AI Paper Atlas

Research map / Benchmarks / R2R-CE Val-Unseen

R2R-CE Val-Unseen leaderboard

33 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on R2R-CE Val-Unseen. Best success rate: 70.3 (Image2Nav (180° FOV), NeurIPS 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

R2R-CE Val-Unseen · success rate

31 papers, each with its best reported number for its own method; higher is better.

#success rate ↑MethodPaperTopic
170.3Image2Nav (180° FOV)
Qwen3-VL-4B-Instruct
Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator
NeurIPS 2026 · Table 2
3D generation
269.5OmniNav
Qwen2.5-VL-3B-Instruct
OmniNav: A Unified Framework for Prospective Exploration and Visual-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
368.3MobileVLA-R1
NaVILA (LLaMA3-8B)
MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile Robots
ECCV 2026 · Table 2
Robot action and navigation
467.2AgentVLN-3B
Qwen2.5-VL-3B
AgentVLN: Towards Agentic Vision-and-Language Navigation
ECCV 2026 · Table 1
Robot action and navigation
566.2NavForeseeNavForesee: A Unified Vision-Language World Model for Hierarchical Planning and Dual-Horizon Navigation Prediction
CVPR 2026 · Table 1
Robot action and navigation
665.6Ours (Dual-Anchoring)Dual-Anchoring: Addressing State Drift in Vision-Language Navigation
ECCV 2026 · Table 2
Robot action and navigation
765.4AwareVLN (Ours)AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation
CVPR 2026 · Table 1
Robot action and navigation
865.1CorrectNavCorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model
AAAI 2026 · Table 1
Robot action and navigation
964.3DualVLN
Qwen2.5-VL-7B
Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
1062.58ETPNav w/ Phi-Nav
ETPNav
Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation
ECCV 2026 · Table 1
Robot action and navigation
21 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

R2R-CE Val-Unseen · SPL

25 papers, each with its best reported number for its own method; higher is better.

#SPL ↑MethodPaperTopic
166.1OmniNav
Qwen2.5-VL-3B-Instruct
OmniNav: A Unified Framework for Prospective Exploration and Visual-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
265.6Image2Nav (180° FOV)
Qwen3-VL-4B-Instruct
Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator
NeurIPS 2026 · Table 2
3D generation
365.2MobileVLA-R1
NaVILA (LLaMA3-8B)
MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile Robots
ECCV 2026 · Table 2
Robot action and navigation
464.7AgentVLN-3B
Qwen2.5-VL-3B
AgentVLN: Towards Agentic Vision-and-Language Navigation
ECCV 2026 · Table 1
Robot action and navigation
562.3CorrectNavCorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model
AAAI 2026 · Table 1
Robot action and navigation
662.1Ours (Dual-Anchoring)Dual-Anchoring: Addressing State Drift in Vision-Language Navigation
ECCV 2026 · Table 2
Robot action and navigation
759.7NavForeseeNavForesee: A Unified Vision-Language World Model for Hierarchical Planning and Dual-Horizon Navigation Prediction
CVPR 2026 · Table 1
Robot action and navigation
858.5DualVLN
Qwen2.5-VL-7B
Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
958LatentPilotLatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning
ECCV 2026 · Table 1
Robot action and navigation
1056.8JanusVLNJanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
15 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

R2R-CE Val-Unseen · NE

16 papers, each with its best reported number for its own method; lower is better.

#NE ↓MethodPaperTopic
13.74OmniNav
Qwen2.5-VL-3B-Instruct
OmniNav: A Unified Framework for Prospective Exploration and Visual-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
23.88AgentVLN-3B
Qwen2.5-VL-3B
AgentVLN: Towards Agentic Vision-and-Language Navigation
ECCV 2026 · Table 1
Robot action and navigation
33.94NavForeseeNavForesee: A Unified Vision-Language World Model for Hierarchical Planning and Dual-Horizon Navigation Prediction
CVPR 2026 · Table 1
Robot action and navigation
44.02AwareVLN (Ours)AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation
CVPR 2026 · Table 1
Robot action and navigation
54.05DualVLN
Qwen2.5-VL-7B
Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-Language Navigation
ICLR 2026 · Table 1
Robot action and navigation
64.05MobileVLA-R1
NaVILA (LLaMA3-8B)
MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile Robots
ECCV 2026 · Table 2
Robot action and navigation
74.15Ours (Dual-Anchoring)Dual-Anchoring: Addressing State Drift in Vision-Language Navigation
ECCV 2026 · Table 2
Robot action and navigation
84.29ETPNav w/ Phi-Nav
ETPNav
Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation
ECCV 2026 · Table 1
Robot action and navigation
94.41LatentPilotLatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning
ECCV 2026 · Table 1
Robot action and navigation
104.57SACALet's Reward Step-by-Step: Step-Aware Contrastive Alignment for Vision-Language Navigation in Continuous Environments
ECCV 2026 · Table 1
Multimodal LLMs
6 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.