mlsanityData to 7 Oct 2026

A Comprehensive Survey of Cross-Domain Policy Transfer for Embodied Agents

arXiv 2402.04580 ↗First posted 7 Feb 2024cs.RO cs.AI cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

The burgeoning fields of robot learning and embodied AI have triggered an increasing demand for large quantities of data. However, collecting sufficient unbiased data from the target domain remains a challenge due to costly data collection processes and stringent safety requirements. Consequently, researchers often resort to data from easily accessible source domains, such as simulation and laboratory environments, for cost-effective data acquisition and rapid model iteration. Nevertheless, the environments and embodiments of these source domains can be quite different from their target domain counterparts, underscoring the need for effective cross-domain policy transfer approaches. In this paper, we conduct a systematic review of existing cross-domain policy transfer methods. Through a nuanced categorization of domain gaps, we encapsulate the overarching insights and design considerations of each problem setting. We also provide a high-level discussion about the key methodologies used in cross-domain policy transfer problems. Lastly, we summarize the open challenges that lie beyond the capabilities of current paradigms and discuss potential future directions in this field.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Builds on 71 of its 86 references that are in the map. First posted on arXiv 7 Feb 2024.

Builds on

71 of 86 references.

References · 71

  1. 2017Attention Is All You Needcited by 59,932
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  4. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,401
  5. 2022A Generalist Agentcited by 802
  6. 2018Learning Dexterous In-Hand Manipulationcited by 643
  7. 2023BridgeData V2: A Dataset for Robot Learning at Scalecited by 634
  8. 2017Sim-to-Real Transfer of Robotic Control with Dynamics Randomizationcited by 616
  9. 2023End-to-end Autonomous Driving: Challenges and Frontierscited by 485
  10. 2017Time-Contrastive Networks: Self-Supervised Learning from Videocited by 453
  11. 2020Reinforcement Learning with Augmented Datacited by 452
  12. 2022VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Trainingcited by 384
38 more
  1. 2019Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variablescited by 372
  2. 2018Learning to Adapt in Dynamic, Real-World Environments Through Meta-Reinforcement Learningcited by 270
  3. 2021A Survey of Embodied AI: From Simulators to Research Taskscited by 256
  4. 2019Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization without Accessing Target Domain Datacited by 245
  5. 2019VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learningcited by 210
  6. 2018Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experiencecited by 203
  7. 2023LIV: Language-Image Representations and Rewards for Robotic Controlcited by 180
  8. 2017Imitation from Observation: Learning to Imitate Behaviors from Raw Video via Context Translationcited by 167
  9. 2019Active Domain Randomizationcited by 161
  10. 2021Embodied Intelligence via Learning and Evolutioncited by 154
  11. 2019Temporal Cycle-Consistency Learningcited by 151
  12. 2018One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learningcited by 146
  13. 2016EPOpt: Learning Robust Neural Network Policies Using Model Ensemblescited by 136
  14. 2017Preparing for the Unknown: Learning a Universal Policy with Online System Identificationcited by 129
  15. 2019Meta-Sim: Learning to Generate Synthetic Datasetscited by 128
  16. 2018Zero-Shot Visual Imitationcited by 114
  17. 2017Third-Person Imitation Learningcited by 103
  18. 2021XIRL: Cross-embodiment Inverse Reinforcement Learningcited by 101
  19. 2020Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifierscited by 93
  20. 2018Driving Policy Transfer via Modularity and Abstractioncited by 91
  21. 2020State-Only Imitation Learning for Dexterous Manipulationcited by 90
  22. 2020RL-CycleGAN: Reinforcement Learning Aware Simulation-To-Realcited by 89
  23. 2017Learning Invariant Feature Spaces to Transfer Skills with Reinforcement Learningcited by 84
  24. 2017Virtual to Real Reinforcement Learning for Autonomous Drivingcited by 79
  25. 2019BayesSim: adaptive domain randomization via probabilistic inference for robotics simulatorscited by 78
  26. 2022MetaMorph: Learning Universal Controllers with Transformerscited by 78
  27. 2023XSkill: Cross Embodiment Skill Discoverycited by 74
  28. 2019Third-Person Visual Imitation Learning via Decoupled Hierarchical Controllercited by 73
  29. 2019State Alignment-based Imitation Learningcited by 73
  30. 2021Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapescited by 70
  31. 2022When to Trust Your Simulator: Dynamics-Aware Hybrid Offline-and-Online Reinforcement Learningcited by 59
  32. 2016Transfer from Simulation to Real World through Learning Deep Inverse Dynamics Modelcited by 53
  33. 2021A Survey on Deep Domain Adaptation for LiDAR Perceptioncited by 46
  34. 2021Auto-Tuned Sim-to-Real Transfercited by 45
  35. 2020Data-efficient Domain Randomization with Bayesian Optimizationcited by 43
  36. 2020Learning Cross-Domain Correspondence for Control with Dynamics Cycle-Consistencycited by 43
  37. 2021Cross-domain Imitation from Observationscited by 42
  38. 2022DARA: Dynamics-Aware Reward Augmentation in Offline Reinforcement Learningcited by 40

50 most cited of 71 All 71 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 7; 2025: 7; 2026: 2.

16 papers.

Sort:

By area:

  1. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Modelcited by 285
  2. 2024Flow as the Cross-Domain Manipulation Interfacecited by 126
  3. 2025Universal Actions for Enhanced Embodied Foundation Modelscited by 80
  4. 2024Cross-Embodiment Dexterous Grasping with Reinforcement Learningcited by 24
  5. 2024xTED: Cross-Domain Adaptation via Diffusion-Based Trajectory Editingcited by 9
  6. 2024Hybrid Transfer Reinforcement Learning: Provable Sample Efficiency from Shifted-Dynamics Datacited by 8
  7. 2025DmC: Nearest Neighbor Guidance Diffusion Model for Offline Cross-domain Reinforcement Learningcited by 7
  8. 2024Built Different: Tactile Perception to Overcome Cross-Embodiment Capability Differences in Collaborative Manipulationcited by 6
  9. 2025Dual-Robust Cross-Domain Offline Reinforcement Learning Against Dynamics Shiftscited by 5
  10. 2025Hybrid Cross-domain Robust Reinforcement Learningcited by 3
  11. 2024Knowledge Transfer for Cross-Domain Reinforcement Learning: A Systematic Reviewcited by 1
  12. 2025A Multi-Fidelity Control Variate Approach for Policy Gradient Estimationcited by 1
4 more
  1. 2024Representation Alignment from Human Feedback for Cross-Embodiment Reward Learning from Mixed-Quality Demonstrationscited by 0
  2. 2025Policy Regularization on Globally Accessible States in Cross-Dynamics Reinforcement Learningcited by 0
  3. 2026Unifying Value Alignment and Assignment in Cross-Domain Offline Reinforcement Learning with Heterogeneous Datasetscited by 0
  4. 2026Transfer Learning Across Policy Regimes in Adaptive Multi-Agent Systemscited by 0