Object Navigation on HM3D v2 (val)
83SRVLingNav
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VLingNav2026.01 | 83 | 40.5 | |
| STRIVEVenue=ICRA’26, Training Free=Yes, Oracle Free=No, Intermediate Representation=Struct. Text2025.11 | 79.6 | 38.7 | |
| PIGEON-ZeroShotVenue=-, Training Free=Yes, Oracle Free=Yes, Intermediate Representation=No2025.11 | 79.6 | 38.1 | |
| FiLM-Nav2026.01 | 77 | 41.3 | |
| VLingNavtraining=Supervised Fine-Tuning (SFT)2026.01 | 76.4 | 32.6 | |
| ApexNav2026.01 | 76.2 | 38 | |
| ApexNavVenue=RAL’25, Training Free=Yes, Oracle Free=Yes, Intermediate Representation=No2025.11 | 76.2 | 38 | |
| FOM-Nav2025.11 | 75.8 | 47.9 | |
| DSCDZero-Shot=true, Training-Free=true, base VLM=Gemini-2.5-Flash-Lite2026.01 | 73 | 38.7 | |
| WMNavVenue=IROS’25 Oral, Training Free=Yes, Oracle Free=Yes, Intermediate Representation=No2025.11 | 72.2 | 33.3 | |
| PIRLNavreproduced=true2025.11 | 66 | 27 | |
| VLFMreproduced=true2025.11 | 64 | 33 | |
| VLFMZero-Shot=true, Training-Free=false2026.01 | 63.6 | 32.5 | |
| VLFM2026.01 | 63.6 | 32.5 | |
| VLFMVenue=ICRA’24, Training Free=Yes, Oracle Free=Yes, Intermediate Representation=No2025.11 | 63.6 | 32.5 | |
| VLFMsegmentation_model=our segmentation model2025.11 | 60.3 | 33.6 | |
| RIMreproduced=true2025.11 | 58.4 | 22.2 | |
| WMNav Gemini 1.5 ProVision-Language Model=Gemini 1.5 Pro, Zero-shot=true2026.06 | 58.1 | 31.2 | |
| InstructNavZero-Shot=true, Training-Free=true2026.01 | 58 | 20.9 | |
| InstructNav2026.01 | 58 | 20.9 | |
| InstructNavVenue=CoRL’24, Training Free=Yes, Oracle Free=Yes, Intermediate Representation=Struct. Text2025.11 | 58 | 20.9 | |
| InstructNavVision-Language Model=GPT4 + GPT4V, Zero-shot=true2026.06 | 56 | 22.5 | |
| ViTLVision-Language Model=LLaVA-34B, Zero-shot=true2026.06 | 54.1 | 30.9 | |
| SG-Nav-GPTZero-shot=true2026.06 | 54 | 24.9 | |
| WMNav Gemini 1.5 FlashVision-Language Model=Gemini 1.5 Flash, Zero-shot=true2026.06 | 53.5 | 30.7 | |
| GAMapZero-shot=true2026.06 | 53.1 | 26 | |
| VLFMZero-shot=true2026.06 | 52.4 | 30.8 | |
| InstructNavVision-Language Model=GPT4 + LLaVA 34B, Zero-shot=true2026.06 | 50 | 19.4 | |
| SG-NavZero-Shot=true, Training-Free=true2026.01 | 49.6 | 25.5 | |
| SG-Nav2026.01 | 49.6 | 25.5 | |
| L3MVN2026.01 | 36.6 | 15.7 | |
| L3MVNZero-Shot=true, Training-Free=true2026.01 | 36.3 | 15.7 | |
| Habitat-Web2026.01 | 31.6 | 8.5 |