mlsanityData to 7 Oct 2026

Imitation Learning from a Single Temporally Misaligned Video

arXiv 2502.05397 ↗First posted 8 Feb 2025cs.LGcited by 7

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

Abstract (arXiv)

We examine the problem of learning sequential tasks from a single visual demonstration. A key challenge arises when demonstrations are temporally misaligned due to variations in timing, differences in embodiment, or inconsistencies in execution. Existing approaches treat imitation as a distribution-matching problem, aligning individual frames between the agent and the demonstration. However, we show that such frame-level matching fails to enforce temporal ordering or ensure consistent progress. Our key insight is that matching should instead be defined at the level of sequences. We propose that perfect matching occurs when one sequence successfully covers all the subgoals in the same order as the other sequence. We present ORCA (ORdered Coverage Alignment), a dense per-timestep reward function that measures the probability of the agent covering demonstration frames in the correct order. On temporally misaligned demonstrations, we show that agents trained with the ORCA reward achieve 4.5x improvement (0.11 rightarrow 0.50 average normalized returns) for Meta-world tasks and 6.6x improvement (6.55 rightarrow 43.3 average returns) for Humanoid-v4 tasks compared to the best frame-level matching algorithms. We also provide empirical analysis showing that ORCA is robust to varying levels of temporal misalignment. Our code is available at https://github.com/portal-cornell/orca/

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 34 of its 42 references that are in the map. First posted on arXiv 8 Feb 2025.

Builds on

34 of 42 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Keypoint Action Tokens Enable In-Context Imitation Learning in Robotics28 Mar
  2. 2023Teach a Robot to FISH: Versatile Imitation from One Minute of Demonstrations2 Mar

Other references · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  3. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  4. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
  5. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  6. 2018Computational Optimal Transportcited by 765
  7. 2017Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classificationcited by 521
  8. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 359
  9. 2017One-Shot Imitation Learningcited by 232
  10. 2018Generative Adversarial Imitation from Observationcited by 183
  11. 2023LIV: Language-Image Representations and Rewards for Robotic Controlcited by 180
  12. 2023Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learningcited by 155
20 more
  1. 2022Prompting Decision Transformer for Few-Shot Policy Generalizationcited by 143
  2. 2023RoboCLIP: One Demonstration is Enough to Learn Robot Policiescited by 128
  3. 2020Primal Wasserstein Imitation Learningcited by 120
  4. 2021Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gapcited by 80
  5. 2019Provably Efficient Imitation Learning from Observation Alonecited by 78
  6. 2022Watch and Match: Supercharging Imitation with Regularized Optimal Transportcited by 64
  7. 2019Wasserstein Adversarial Imitation Learningcited by 59
  8. 2024Vid2Robot: End-to-end Video-conditioned Policy Learning with Cross-Attention Transformerscited by 47
  9. 2023Optimal Transport for Offline Imitation Learningcited by 42
  10. 2023Fast Traversability Estimation for Wild Visual Navigationcited by 41
  11. 2023See to Touch: Learning Tactile Dexterity through Visual Incentivescited by 33
  12. 2020Imitation Learning with Sinkhorn Distancescited by 30
  13. 2024Rank2Reward: Learning Shaped Reward Functions from Passive Videocited by 30
  14. 2024FuRL: Visual-Language Models as Fuzzy Rewards for Reinforcement Learningcited by 26
  15. 2023For Pre-Trained Vision Models in Motor Control, Not All Policy Learning Methods are Created Equalcited by 22
  16. 2023What Matters to You? Towards Visual Representation Alignment for Robot Learningcited by 14
  17. 2024One-Shot Imitation under Mismatched Executioncited by 11
  18. 2023Imitation Learning from Observation with Automatic Discount Schedulingcited by 10
  19. 2024Robot Policy Learning with Temporal Optimal Transport Rewardcited by 8
  20. 2024Align Your Intents: Offline Imitation Learning via Optimal Transportcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 4.

7 papers.

  1. 2025X-Diffusion: Training Diffusion Policies on Cross-Embodiment Human Demonstrationscited by 6
  2. 2025TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distancecited by 5
  3. 2025Minimal Ingredients for Reward Assignment from Expert Demonstrationscited by 0
  4. 2026Cross-Domain Demo-to-Code via Neurosymbolic Counterfactual Reasoningcited by 0
  5. 2026FocalPolicy: Frequency-Optimized Chunking and Locally Anchored Flow Matching for Coherent Visuomotor Policycited by 0
  6. 2026Support-Constrained RL Enables Real-World Policy Improvement without Real-World Experiencecited by 0
  7. 2026Rank-Then-Act: Reward-Free Control from Frame-Order Progresscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Perfect Demo Makes Poor Teacher: Learning Robust Alignment from Critical Motion Segments
  2. 2021XIRL: Cross-embodiment Inverse Reinforcement Learning
  3. 2026Task-Centric Policy Optimization from Misaligned Motion Priors
  4. 2026Difference-Aware Retrieval Policies for Imitation Learning
  5. 2022Eliciting Compatible Demonstrations for Multi-Human Imitation Learning
  6. 2025Long-Horizon Visual Imitation Learning via Plan and Code Reflection
  7. 2021The Surprising Effectiveness of Representation Learning for Visual Imitation
  8. 2024Learning from 10 Demos: Generalisable and Sample-Efficient Policy Learning with Oriented Affordance Frames