World models for navigation
Key facts. World models for navigation: 139 papers on arXiv since 2015 (data through 7 Oct 2026). 35 of its 139 mapped papers (25%) cite “Navigation World Models” (2024, arXiv 2412.03572). 20 of its 139 mapped papers (14%) cite “Pathdreamer: A World Model for Indoor Navigation” (2021, arXiv 2105.08756). 15 new papers 17 Sep–7 Oct.
What’s new
17 Sep – 7 Oct 2026
15 new papers, 17 Sep – 7 Oct; 2 this week.
- This week · 2
- SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World Models
- How Much Planning Is Enough? Reducing Search and Computation in World-Model Planning also fits: Latent predictive world models
- Earlier · 13
- EVEWorld: Physical Evolution Supervision for Embodied World Models also fits: Embodied and robot world models
- Social-WM: Safety-Aware Latent World Models for Robot Social Navigation also fits: Embodied and robot world models
- DiffWAM: A Fast and Efficient Navigation World Action Model
- DSWM: Decomposed Spatio-Temporal World Model for Demand-Driven UAV Base Station Repositioning also fits: Latent dynamics models and model-based RL
- AquaWAM: A Dynamics-aware World Action Model for Underwater Embodied Agents
- ForeFly: A Dual-Horizon World Action Model for Aerial Vision-Language Navigation
- DroneWAM: Efficient World Action Model for Drone Visual Navigation
- World SLAM Model: Joint World Modeling for SLAM and Navigation
All 15
- Aim Short to Reach Far: Your Frozen World Model Can Plan Better Than You Think also fits: Latent predictive world models
- Skytopia: Monocular Drone Navigation with Action-Conditioned Latent World Models
- WOLF: World Model Guided LiDAR Exploration with Predictive Frontiers also fits: World modeling from LiDAR generation
- Verti-WM: A Physics-Aided Exteroceptive World Model for Off-Road Reinforcement Learning
- Feeling Terrain Before Crossing: World Models for Off-Road Navigation
Foundations
Most cited by papers here
- Navigation World Models2024cited by 35 here
- Pathdreamer: A World Model for Indoor Navigation2021cited by 20 here
- Habitat: A Platform for Embodied AI Research2019cited by 19 here
- NavMorph: A Self-Evolving World Model for Vision-and-Language Navigation in Continuous Environments2025cited by 13 here
- A Comprehensive Survey on World Models for Embodied AI2025cited by 8 here
Lineage
The papers this direction builds on, oldest first.
Each paper on the line cites the one before.
- 2018World Modelscited by 45 here
- 2018Learning Latent Dynamics for Planning from Pixelscited by 34 here
- 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 40 here
- 2020Mastering Atari with Discrete World Modelscited by 31 here
- 2023Mastering Diverse Domains through World Modelscited by 38 here
- 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 15 here
- 2024DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planningcited by 15 here
- 2024Navigation World Modelscited by 35 here
- 2025Cosmos World Foundation Model Platform for Physical AIcited by 15 here
15 more it builds on, not in this chain
- 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 16 here
- 2018Recurrent World Models Facilitate Policy Evolutioncited by 23 here
- 2019Habitat: A Platform for Embodied AI Researchcited by 19 here
- 2020Denoising Diffusion Probabilistic Modelscited by 14 here
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 15 here
- 2021Pathdreamer: A World Model for Indoor Navigationcited by 20 here
- 2022DayDreamer: World Models for Physical Robot Learningcited by 24 here
- 2022Flow Matching for Generative Modelingcited by 14 here
- 2022Scalable Diffusion Models with Transformerscited by 20 here
- 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 19 here
- 2023GAIA-1: A Generative World Model for Autonomous Drivingcited by 14 here
- 2024Genie: Generative Interactive Environmentscited by 29 here
- 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 15 here
- 2024Understanding World or Predicting Future? A Comprehensive Survey of World Modelscited by 18 here
- 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 18 here
The 24 most cited of 561.
Browse
30 of 139, ranked 6–35
- 6WMNav: Integrating Vision-Language Models into World Models for Object Goal Navigation20257
- 7WorldArena: A Unified Benchmark for Evaluating Perception and Functional Utility of Embodied World Models20266
- 8X-MOBILITY: End-To-End Generalizable Navigation via World Modeling20246
- 9An Efficient and Multi-Modal Navigation System with One-Step World Model20265
- 10RAE-NWM: Navigation World Model in Dense Visual Representation Space20265
- 11Aerial World Model for Long-horizon Visual Generation and Navigation in 3D Space20255
- 12DREAMWALKER: Mental Planning for Continuous Vision-Language Navigation20235
- 13A Survey: Learning Embodied Intelligence from Physical Simulators and World Models20254
- 14MWM: Mobile World Models for Action-Conditioned Consistent Prediction20263
- 15World Model for Robot Learning: A Comprehensive Survey20263
All 30
- 16WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation20263
- 17MAD: Mapping-Aware World Models for Agile Quadrotor Flight20263
- 18AirScape: An Aerial Generative World Model with Motion Controllability20253
- 19Towards Unified World Models for Visual Navigation via Memory-Augmented Planning and Foresight20253
- 20Efficient Image-Goal Navigation with Representative Latent World Model20253
- 21The Essential Role of Causality in Foundation World Models for Embodied AI20243
- 22World Models and Predictive Coding for Cognitive and Developmental Robotics: Frontiers and Challenges20233
- 23Wow, wo, val! A Comprehensive Embodied World Model Evaluation Turing Test20262
- 24ImagineUAV: Aerial Vision-Language Navigation via World-Action Modeling and Kinodynamic Planning20262
- 25WAM-Nav: Asymmetric Latent World-Action Modeling for Unified Visual Navigation20262
- 26Embodied AI Agents: Modeling the World20252
- 27Embodied AI: From LLMs to World Models20252
- 28NavForesee: A Unified Vision-Language World Model for Hierarchical Planning and Dual-Horizon Navigation Prediction20252
- 29AstraNav-World: World Model for Foresight Control and Consistency20252
- 30WHALE: Towards Generalizable and Scalable World Models for Embodied Decision-making20242
- 31Tracking and Planning with Spatial World Models20222
- 32Active Inference-Driven World Modeling for Adaptive UAV Swarm Trajectory Design20261
- 33MotionScape: A Motion-Stratified UAV Video Benchmark for World Modeling and Future Video Generation20261
- 34DeTrack: A Benchmark and Altitude-Aware Dual World Model for Drone-embodied Tracking20261
- 35FlyMirage: A Fully Automated Generation Pipeline for Diverse and Scalable UAV Flight Data via Generative World Model20261