavigatorTraining Free=true2025.12 | 60.9 | 23.7 | 2.23 | |
| FAINTInitialization=On-Trajectory, Camera Setup=Matched2026.02 | 60.3 | 59 | — | |
| SGMTraining Free=false2025.12 | 60.2 | 30.8 | — | |
| ApexNavVenue=RAL’25, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=DeepSeek-V3 [23]2026.06 | 59.6 | 33 | — | |
| ImagineNav++Open-Vocabulary=true, Mapless=true2025.12 | 58.5 | 26.6 | — | |
| MFNPVenue=ICRA’25, Zero-shot=true, Vision Interpolator=Qwen-VLChat-Int4 [2], Language Interpolator=Qwen2-7B [47]2026.06 | 58.3 | 26.7 | — | |
| WMNavGeminiZero-shot=true, Universal=false2026.05 | 58.1 | 31.2 | — | |
| InstructNavVenue=CoRL’24, Zero-shot=true, Vision Interpolator=GPT-4V [1], Language Interpolator=GPT-4 [1]2026.06 | 58 | 20.9 | — | |
| RIMVenue=IROS’23, Zero-shot=false, Vision Interpolator=-, Language Interpolator=-2026.06 | 57.8 | 27.2 | — | |
| CompassNavVenue=ICLR’26, E2E=true, ME=false, Backbone Models=Qwen2.5-VL-7B2025.10 | 56.6 | 27.6 | — | |
| TriHelperTraining Free=true2025.12 | 56.5 | 25.3 | 3.87 | |
| PlaceNavInitialization=On-Trajectory, Camera Setup=Matched2026.02 | 55.4 | 54.5 | — | |
| OpenFMNavVenue=NAACL-F’24, Zero-shot=true, Vision Interpolator=GPT-4V [1], Language Interpolator=GPT-4 [1]2026.06 | 54.9 | 24.4 | — | |
| MCNavZero-shot=true, Universal=true2026.05 | 54.8 | 25.1 | — | |
| UniGoalOpen-Vocabulary=true, Mapless=false2025.12 | 54.5 | 25.1 | — | |
| UniGoalVenue=CVPR’25, E2E=false, ME=true, Vision Backbone=LLaVA-1.6-7B, Text Backbone=LLaMA-2-7B2025.10 | 54.5 | 25.1 | — | |
| UniGoalZero-shot=true, Universal=true2026.05 | 54.5 | 25.1 | — | |
| SG-NavOpen-Vocabulary=true, Mapless=false2025.12 | 54 | 24.9 | — | |
| SG-NavVenue=NeurIPS’24, E2E=false, ME=true, Vision Backbone=LLaVA-1.6-7B, Text Backbone=GPT-42025.10 | 54 | 24.9 | — | |
| SG-NavZero-shot=true, Universal=false2026.05 | 54 | 24.9 | — | |
| SG-NavVenue=NeurIPS’24, Zero-shot=true, Vision Interpolator=LLaVA-1.6-7B [24], Language Interpolator=GPT-4 [1]2026.06 | 54 | 24.9 | — | |
| GAMapTraining Free=true2025.12 | 53.1 | 26 | — | |
| GAMapVenue=NeurIPS’24, E2E=false, ME=true, Vision Backbone=CLIP, Text Backbone=GPT-42025.10 | 53.1 | 26 | — | |
| ImagineNavOpen-Vocabulary=true, Mapless=true2025.12 | 53 | 23.8 | — | |
| VLFMTraining Free=true2025.12 | 52.5 | 30.4 | — | |
| VLFMOpen-Vocabulary=true, Mapless=false2025.12 | 52.5 | 30.4 | — | |
| OpenFMNavZero-shot=true, Universal=false2026.05 | 52.5 | 24.1 | — | |
| VLFMVenue=ICRA’24, Zero-shot=false, Vision Interpolator=BLIP-2 [22], Language Interpolator=-2026.06 | 52.5 | 30.4 | — | |
| VLFMVenue=ICRA’24, E2E=false, ME=true, Vision Backbone=BLIP-22025.10 | 52.4 | 30.4 | — | |
| VLFMZero-shot=true, Universal=false2026.05 | 52.4 | 30.3 | — | |
| InstructNavTraining Free=true2025.12 | 51 | 18.7 | 2.89 | |
| VLFMVenue=ICRA’24, Zero-shot=true, Vision Interpolator=BLIP-2 [22], Language Interpolator=-2026.06 | 50.9 | 23.6 | — | |
| GoatOpen-Vocabulary=true, Mapless=false2025.12 | 50.6 | 24.1 | — | |
| GOATZero-shot=false, Universal=true2026.05 | 50.6 | 24.1 | — | |
| L3MVNTraining Free=true2025.12 | 50.4 | 23.1 | 4.43 | |
| L3MVNVenue=IROS’23, E2E=false, ME=true, Text Backbone=GPT-22025.10 | 50.4 | 23.1 | — | |
| L3MVNZero-shot=true, Universal=false2026.05 | 50.4 | 23.1 | — | |
| L3MVNVenue=IROS’23, Zero-shot=true, Vision Interpolator=-, Language Interpolator=RoBERTa-large [25]2026.06 | 50.4 | 23.1 | — | |
| InstructNavVenue=CoRL’24, E2E=false, ME=true, Vision Backbone=GPT-4V, Text Backbone=LLaMA3-70B2025.10 | 50 | 20.9 | — | |
| FAINTInitialization=Off-Trajectory, Camera Setup=Matched2026.02 | 46.1 | 40.6 | — | |
| WMNavQwen2.5Zero-shot=true, Universal=false2026.05 | 46.1 | 20.7 | — | |
| InstructNav*Training Free=true, LLM/VLM=GPT-4o, Representation=3D Gaussian2025.12 | 45.3 | 18.6 | 3.38 | |
| PSLOpen-Vocabulary=true, Mapless=true2025.12 | 42.4 | 19.2 | — | |
| PSLVenue=ECCV’24, E2E=false, ME=false, Vision Backbone=CLIP2025.10 | 42.4 | 19.2 | — | |
| PSLZero-shot=false, Universal=true2026.05 | 42.4 | 19.2 | — | |
| VoroNavTraining Free=true2025.12 | 42 | 26 | — | |
| VoroNavOpen-Vocabulary=true, Mapless=false2025.12 | 42 | 26 | — | |
| VoroNavVenue=ICML’24, E2E=false, ME=true, Vision Backbone=BLIP, Text Backbone=GPT-3.52025.10 | 42 | 26 | — | |
| Habitat-WebOpen-Vocabulary=false, Mapless=true2025.12 | 41.5 | 16 | — | |
| Habitat-WebVenue=CVPR’22, E2E=false, ME=false2025.10 | 41.5 | 16 | — | |
| ESCTraining Free=true2025.12 | 39.2 | 22.3 | — | |
| ESCOpen-Vocabulary=true, Mapless=false2025.12 | 39.2 | 22.3 | — | |
| ESCVenue=ICML’23, E2E=false, ME=true, Text Backbone=GPT-3.52025.10 | 39.2 | 22.3 | — | |
| ESCZero-shot=true, Universal=false2026.05 | 39.2 | 22.3 | — | |
| PixNavTraining Free=false2025.12 | 37.9 | 20.5 | — | |
| SemExpOpen-Vocabulary=false, Mapless=false2025.12 | 37.9 | 18.8 | — | |
| PixNavOpen-Vocabulary=true, Mapless=true2025.12 | 37.9 | 20.5 | — | |
| Pixel-NavVenue=ICRA’24, E2E=false, ME=true, Vision Backbone=LLaMA-Adapter, Text Backbone=GPT-42025.10 | 37.9 | 20.5 | — | |
| PixNavVenue=ICRA’24, Zero-shot=true, Vision Interpolator=LLaMA-Adapter [59], Language Interpolator=GPT-4 [1]2026.06 | 37.9 | 20.5 | — | |
| FAINTInitialization=On-Trajectory, Camera Setup=Cross2026.02 | 34.8 | 31.6 | — | |
| FBEOpen-Vocabulary=false, Mapless=false2025.12 | 33.7 | 15.3 | — | |
| CowZero-shot=false, Universal=false2026.05 | 32 | 18.1 | — | |
| SPNetTraining Free=false2025.12 | 31.2 | 10.1 | — | |
| FAINTInitialization=Off-Trajectory, Camera Setup=Cross2026.02 | 28.9 | 25.3 | — | |
| ZSONTraining Free=false2025.12 | 25.5 | 12.6 | — | |
| ZSONOpen-Vocabulary=true, Mapless=true2025.12 | 25.5 | 12.6 | — | |
| ZSONZero-shot=false, Universal=false2026.05 | 25.5 | 12.6 | — | |
| ZSONVenue=NeurIPS’22, Zero-shot=true, Vision Interpolator=CLIP [29], Language Interpolator=-2026.06 | 25.5 | 12.6 | — | |
| NoMaDInitialization=On-Trajectory, Camera Setup=Matched2026.02 | 25 | 22.2 | — | |
| ViNTInitialization=Off-Trajectory, Camera Setup=Matched2026.02 | 24 | 20.8 | — | |
| NoMaDInitialization=Off-Trajectory, Camera Setup=Matched2026.02 | 14.2 | 11.2 | — | |
| ViNTInitialization=On-Trajectory, Camera Setup=Cross2026.02 | 12.7 | 12.1 | — | |
| NoMaDInitialization=Off-Trajectory, Camera Setup=Cross2026.02 | 12.7 | 10 | — | |
| ViNTInitialization=Off-Trajectory, Camera Setup=Cross2026.02 | 12.3 | 10.3 | — | |
| PlaceNavInitialization=Off-Trajectory, Camera Setup=Matched2026.02 | 11.3 | 10 | — | |
| NoMaDInitialization=On-Trajectory, Camera Setup=Cross2026.02 | 9.3 | 8.2 | — | |
| PlaceNavInitialization=On-Trajectory, Camera Setup=Cross2026.02 | 7.8 | 7.3 | — | |
| PlaceNavInitialization=Off-Trajectory, Camera Setup=Cross2026.02 | 6.4 | 5 | — |