mlsanityData to 7 Oct 2026

Solving Rubik's Cube with a Robot Hand

arXiv 1910.07113 ↗First posted 16 Oct 2019cs.LG cs.AI cs.CVcited by 699

On the map: World models → Embodied and robot world models

Abstract (arXiv)

We demonstrate that models trained only in simulation can be used to solve a manipulation problem of unprecedented complexity on a real robot. This is made possible by two key components: a novel algorithm, which we call automatic domain randomization (ADR) and a robot platform built for machine learning. ADR automatically generates a distribution over randomized environments of ever-increasing difficulty. Control policies and vision state estimators trained with ADR exhibit vastly improved sim2real transfer. For control policies, memory-augmented models trained on an ADR-generated distribution of environments show clear signs of emergent meta-learning at test time. The combination of ADR with our custom robot platform allows us to solve a Rubik's cube with a humanoid robot hand, which involves both control and state estimation problems. Videos summarizing our results are available: https://openai.com/blog/solving-rubiks-cube/

Key facts. Cited by 699 papers in the map (data through 7 Oct 2026). Placed in Embodied and robot world models, one of 11 directions in World models. Builds on 1 of its 118 references that are in the map. First posted on arXiv 16 Oct 2019.

Builds on

1 of 118 references.

References · 1

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2020: 7; 2021: 132; 2022: 137; 2023: 143; 2024: 97; 2025: 104; 2026: 79.

699 papers.

Sort:

By area:

  1. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 882
  2. 2023BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Datasetcited by 828
  3. 2021CLIPort: What and Where Pathways for Robotic Manipulationcited by 617
  4. 2023Eureka: Human-Level Reward Design via Coding Large Language Modelscited by 542
  5. 2023Augmented Language Models: a Surveycited by 398
  6. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 395
  7. 2020Generative Language Modeling for Automated Theorem Provingcited by 379
  8. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 375
  9. 2022DayDreamer: World Models for Physical Robot Learningcited by 372
  10. 2023Tool Learning with Foundation Modelscited by 302
  11. 2024Training Software Engineering Agents and Verifiers with SWE-Gymcited by 300
  12. 2022Behavior Transformers: Cloning k modes with one stonecited by 299
38 more
  1. 2021PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-trainingcited by 298
  2. 2024A Survey on Vision-Language-Action Models for Embodied AIcited by 274
  3. 2021Scaling Laws for Transfercited by 265
  4. 2022Masked Visual Pre-training for Motor Controlcited by 246
  5. 2023Orbit: A Unified Simulation Framework for Interactive Robot Learning Environmentscited by 239
  6. 2021MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scalecited by 226
  7. 2024DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learningcited by 210
  8. 2021Towards Causal Representation Learningcited by 207
  9. 2023RoboGen: Towards Unleashing Infinite Data for Automated Robot Learning via Generative Simulationcited by 206
  10. 2021DexMV: Imitation Learning for Dexterous Manipulation from Human Videoscited by 194
  11. 2022Will we run out of data? Limits of LLM scaling based on human-generated datacited by 187
  12. 2021Open-Ended Learning Leads to Generally Capable Agentscited by 177
  13. 2023Scaling Robot Learning with Semantically Imagined Experiencecited by 175
  14. 2021Behavior From the Void: Unsupervised Active Pre-Trainingcited by 173
  15. 2022Evolving Curricula with Regret-Based Environment Designcited by 166
  16. 2025VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learningcited by 166
  17. 2021A Survey of Zero-shot Generalisation in Deep Reinforcement Learningcited by 164
  18. 2021BEHAVIOR: Benchmark for Everyday Household Activities in Virtual, Interactive, and Ecological Environmentscited by 161
  19. 2022Compute Trends Across Three Eras of Machine Learningcited by 161
  20. 2024Understanding World or Predicting Future? A Comprehensive Survey of World Modelscited by 161
  21. 2023Foundation Models for Decision Making: Problems, Methods, and Opportunitiescited by 160
  22. 2021Policy Finetuning: Bridging Sample-Efficient Offline and Online Reinforcement Learningcited by 157
  23. 2022Human-to-Robot Imitation in the Wildcited by 157
  24. 2021A System for General In-Hand Object Re-Orientationcited by 155
  25. 2021Embodied Intelligence via Learning and Evolutioncited by 154
  26. 2021Collaborating with Humans without Human Datacited by 139
  27. 2021URLB: Unsupervised Reinforcement Learning Benchmarkcited by 138
  28. 2022DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Realitycited by 136
  29. 2023Human-Timescale Adaptation in an Open-Ended Task Spacecited by 136
  30. 2023Real-World Humanoid Locomotion with Reinforcement Learningcited by 132
  31. 2023FurnitureBench: Reproducible Real-World Benchmark for Long-Horizon Complex Manipulationcited by 132
  32. 2022Rapid Locomotion via Reinforcement Learningcited by 130
  33. 2022DexGraspNet: A Large-Scale Robotic Dexterous Grasp Dataset for General Objects Based on Simulationcited by 129
  34. 2022Goal Misgeneralization: Why Correct Specifications Aren't Enough For Correct Goalscited by 120
  35. 2023GenSim: Generating Robotic Simulation Tasks via Large Language Modelscited by 120
  36. 2023UniDexGrasp: Universal Robotic Dexterous Grasping via Learning Diverse Proposal Generation and Goal-Conditioned Policycited by 117
  37. 2022RAMBO-RL: Robust Adversarial Model-Based Offline Reinforcement Learningcited by 115
  38. 2022Towards Human-Level Bimanual Dexterous Manipulation with Reinforcement Learningcited by 113

50 most cited of 699 All 699 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Structured Task Solving via Modular Embodied Intelligence: A Case Study on Rubik's Cube
  2. 2026Memorize, Adapt, Ignore: Diagnosing Robot Learning Mechanisms under Training Data Variation
  3. 2020Grasping with Chopsticks: Combating Covariate Shift in Model-free Imitation Learning for Fine Manipulation
  4. 2026KinDER: A Physical Reasoning Benchmark for Robot Learning and Planning
  5. 2025HumanoidGen: Data Generation for Bimanual Dexterous Manipulation via LLM Reasoning
  6. 2024Screw Geometry Meets Bandits: Incremental Acquisition of Demonstrations to Generate Manipulation Plans
  7. 2024Manipulate-Anything: Automating Real-World Robots using Vision-Language Models
  8. 2026PlayWorld: Learning Robot World Models from Autonomous Play