mlsanityData to 7 Oct 2026

Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics

arXiv 2501.10100 ↗First posted 17 Jan 2025cs.RO cs.AI cs.LGcited by 47

On the map: World models → Latent dynamics models and model-based RL

Abstract (arXiv)

Learning robust and generalizable world models is crucial for enabling efficient and scalable robotic control in real-world environments. In this work, we introduce a novel framework for learning world models that accurately capture complex, partially observable, and stochastic dynamics. The proposed method employs a dual-autoregressive mechanism and self-supervised training to achieve reliable long-horizon predictions without relying on domain-specific inductive biases, ensuring adaptability across diverse robotic tasks. We further propose a policy optimization framework that leverages world models for efficient training in imagined environments and seamless deployment in real-world systems. This work advances model-based reinforcement learning by addressing the challenges of long-horizon prediction, error accumulation, and sim-to-real transfer. By providing a scalable and robust framework, the introduced methods pave the way for adaptive and efficient robotic systems in real-world applications.

Key facts. Cited by 47 papers in the map (data through 7 Oct 2026). Placed in Latent dynamics models and model-based RL, one of 11 directions in World models. Builds on 38 of its 47 references that are in the map. First posted on arXiv 17 Jan 2025.

Builds on

38 of 47 references.

Closest ancestors · 14

Same direction, newest first

  1. 2024Diffusion for World Modeling: Visual Details Matter in Atari20 May
  2. 2023TD-MPC2: Scalable, Robust World Models for Continuous Control25 Oct
  3. 2023Finetuning Offline World Models in the Real World24 Oct
  4. 2023STORM: Efficient Stochastic Transformer based World Models for Reinforcement Learning14 Oct
  5. 2023Mastering Diverse Domains through World Models10 Jan
  6. 2022Transformers are Sample-Efficient World Models1 Sep
  7. 2022DayDreamer: World Models for Physical Robot Learning28 Jun
  8. 2022Curious Exploration via Structured World Models Yields Zero-Shot Object Manipulation22 Jun
  9. 2020Mastering Atari with Discrete World Models5 Oct
  10. 2019Dream to Control: Learning Behaviors by Latent Imagination3 Dec
  11. 2019When to Trust Your Model: Model-Based Policy Optimization19 Jun
  12. 2018Recurrent World Models Facilitate Policy Evolution4 Sep
2 more
  1. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models30 May
  2. 2017Neural Network Dynamics for Model-Based Deep Reinforcement Learning with Model-Free Fine-Tuning8 Aug

Other references · 24

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  3. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  4. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,221
  5. 2015End-to-End Training of Deep Visuomotor Policiescited by 847
  6. 2022A Generalist Agentcited by 802
  7. 2020MOPO: Model-based Offline Policy Optimizationcited by 630
  8. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 494
  9. 2022Temporal Difference Learning for Model Predictive Controlcited by 375
  10. 2018Sim-to-Real: Learning Agile Locomotion For Quadruped Robotscited by 277
  11. 2016Deep Visual Foresight for Planning Robot Motioncited by 276
  12. 2016Unsupervised Learning for Physical Interaction through Video Predictioncited by 256
12 more
  1. 2023Orbit: A Unified Simulation Framework for Interactive Robot Learning Environmentscited by 239
  2. 2020Learning Agile Robotic Locomotion Skills by Imitating Animalscited by 229
  3. 2018Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Controlcited by 220
  4. 2022Accelerated Policy Learning with Parallel Differentiable Simulationcited by 92
  5. 2019Data Efficient Reinforcement Learning for Legged Robotscited by 60
  6. 2022Learning Agile Skills via Adversarial Imitation of Rough Partial Demonstrationscited by 30
  7. 2024Learning Quadruped Locomotion Using Differentiable Simulationcited by 22
  8. 2024Adaptive Horizon Actor-Critic for Policy Learning in Contact-Rich Differentiable Simulationcited by 18
  9. 2024FLD: Fourier Latent Dynamics for Structured Motion Representation and Learningcited by 14
  10. 2024Learning to Walk from Three Minutes of Real-World Data with Semi-structured Dynamics Modelscited by 14
  11. 2022Versatile Skill Control via Self-supervised Adversarial Imitation of Unlabeled Mixed Motionscited by 13
  12. 2023Sample-Efficient Learning to Solve a Real-World Labyrinth Game Using Data-Augmented Model-Based Reinforcement Learningcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 14; 2026: 33.

47 papers.

Sort:

By area:

  1. 2026PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulationcited by 74
  2. 2025VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards in World Simulatorscited by 65
  3. 2025A Comprehensive Survey on World Models for Embodied AIcited by 65
  4. 2025WorldEval: World Model as Real-World Robot Policies Evaluatorcited by 64
  5. 2025A Survey: Learning Embodied Intelligence from Physical Simulators and World Modelscited by 50
  6. 2026RISE: Self-Improving Robot Policy with Compositional World Modelcited by 39
  7. 2025What about gravity in video generation? Post-Training Newton's Laws with Verifiable Rewardscited by 23
  8. 2026PlayWorld: Learning Robot World Models from Autonomous Playcited by 16
  9. 2025Uncertainty-Aware Robotic World Model Makes Offline Model-Based Reinforcement Learning Work on Real Robotscited by 15
  10. 2026Action Images: End-to-End Policy Learning via Multiview Video Generationcited by 10
  11. 2026Is the Future Compatible? Diagnosing Dynamic Consistency in World Action Modelscited by 10
  12. 2026SkyJEPA: Learning Long-Horizon World Models for Zero-Shot Sim-to-Real Control of Quadrotorscited by 9
35 more
  1. 2026Hi-WM: Human-in-the-World-Model for Scalable Robot Post-Trainingcited by 8
  2. 2025AirScape: An Aerial Generative World Model with Motion Controllabilitycited by 7
  3. 2025Neural Robot Dynamicscited by 7
  4. 2025DreamPolicy: A Unified World-model Policy for Scalable Humanoid Locomotioncited by 5
  5. 2025Ego-Vision World Model for Humanoid Contact Planningcited by 5
  6. 2026Reparameterization Flow Policy Optimizationcited by 4
  7. 2026HAIC: Humanoid Agile Object Interaction Control via Dynamics-Aware World Modelcited by 3
  8. 2026Safety, Security, and Cognitive Risks in World Modelscited by 3
  9. 2026Morphology-Conditioned World Model for Cross-Embodiment Quadrupedal Locomotioncited by 3
  10. 2026World Models: A Comprehensive Survey of Architectures, Methodologies, Reasoning Paradigms, and Applicationscited by 3
  11. 2025MOSAIC: Skill-Centric Manipulation Planning with Physics Simulationcited by 2
  12. 2025Learning Soft Robotic Dynamics with Active Explorationcited by 2
  13. 2025Target-Bench: Can Video World Models Achieve Mapless Path Planning with Semantic Targets?cited by 2
  14. 2026Learning Visual Feature-Based World Models via Residual Latent Actioncited by 2
  15. 2026SplitAdapter: Load-Aware Humanoid Loco-Manipulation via Factorized Adaptationcited by 2
  16. 2026Describe-Then-Act: Proactive Agent Steering via Distilled Language-Action World Modelscited by 1
  17. 2026Cortex 2.0: Grounding World Models in Real-World Industrial Deploymentcited by 1
  18. 2026Self-supervised Hierarchical Visual Reasoning with World Modelcited by 1
  19. 2026DreamMimic: Learning Visuomotor Whole-Body Loco-Manipulation via World Modelcited by 1
  20. 2026World-Action Models for Robot Learning and Control: A Surveycited by 1
  21. 2026World Models for Embodied Intelligence: From Plausible to Controllable to Actionablecited by 1
  22. 2025Pretraining in Actor-Critic Reinforcement Learning for Locomotioncited by 0
  23. 2026Entropy-Guided k-Guard Sampling for Long-Horizon Autoregressive Video Generationcited by 0
  24. 2026Tempered Sequential Monte Carlo for Trajectory and Policy Optimization with Differentiable Dynamicscited by 0
  25. 2026Learning What Matters: Adaptive Information-Theoretic Objectives for Robot Explorationcited by 0
  26. 2026CaLiSym: Learning Symplectic Dynamics of Real-World Systems through Structured Canonical Liftscited by 0
  27. 2026LUCID: Latent-Skill Unified Control via Imagined Dynamics for Long-Horizon Humanoid Loco-Manipulationcited by 0
  28. 2026Predicting Consequences and Reinforcing Navigation Policies with Latent World Modelscited by 0
  29. 2026Toward Unified Robot Learning: Bridging Representation, Vision-Language-Action, and World Modelscited by 0
  30. 2026Coupled Control and Wireless World Models for Resilient Remote Robotic Controlcited by 0
  31. 2026WM-Craftnet: World Synesthesia Model for Generalizable and Robust Dexterous In-Hand Manipulationcited by 0
  32. 2026Adaptive Rollout Truncation Based on Epistemic Uncertainty for Efficient Offline World Model Trainingcited by 0
  33. 2026Verti-WM: A Physics-Aided Exteroceptive World Model for Off-Road Reinforcement Learningcited by 0
  34. 2026When World Models Lie: Adaptive Safety Analysis Under Wrong Imaginationscited by 0
  35. 2026LocoWM: High-Precision Locomotion through World-Model-Guided Residual Adaptationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Accelerating Model-Based Reinforcement Learning with State-Space World Models
  2. 2026World Model for Robot Learning: A Comprehensive Survey
  3. 2025WMPO: World Model-based Policy Optimization for Vision-Language-Action Models
  4. 2024PWM: Policy Learning with Multi-Task World Models
  5. 2026Feedback World Model Enables Precise Guidance of Diffusion Policy
  6. 2025World Models as Reference Trajectories for Rapid Motor Adaptation
  7. 2025World4RL: Diffusion World Models for Policy Refinement with Reinforcement Learning for Robotic Manipulation
  8. 2026Persistent Robot World Models: Stabilizing Multi-Step Rollouts via Reinforcement Learning