# Mem4Nav: Boosting Vision-and-Language Navigation in Urban Environments with a Hierarchical Spatial-Cognition Long-Short Memory System

arXiv: https://arxiv.org/abs/2506.19433 · HTML: https://mlsanity.com/paper/2506.19433/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 39 references that are in the map.
- First posted on arXiv 24 Jun 2025.

## Builds on (most cited first, 30 of 31)

- [Vision-and-Language Navigation: Interpreting visually-grounded navigation instructions in real environments](https://mlsanity.com/paper/1711.07280/) (2017)
- [ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks](https://mlsanity.com/paper/1912.01734/) (2019)
- [Beyond the Nav-Graph: Vision-and-Language Navigation in Continuous Environments](https://mlsanity.com/paper/2004.02857/) (2020)
- [LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and Action](https://mlsanity.com/paper/2207.04429/) (2022)
- [NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models](https://mlsanity.com/paper/2305.16986/) (2023)
- [Touchdown: Natural Language Navigation and Spatial Reasoning in Visual Street Environments](https://mlsanity.com/paper/1811.12354/) (2018)
- [NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigation](https://mlsanity.com/paper/2402.15852/) (2024)
- [Think Global, Act Local: Dual-scale Graph Transformer for Vision-and-Language Navigation](https://mlsanity.com/paper/2202.11742/) (2022)
- [ETPNav: Evolving Topological Planning for Vision-Language Navigation in Continuous Environments](https://mlsanity.com/paper/2304.03047/) (2023)
- [Bridging the Gap Between Learning in Discrete and Continuous Environments for Vision-and-Language Navigation](https://mlsanity.com/paper/2203.02764/) (2022)
- [Airbert: In-domain Pretraining for Vision-and-Language Navigation](https://mlsanity.com/paper/2108.09105/) (2021)
- [AerialVLN: Vision-and-Language Navigation for UAVs](https://mlsanity.com/paper/2308.06735/) (2023)
- [Vision-and-Language Navigation: A Survey of Tasks, Methods, and Future Directions](https://mlsanity.com/paper/2203.12667/) (2022)
- [Topological Planning with Transformers for Vision-and-Language Navigation](https://mlsanity.com/paper/2012.05292/) (2020)
- [Learning to Navigate in Cities Without a Map](https://mlsanity.com/paper/1804.00168/) (2018)
- [NavGPT-2: Unleashing Navigational Reasoning Capability for Large Vision-Language Models](https://mlsanity.com/paper/2407.12366/) (2024)
- [Language-Aligned Waypoint (LAW) Supervision for Vision-and-Language Navigation in Continuous Environments](https://mlsanity.com/paper/2109.15207/) (2021)
- [MapGPT: Map-Guided Prompting with Adaptive Path Planning for Vision-and-Language Navigation](https://mlsanity.com/paper/2401.07314/) (2024)
- [VELMA: Verbalization Embodiment of LLM Agents for Vision and Language Navigation in Street View](https://mlsanity.com/paper/2307.06082/) (2023)
- [Lookahead Exploration with Neural Radiance Representation for Continuous Vision-Language Navigation](https://mlsanity.com/paper/2404.01943/) (2024)
- [The Road to Know-Where: An Object-and-Room Informed Sequential BERT for Indoor Vision-Language Navigation](https://mlsanity.com/paper/2104.04167/) (2021)
- [KERM: Knowledge Enhanced Reasoning for Vision-and-Language Navigation](https://mlsanity.com/paper/2303.15796/) (2023)
- [Sim-to-Real Transfer via 3D Feature Fields for Vision-and-Language Navigation](https://mlsanity.com/paper/2406.09798/) (2024)
- [Perceive, Reflect, and Plan: Designing LLM Agent for Goal-Directed City Navigation without Instructions](https://mlsanity.com/paper/2408.04168/) (2024)
- [Analyzing Generalization of Vision and Language Navigation to Unseen Outdoor Areas](https://mlsanity.com/paper/2203.13838/) (2022)
- [GeoVLN: Learning Geometry-Enhanced Visual Representation with Slot Attention for Vision-and-Language Navigation](https://mlsanity.com/paper/2305.17102/) (2023)
- [Generative Language-Grounded Policy in Vision-and-Language Navigation with Bayes' Rule](https://mlsanity.com/paper/2009.07783/) (2020)
- [Generating Landmark Navigation Instructions from Maps as a Graph-to-Text Problem](https://mlsanity.com/paper/2012.15329/) (2020)
- [Exploring Spatial Representation to Enhance LLM Reasoning in Aerial Vision-Language Navigation](https://mlsanity.com/paper/2410.08500/) (2024)
- [A Survey of Large Language Model-Powered Spatial Intelligence Across Scales: Advances in Embodied Agents, Smart Cities, and Earth Science](https://mlsanity.com/paper/2504.09848/) (2025)

## Built on by (most cited first, 3 of 3)

- [EvoMemNav: Efficient Self-Evolving Fine-Grained Memory for Zero-Shot Embodied Navigation](https://mlsanity.com/paper/2606.03509/) (2026)
- [Mosaic: Runtime-Efficient Multi-Agent Embodied Planning](https://mlsanity.com/paper/2607.09603/) (2026)
- [Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning](https://mlsanity.com/paper/2606.00963/) (2026)

## Abstract (arXiv)

Vision-and-Language Navigation (VLN) in large-scale urban environments requires embodied agents to ground linguistic instructions in complex scenes and recall relevant experiences over extended time horizons. Prior modular pipelines offer interpretability but lack unified memory, while end-to-end (M)LLM agents excel at fusing vision and language yet remain constrained by fixed context windows and implicit spatial reasoning. We introduce Mem4Nav, a hierarchical spatial-cognition long-short memory system that can augment any VLN backbone. Mem4Nav fuses a sparse octree for fine-grained voxel indexing with a semantic topology graph for high-level landmark connectivity, storing both in trainable memory tokens embedded via a reversible Transformer. Long-term memory (LTM) compresses and retains historical observations at both octree and graph nodes, while short-term memory (STM) caches recent multimodal entries in relative coordinates for real-time obstacle avoidance and local planning. At each step, STM retrieval sharply prunes dynamic context, and, when deeper history is needed, LTM tokens are decoded losslessly to reconstruct past embeddings. Evaluated on Touchdown and Map2Seq across three backbones (modular, state-of-the-art VLN with prompt-based LLM, and state-of-the-art VLN with strided-attention MLLM), Mem4Nav yields 7-13 pp gains in Task Completion, sufficient SPD reduction, and >10 pp nDTW improvement. Ablations confirm the indispensability of both the hierarchical map and dual memory modules. Our codes are open-sourced via https://github.com/tsinghua-fib-lab/Mem4Nav.