mlsanityData to 14 Oct 2026

EgoBridge: Domain Adaptation for Generalizable Imitation from Egocentric Human Data

arXiv 2509.19626 ↗First posted 23 Sep 2025cs.RO cs.CV cs.LGcited by 29

On the map: Offline RL and imitation learning from data → Robot manipulation imitation learning

Abstract (arXiv)

Egocentric human experience data presents a vast resource for scaling up end-to-end imitation learning for robotic manipulation. However, significant domain gaps in visual appearance, sensor modalities, and kinematics between human and robot impede knowledge transfer. This paper presents EgoBridge, a unified co-training framework that explicitly aligns the policy latent spaces between human and robot data using domain adaptation. Through a measure of discrepancy on the joint policy latent features and actions based on Optimal Transport (OT), we learn observation representations that not only align between the human and robot domain but also preserve the action-relevant information critical for policy learning. EgoBridge achieves a significant absolute policy success rate improvement by 44% over human-augmented cross-embodiment baselines in three real-world single-arm and bimanual manipulation tasks. EgoBridge also generalizes to new objects, scenes, and tasks seen only in human data, where baselines fail entirely. Videos and additional information can be found at https://ego-bridge.github.io

Key facts. Cited by 29 papers in the map (data through 14 Oct 2026). Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 31 of its 39 references that are in the map. First posted on arXiv 23 Sep 2025.

Builds on

31 of 39 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Empirical Analysis of Sim-and-Real Cotraining of Diffusion Policies for Planar Pushing from Pixels28 Mar
  2. 2025Motion Tracks: A Unified Representation for Human-Robot Transfer in Few-Shot Imitation Learning13 Jan
  3. 2024EgoMimic: Scaling Imitation Learning via Egocentric Video31 Oct
  4. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware23 Apr
  5. 2023MimicPlay: Long-Horizon Imitation Learning by Watching Human Play24 Feb

Other references · 26

  1. 2020End-to-End Object Detection with Transformerscited by 7,414
  2. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  3. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
  4. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  5. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,407
  6. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  7. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  8. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,761
  9. 2017Adversarial Discriminative Domain Adaptationcited by 1,311
  10. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,225
  11. 2024DROID: A Large-Scale In-The-Wild Robot Manipulation Datasetcited by 852
  12. 2016Deep Transfer Learning with Joint Adaptation Networkscited by 612
14 more
  1. 2023CoTracker: It is Better to Track Togethercited by 522
  2. 2015Optimal Transport for Domain Adaptationcited by 463
  3. 2024Latent Action Pretraining from Videoscited by 321
  4. 2017Joint Distribution Optimal Transportation for Domain Adaptationcited by 282
  5. 2018DeepJDOT: Deep Joint Distribution Optimal Transport for Unsupervised Domain Adaptationcited by 214
  6. 2023Project Aria: A New Tool for Egocentric Multi-Modal AI Researchcited by 208
  7. 2024HumanPlus: Humanoid Shadowing and Imitation from Humanscited by 185
  8. 2022Human-to-Robot Imitation in the Wildcited by 161
  9. 2024Scaling Proprioceptive-Visual Learning with Heterogeneous Pre-trained Transformerscited by 148
  10. 2024Track2Act: Predicting Point Tracks from Internet Videos enables Generalizable Robot Manipulationcited by 124
  11. 2019AVID: Learning Multi-Stage Tasks via Pixel-Level Translation of Human Videoscited by 105
  12. 2025Humanoid Policy ~ Human Policycited by 78
  13. 2025Phantom: Training Robots Without Robots Using Only Human Videoscited by 70
  14. 2024Natural Language Can Help Bridge the Sim2Real Gapcited by 21

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 26.

29 papers.

Sort:

By area:

  1. 2026Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalizationcited by 66
  2. 2026EgoVerse: An Egocentric Human Dataset for Robot Learning from Around the Worldcited by 57
  3. 2025Emergence of Human to Robot Transfer in Vision-Language-Action Modelscited by 42
  4. 2025In-N-On: Scaling Egocentric Manipulation with in-the-wild and on-task Datacited by 26
  5. 2025ActiveUMI: Robotic Manipulation with Active Perception from Robot-Free Human Demonstrationscited by 19
  6. 2026RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulationcited by 19
  7. 2026HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videoscited by 16
  8. 2026HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulationcited by 10
  9. 2026Cross-Hand Latent Representation for Vision-Language-Action Modelscited by 7
  10. 2026Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Trainingcited by 6
  11. 2026Learning Dexterous Manipulation Using Contact Wrench Guidance From Human Demonstrationcited by 6
  12. 2026World-Action Models for Robot Learning and Control: A Surveycited by 6
17 more
  1. 2026A Mechanistic Analysis of Sim-and-Real Co-Training in Generative Robot Policiescited by 5
  2. 2026Retrieve, Don't Retrain: Extending Vision Language Action Models to New Tasks at Test Timecited by 5
  3. 2026Ambient Diffusion Policy: Imitation Learning from Suboptimal Data in Roboticscited by 4
  4. 2026ACE-Data-0: Human-Centric Ambient Capture as Embodied Data Enginecited by 3
  5. 2026SiMDex: Mining Similar Egocentric Videos for Cross-Embodiment Dexterous Manipulationcited by 3
  6. 2026What Matters When Cotraining Robot Manipulation Policies on Everyday Human Videos?cited by 2
  7. 2026Translation as a Bridging Action: Transferring Manipulation Skills from Humans to Robotscited by 1
  8. 2026Sim-and-Human Co-training for Data-Efficient and Scene-Generalizable Bimanual Manipulationcited by 0
  9. 2026Human2Any: Human-to-Robot Transfer via Constraint-Aware Compositional Planningcited by 0
  10. 2026EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstrationcited by 0
  11. 2026IM-ENGINE: Image Editing for Embodied Data Generationcited by 0
  12. 2026ActiveScale: Scaling Active Perception for Robots across Model, Data, and Hardwarecited by 0
  13. 2026CE4L: Continual Ego, Exo, and Ego-Exo Learningcited by 0
  14. 2026EgoSpeedUp: Transferring Human Manipulation Tempo to Robot Policiescited by 0
  15. 2026EgoHumanoid-V2: Human-to-Humanoid Transfer of Coordinated Whole-Body Skills for Loco-Manipulationcited by 0
  16. 2026EgoLAP: Learning from Egocentric Human Data through Language-Action Reasoningcited by 0
  17. 2026Learning Transferable World-Action Models from Task-Paired Human Videoscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Ego2Robot: Scalable Robot Data Synthesis from Egocentric Human Data
  2. 2026EgoScale: Scaling Dexterous Manipulation with Diverse Egocentric Human Data
  3. 2025EgoMe: A New Dataset and Challenge for Following Me via Egocentric View in Real World
  4. 2025EgoMI: Learning Active Vision and Whole-Body Manipulation from Egocentric Human Demonstrations
  5. 2025EgoDemoGen: Egocentric Demonstration Generation for Viewpoint Generalization in Robotic Manipulation
  6. 2026EgoSteer: An Open-Source Full-Stack System Towards Steerable Dexterous Manipulation from Egocentric Videos
  7. 2026EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Data
  8. 2026UMI-Bridge: Action-Anchored Latent Alignment across Human and Robot Manipulation Data ↗