mlsanityData to 14 Oct 2026

Robust Offline Imitation Learning Through State-level Trajectory Stitching

arXiv 2503.22524 ↗First posted 28 Mar 2025cs.RO cs.AIcited by 1

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Imitation learning (IL) has proven effective for enabling robots to acquire visuomotor skills through expert demonstrations. However, traditional IL methods are limited by their reliance on high-quality, often scarce, expert data, and suffer from covariate shift. To address these challenges, recent advances in offline IL have incorporated suboptimal, unlabeled datasets into the training. In this paper, we propose a novel approach to enhance policy learning from mixed-quality offline datasets by leveraging task-relevant trajectory fragments and rich environmental dynamics. Specifically, we introduce a state-based search framework that stitches state-action pairs from imperfect demonstrations, generating more diverse and informative training trajectories. Experimental results on standard IL benchmarks and real-world robotic tasks showcase that our proposed method significantly improves both generalization and performance.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 29 of its 40 references that are in the map. First posted on arXiv 28 Mar 2025.

Builds on

29 of 40 references.

Closest ancestors · 7

Same direction, newest first

  1. 2024Offline Imitation Learning with Model-based Reverse Augmentation18 Jun
  2. 2024How to Leverage Diverse Demonstrations in Offline Imitation Learning24 May
  3. 2024SEABO: A Simple Search-Based Method for Offline Imitation Learning6 Feb
  4. 2022Discriminator-Weighted Offline Imitation Learning from Suboptimal Demonstrations20 Jul
  5. 2022Discriminator-Guided Model-Based Offline Imitation Learning1 Jul
  6. 2022Versatile Offline Imitation from Observations and Examples via Regularized State-Occupancy Matching4 Feb
  7. 2020Toward the Fundamental Limits of Imitation Learning13 Sep

Other references · 22

  1. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,838
  2. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,920
  3. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,809
  4. 2016Generative Adversarial Imitation Learningcited by 1,597
  5. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,568
  6. 2024Octo: An Open-Source Generalist Robot Policycited by 1,484
  7. 2020D4RL: Datasets for Deep Data-Driven Reinforcement Learningcited by 1,427
  8. 2021Offline Reinforcement Learning with Implicit Q-Learningcited by 1,293
  9. 2023Mastering Diverse Domains through World Modelscited by 1,253
  10. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 694
  11. 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 512
  12. 2023Vision-Language Foundation Models as Effective Robot Imitatorscited by 349
10 more
  1. 2023RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunkingcited by 169
  2. 2023STORM: Efficient Stochastic Transformer based World Models for Reinforcement Learningcited by 89
  3. 2021Q-attention: Enabling Efficient Learning for Vision-based Robotic Manipulationcited by 85
  4. 2024BAKU: An Efficient Transformer for Multi-Task Policy Learningcited by 80
  5. 2023Data Quality in Imitation Learningcited by 75
  6. 2022How to Leverage Unlabeled Data in Offline Reinforcement Learningcited by 62
  7. 2023Behavior Retrieval: Few-Shot Imitation Learning by Querying Unlabeled Datasetscited by 58
  8. 2023Optimal Transport for Offline Imitation Learningcited by 42
  9. 2023Universal Visual Decomposer: Long-Horizon Manipulation Made Easycited by 25
  10. 2024Offline Imitation Learning Through Graph Search and Retrievalcited by 14

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026NEEDLEWORK: Offline Rewriting of Robot Data with Verified Local Stitchescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Offline Imitation Learning upon Arbitrary Demonstrations by Pre-Training Dynamics Representations
  2. 2025Using Non-Expert Data to Robustify Imitation Learning via Offline Reinforcement Learning
  3. 2021Mitigating Covariate Shift in Imitation Learning via Offline Data Without Great Coverage
  4. 2026Imitation from Observations with Trajectory-Level Generative Embeddings
  5. 2023A Simple Solution for Offline Imitation from Observations and Examples with Possibly Incomplete Trajectories
  6. 2023GO-DICE: Goal-Conditioned Option-Aware Offline Imitation Learning via Stationary Distribution Correction Estimation
  7. 2022Learning and Retrieval from Prior Data for Skill-based Imitation Learning
  8. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulation