mlsanityData to 14 Oct 2026

JanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation

arXiv 2509.22548 ↗First posted 26 Sep 2025cs.CV cs.ROcited by 89

Not on the map. Why?

Abstract (arXiv)

Vision-and-Language Navigation requires an embodied agent to navigate through unseen environments, guided by natural language instructions and a continuous video stream. Recent advances in VLN have been driven by the powerful semantic understanding of Multimodal Large Language Models. However, these methods typically rely on explicit semantic memory, such as building textual cognitive maps or storing historical visual frames. This type of method suffers from spatial information loss, computational redundancy, and memory bloat, which impede efficient navigation. Inspired by the implicit scene representation in human navigation, analogous to the left brain's semantic understanding and the right brain's spatial cognition, we propose JanusVLN, a novel VLN framework featuring a dual implicit neural memory that models spatial-geometric and visual-semantic memory as separate, compact, and fixed-size neural representations. This framework first extends the MLLM to incorporate 3D prior knowledge from the spatial-geometric encoder, thereby enhancing the spatial reasoning capabilities of models based solely on RGB input. Then, the historical key-value caches from the spatial-geometric and visual-semantic encoders are constructed into a dual implicit memory. By retaining only the KVs of tokens in the initial and sliding window, redundant computation is avoided, enabling efficient incremental updates. Extensive experiments demonstrate that JanusVLN outperforms over 20 recent methods to achieve SOTA performance. For example, the success rate improves by 10.5-35.5 compared to methods using multiple data types as input and by 3.6-10.8 compared to methods using more RGB training data. This indicates that the proposed dual implicit neural memory, as a novel paradigm, explores promising new directions for future VLN research. Ours project page: https://miv-xjtu.github.io/JanusVLN.github.io/.

Key facts. Cited by 89 papers in the map (data through 14 Oct 2026). Builds on 72 of its 120 references that are in the map. First posted on arXiv 26 Sep 2025.

Builds on

72 of 120 references.

References · 72

  1. 2025Qwen2.5-VL Technical Reportcited by 5,587
  2. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  3. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,186
  4. 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
  5. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,376
  6. 2019Habitat: A Platform for Embodied AI Researchcited by 1,191
  7. 2025SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Featurescited by 1,188
  8. 2017Vision-and-Language Navigation: Interpreting visually-grounded navigation instructions in real environmentscited by 1,027
  9. 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilitiescited by 706
  10. 2020Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environmentscited by 441
  11. 2020Room-Across-Room: Multilingual Vision-and-Language Navigation with Dense Spatiotemporal Groundingcited by 414
  12. 2025Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Passcited by 367
38 more
  1. 2024NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigationcited by 233
  2. 2025Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligencecited by 217
  3. 2024NaVILA: Legged Robot Vision-Language-Action Model for Navigationcited by 211
  4. 2023LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding, Reasoning, and Planningcited by 210
  5. 2023VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigationcited by 205
  6. 2025FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Drivingcited by 166
  7. 2024InstructNav: Zero-shot System for Generic Instruction Navigation in Unexplored Environmentcited by 157
  8. 2024Uni-NaVid: A Video-based Vision-Language-Action Model for Unifying Embodied Navigation Taskscited by 156
  9. 2022Bridging the Gap Between Learning in Discrete and Continuous Environments for Vision-and-Language Navigationcited by 132
  10. 2021Waypoint Models for Instruction-guided Navigation in Continuous Environmentscited by 129
  11. 2023GridMM: Grid Memory Map for Vision-and-Language Navigationcited by 124
  12. 2025Learning from Videos for 3D World: Enhancing MLLMs with 3D Vision Geometry Priorscited by 120
  13. 2025StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modelingcited by 118
  14. 2023Scaling Data Generation in Vision-and-Language Navigationcited by 117
  15. 2021Language-Aligned Waypoint (LAW) Supervision for Vision-and-Language Navigation in Continuous Environmentscited by 94
  16. 2022Sim-2-Sim Transfer for Vision-and-Language Navigation in Continuous Environmentscited by 86
  17. 2023DREAMWALKER: Mental Planning for Continuous Vision-Language Navigationcited by 85
  18. 2025CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsificationcited by 74
  19. 2024HM3D-OVON: A Dataset and Benchmark for Open-Vocabulary Object Goal Navigationcited by 71
  20. 2024Affordances-Oriented Planning using Foundation Models for Continuous Vision-Language Navigationcited by 65
  21. 2025MapNav: A Novel Memory Representation via Annotated Semantic Maps for Vision-and-Language Navigationcited by 60
  22. 2025LION-FS: Fast & Slow Video-Language Thinker as Online Video Assistantcited by 58
  23. 2025Frequency-Aligned Knowledge Distillation for Lightweight Spatiotemporal Forecastingcited by 55
  24. 2025Move to Understand a 3D Scene: Bridging Visual Grounding and Exploration for Efficient and Versatile Embodied Navigationcited by 51
  25. 2025NavMorph: A Self-Evolving World Model for Vision-and-Language Navigation in Continuous Environmentscited by 45
  26. 2023Learning Navigational Visual Representations with Semantic Map Supervisioncited by 43
  27. 2024Sim-to-Real Transfer via 3D Feature Fields for Vision-and-Language Navigationcited by 39
  28. 2025Chat-Driven Text Generation and Interaction for Person Retrievalcited by 34
  29. 2025Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Modelscited by 32
  30. 2025Contextual Integrity in LLMs via Reasoning and Reinforcement Learningcited by 31
  31. 2025Towards Universal Modal Tracking with Online Dense Temporal Token Learningcited by 30
  32. 2025EmbodieDreamer: Advancing Real2Sim2Real Transfer for Policy Training via Embodied World Modelingcited by 29
  33. 2024MapDistill: Boosting Efficient Camera-based HD Map Construction via Camera-LiDAR Fusion Model Distillationcited by 26
  34. 2024g3D-LF: Generalizable 3D-Language Feature Fields for Embodied Taskscited by 22
  35. 2025PromptSculptor: Multi-Agent Based Text-to-Image Prompt Optimizationcited by 19
  36. 2025Selective KV-Cache Sharing to Mitigate Timing Side-Channels in LLM Inferencecited by 18
  37. 2025Knowledge-Aligned Counterfactual-Enhancement Diffusion Perception for Unsupervised Cross-Domain Visual Emotion Recognitioncited by 17
  38. 2024YOLO-PPA based Efficient Traffic Sign Detection for Cruise Control in Autonomous Drivingcited by 16

50 most cited of 72 All 72 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 13; 2026: 76.

89 papers.

Sort:

By area:

  1. 2025Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Viewscited by 65
  2. 2025HUD: Hierarchical Uncertainty-Aware Disambiguation Network for Composed Video Retrievalcited by 32
  3. 2026VGGT-Omegacited by 26
  4. 2026SCRIBE: Structured Mid-Level Supervision for Tool-Using Language Modelscited by 21
  5. 2026Air-Know: Arbiter-Calibrated Knowledge-Internalizing Robust Network for Composed Image Retrievalcited by 18
  6. 2026ConeSep: Cone-based Robust Noise-Unlearning Compositional Network for Composed Image Retrievalcited by 17
  7. 2025Learning How to Use Tools, Not Just When: Pattern-Aware Tool-Integrated Reasoningcited by 15
  8. 2025AMAS: Adaptively Determining Communication Topology for LLM-based Multi-Agent Systemcited by 15
  9. 2025FedAdamW: A Communication-Efficient Optimizer with Convergence and Generalization Guarantees for Federated Large Modelscited by 15
  10. 2026VLingNav: Embodied Navigation with Adaptive Reasoning and Visual-Assisted Linguistic Memorycited by 10
  11. 2026TIC-VLA: A Think-in-Control Vision-Language-Action Model for Robot Navigation in Dynamic Environmentscited by 10
  12. 2026AgentVLN: Towards Agentic Vision-and-Language Navigationcited by 9
38 more
  1. 2026Seeing with You: Perception-Reasoning Coevolution for Multimodal Reasoningcited by 9
  2. 2025Reward Auditor: Inference on Reward Modeling Suitability in Real-World Perturbed Scenarioscited by 8
  3. 2025FT-MDT: Extracting Decision Trees from Medical Texts via a Novel Low-rank Adaptation Methodcited by 7
  4. 2026ABot-Claw: A Foundation for Persistent, Cooperative, and Self-Evolving Robotic Agentscited by 7
  5. 2026Recursive Self-Evolving Agents via Held-Out Selectioncited by 7
  6. 2025CATCH: A Modular Cross-domain Adaptive Template with Hookcited by 6
  7. 2026RoboRouter: Training-Free Policy Routing for Robotic Manipulationcited by 6
  8. 2026When Models Learn to Ask Why: Adaptive Causal Reasoning for Trustworthy Medical Vision-Language Modelscited by 6
  9. 2026One Token Per Frame: Reconsidering Visual Bandwidth in World Models for VLA Policycited by 6
  10. 2026MindVLA-U1: VLA Beats VA with Unified Streaming Architecture for Autonomous Drivingcited by 6
  11. 2026From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AIcited by 6
  12. 2026textscNaVIDA: Vision-Language Navigation with Inverse Dynamics Augmentationcited by 5
  13. 2026MapDream: Task-Driven Map Learning for Vision-Language Navigationcited by 4
  14. 2026FineCog-Nav: Integrating Fine-grained Cognitive Modules for Zero-shot Multimodal UAV Navigationcited by 4
  15. 2026Embodied3DBench: Benchmarking Low-Level Embodied Spatial Intelligence of Vision Language Modelscited by 4
  16. 2025IdealTSF: Can Non-Ideal Data Contribute to Enhancing the Performance of Time Series Forecasting Models?cited by 3
  17. 2026Adaptive Debiasing Tsallis Entropy for Test-Time Adaptationcited by 3
  18. 2026PROSPECT: Unified Streaming Vision-Language Navigation via Semantic--Spatial Fusion and Latent Predictive Representationcited by 3
  19. 2026RegFormer++: An Efficient Large-Scale 3D LiDAR Point Registration Network with Projection-Aware 2D Transformercited by 3
  20. 2026FAST: A Synergistic Framework of Attention and State-space Models for Spatiotemporal Traffic Predictioncited by 3
  21. 2026Learning Action Manifold with Multi-view Latent Priors for Robotic Manipulationcited by 3
  22. 2026ReAlign: Generalizable Image Forgery Detection via Reasoning-Aligned Representationcited by 3
  23. 2026Uni-LaViRA: Language-Vision-Robot Actions Translation for Unified Embodied Navigationcited by 3
  24. 2026HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harnesscited by 3
  25. 2025Efficient-VLN: A Simple yet Strong Baseline for Efficient Vision-Language Navigationcited by 2
  26. 2026FARTrack: Fast Autoregressive Visual Tracking with High Performancecited by 2
  27. 2026MerNav: A Highly Generalizable Memory-Execute-Review Framework for Zero-Shot Object Goal Navigationcited by 2
  28. 2026Does Peer Observation Help? Vision-Sharing Collaboration for Vision-Language Navigationcited by 2
  29. 2026MathAgent: Adversarial Evolution of Constraint Graphs for Mathematical Reasoning Data Synthesiscited by 2
  30. 2026OC-Distill: Ontology-aware Contrastive Learning with Cross-Modal Distillation for ICU Risk Predictioncited by 2
  31. 2026SpaAct: Spatially-Activated Transition Learning with Curriculum Adaptation for Vision-Language Navigationcited by 2
  32. 2026World-Ego Modeling for Long-Horizon Evolution in Hybrid Embodied Taskscited by 2
  33. 2026PSG-Nav: Probabilistic Scene Graph Navigation via Multiverse Decision Makingcited by 2
  34. 2026Goal2Pixel: Grounding Goals to Pixels for Vision-Language Navigationcited by 2
  35. 2026LIME: Learning Intent-aware Camera Motion from Egocentric Videocited by 2
  36. 2025DiagnoLLM: A Hybrid Bayesian Neural Language Framework for Interpretable Disease Diagnosiscited by 1
  37. 2025Synergy over Discrepancy: A Partition-Based Approach to Multi-Domain LLM Fine-Tuningcited by 1
  38. 2026BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviationcited by 1

50 most cited of 89 All 89 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Mem4Nav: Boosting Vision-and-Language Navigation in Urban Environments with a Hierarchical Spatial-Cognition Long-Short Memory System
  2. 2026SparseNav: Instruction-conditioned Sparse Semantic Perception for Training-Free Vision-Language Navigation
  3. 2025MSNav: Zero-Shot Vision-and-Language Navigation with Dynamic Memory and LLM Spatial Reasoning
  4. 2026SpaceVLN: A Zero-Shot Vision-and-Language Navigation Agent with Online Spatial Cognitive Memory and Reasoning
  5. 2026Towards Dual-Brain Minimal Sufficient Representation for Vision-Language Navigation
  6. 2026Bridging the 2D-3D Gap: A Hierarchical Semantic-Geometric Map for Vision Language Navigation
  7. 2026ABot-N1: Toward a General Visual Language Navigation Foundation Model
  8. 2026AdaGeoVLN: Selective Geometry Across Representation Depth and Navigation Time for Vision-Language Navigation ↗