List your company

Vision-and-Language Navigation on R2R (val seen)

83.74Success Rate (SR)GOAT
13.051231.403149.75568.1069Aug 20, 2021May 27, 2022Mar 3, 2023Dec 8, 2023Sep 13, 2024Jun 20, 2025Mar 27, 2026

Evaluation results

MethodLinks
GOAT2024.0483.741.7988.6479.48—————
TSOR2026.03822.05—76—————
VLN-BERTSpeaker=true, Follower=true2021.0881.862.3586.5778————10.61
AirbertSpeaker=true, Follower=true2021.0881.42.1387.1777————10.63
DSRG2024.04812.238876—————
BEVBert2024.04812.178874—————
BEVBertSetting=Cross-modal backbone, Year=20222026.01812.178874———13.56—
ScaleVLNSetting=Cross-modal backbone, Year=2023, multi-task pretraining=true2026.01812.128775———13.24—
BEVBert2026.02812.178874———13.56—
ScaleVLN2026.02812.128775———13.24—
KESU2026.03812.19—75—————
BEVBert2026.03812.17—74—————
BTK2026.03812.098875—————
GridMM2026.03802.34—74—————
KERM2026.03802.19—74—————
DUET2024.04792.288673—————
GeoVLN2024.04792.22—76—————
DUETSetting=Cross-modal backbone, Year=20222026.01792.288673———12.32—
DUET2026.02792.288673———12.32—
DUET2026.03792.288673—————
CONSOLE2026.03792.17—73—————
ViTeC2026.03792.22—74—————
VLN-SIGVisual Features=CLIP-ViT/162023.0478.6——74.8—81.872.2——
HOP+2026.03782.33—73—————
TD-STP2024.04772.348373—————
EnvEdit2024.04772.32—74—————
TD-STP2026.03772.348373—————
HAMT2024.04762.518272—————
GELA2024.04762.39—73—————
HAMTSetting=Cross-modal backbone, Year=20212026.01762.52—72———11.15—
HAMT2026.02762.52—72———11.15—
HAMT2026.03762.348272—————
Esceme2026.03762.57—73—————
DUET-CLIPVisual Features=CLIP-ViT/162023.0475.2——69.1—7666.5——
AirbertSpeaker=false, Follower=true2021.0874.222.7282.9470————10.69
NavGPT-2Setting=Prompting-based LLM agents, Year=2024, additional synthetic trajectories=true2026.01742.848363———14.13—
NavGPT22026.03742.848363—————
AirbertSpeaker=false, Follower=false2021.0873.853.2180.7169————10.59
LSAL2026.03732.88—70—————
VLN-BERTSpeaker=false, Follower=false2021.0870.23.7376.4766————10.28
Speaker FollowerSetting=Cross-modal backbone, Year=20182026.01663.3674——————
SpeakerFollower2026.02663.3674——————
EnvDrop2026.03623.99—59—————
Self-MonitoringTraining Paradigm=Naïve CL2021.11614.0869.854.664.9————
Self-MonitoringTraining Paradigm=Naïve CL2021.11614.0869.854.664.966.253.9——
Self-MonitoringTraining Paradigm=Self-Paced CL2021.1158.84.1968.253.365.4————
Self-MonitoringTraining Paradigm=Self-Paced CL2021.1158.84.1968.253.365.466.352.2——
FollowerTraining Paradigm=Self-Paced CL2021.1158.74.2369.251.163.3————
FollowerTraining Paradigm=Self-Paced CL2021.1158.74.2369.251.163.364.550.5——
Self-MonitoringTraining Paradigm=Machine Learning2021.1158.44.276751.964.1————
Self-MonitoringTraining Paradigm=Base2021.1158.44.276751.964.165.451.5——
EnvDropTraining Paradigm=Self-Paced CL2021.1158.14.4265.654.867.4————
EnvDropTraining Paradigm=Self-Paced CL2021.1158.14.4265.654.867.467.751.6——
EnvDropTraining Paradigm=Naïve CL2021.1157.84.4963.154.867.2————
EnvDropTraining Paradigm=Naïve CL2021.1157.84.4963.154.867.267.551.5——
EnvDropTraining Paradigm=Machine Learning2021.1157.74.5565.654.467.2————
EnvDropTraining Paradigm=Base2021.1157.74.5565.654.467.267.251.4——
DV-VLNSetting=Trainable language-only agents, Year=20252026.01545.036649———10.53—
FollowerTraining Paradigm=Machine Learning2021.1152.34.8565.344.358.2————
FollowerTraining Paradigm=Base2021.1152.34.8565.344.358.259.444.1——
FollowerTraining Paradigm=Naïve CL2021.1148.65.036240.455.9————
FollowerTraining Paradigm=Naïve CL2021.1148.65.036240.455.957.440.5——
NavCoTSetting=Trainable language-only agents, Year=20252026.01416.464838———10.08—
Seq2SeqSetting=Cross-modal backbone, Year=20182026.01396.0153————11.33—
Seq2Seq2026.02396.0153————11.33—
Seq2Seq2026.03396.01———————
Retrieval-augmented LLM navigatorBackbone=Qwen3-8B2026.0219.888.9639.8613.29———16.2—
NavGPTBackbone=Qwen3-8B2026.0215.778.8129.8710.3———15.29—
EnvDrop2022.03————67.267.2———
HOP2022.03————73.376.1———
Self-Monitoring2022.03————64.165.4———
Syntax2022.03————7070———