mlsanityData to 7 Oct 2026

Incremental Learning of Retrievable Skills For Efficient Continual Task Adaptation

arXiv 2410.22658 ↗First posted 30 Oct 2024cs.LG cs.AIcited by 19

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

Abstract (arXiv)

Continual Imitation Learning (CiL) involves extracting and accumulating task knowledge from demonstrations across multiple stages and tasks to achieve a multi-task policy. With recent advancements in foundation models, there has been a growing interest in adapter-based CiL approaches, where adapters are established parameter-efficiently for tasks newly demonstrated. While these approaches isolate parameters for specific tasks and tend to mitigate catastrophic forgetting, they limit knowledge sharing among different demonstrations. We introduce IsCiL, an adapter-based CiL framework that addresses this limitation of knowledge sharing by incrementally learning shareable skills from different demonstrations, thus enabling sample-efficient task adaptation using the skills particularly in non-stationary CiL environments. In IsCiL, demonstrations are mapped into the state embedding space, where proper skills can be retrieved upon input states through prototype-based memory. These retrievable skills are incrementally learned on their corresponding adapters. Our CiL experiments with complex tasks in Franka-Kitchen and Meta-World demonstrate robust performance of IsCiL in both task adaptation and sample-efficiency. We also show a simple extension of IsCiL for task unlearning scenarios.

Key facts. Cited by 19 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 36 of its 42 references that are in the map. First posted on arXiv 30 Oct 2024.

Builds on

36 of 42 references.

Closest ancestors · 6

Same direction, newest first

  1. 2023LOTUS: Continual Imitation Learning for Robot Manipulation Through Unsupervised Skill Discovery3 Nov
  2. 2023TAIL: Task-specific Adapters for Imitation Learning with Large Pretrained Models9 Oct
  3. 2023Data Quality in Imitation Learning4 Jun
  4. 2022Learning and Retrieval from Prior Data for Skill-based Imitation Learning20 Oct
  5. 2021CRIL: Continual Robot Imitation Learning via Generative and Prediction Model17 Jun
  6. 2019Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning25 Oct

Other references · 30

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  3. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  4. 2020D4RL: Datasets for Deep Data-Driven Reinforcement Learningcited by 1,406
  5. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  6. 2016Progressive Neural Networkscited by 1,332
  7. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  8. 2021Learning to Prompt for Continual Learningcited by 938
  9. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 882
  10. 2019ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Taskscited by 773
  11. 2022DualPrompt: Complementary Prompting for Rehearsal-free Continual Learningcited by 635
  12. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learningcited by 543
18 more
  1. 2022CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learningcited by 446
  2. 2019Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgettingcited by 302
  3. 2020Accelerating Reinforcement Learning with Learned Skill Priorscited by 220
  4. 2022Multi-Game Decision Transformerscited by 212
  5. 2019Continual Unsupervised Representation Learningcited by 147
  6. 2022Continual Learning and Private Unlearningcited by 124
  7. 2023A Unified Continual Learning Framework with General Parameter-Efficient Tuningcited by 124
  8. 2021Learning Language-Conditioned Robot Behavior from Offline Data and Crowd-Sourced Annotationcited by 117
  9. 2021Why Generalization in RL is Difficult: Epistemic POMDPs and Implicit Partial Observabilitycited by 102
  10. 2023Controllability-Aware Unsupervised Skill Discoverycited by 49
  11. 2021Hierarchical Few-Shot Imitation with Skill Transition Modelscited by 37
  12. 2022LISA: Learning Interpretable Skill Abstractions from Languagecited by 30
  13. 2023Learning to Modulate pre-trained Models in RLcited by 28
  14. 2023Continual Learning with Low Rank Adaptationcited by 27
  15. 2024Online Continual Learning For Interactive Instruction Following Agentscited by 21
  16. 2024OVOR: OnePrompt with Virtual Outlier Regularization for Rehearsal-Free Class-Incremental Learningcited by 20
  17. 2022Dynamics-Adaptive Continual Reinforcement Learning via Progressive Contextualizationcited by 13
  18. 2024One-shot Imitation in a Non-Stationary Environment via Multi-Modal Skillcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 10; 2026: 9.

19 papers.

Sort:

By area:

  1. 2025NeSyC: A Neuro-symbolic Continual Learner For Complex Embodied Tasks In Open Domainscited by 15
  2. 2026AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robotscited by 15
  3. 2026Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuningcited by 11
  4. 2025SPECI: Skill Prompts based Hierarchical Continual Imitation Learning for Robot Manipulationcited by 9
  5. 2025Continual Learning for Generative AI: From LLMs to MLLMs and Beyondcited by 9
  6. 2025When Continue Learning Meets Multimodal Large Language Model: A Surveycited by 8
  7. 2025Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learningcited by 8
  8. 2025Do Not Mimic My Voice: Speaker Identity Unlearning for Zero-Shot Text-to-Speechcited by 6
  9. 2025Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learningcited by 5
  10. 2026DreamProver: Evolving Transferable Lemma Libraries via a Wake-Sleep Theorem-Proving Agentcited by 4
  11. 2025In Praise of Stubbornness: An Empirical Case for Cognitive-Dissonance Aware Continual Update of Knowledge in LLMscited by 2
  12. 2025Policy Compatible Skill Incremental Learning via Lazy Learning Interfacecited by 2
7 more
  1. 2026Lifelong Imitation Learning with Multimodal Latent Replay and Incremental Adjustmentcited by 1
  2. 2026Learning New Tasks via Reusable Skills: Skill-Compositional Experts for Embodied Continual Learningcited by 1
  3. 2026LiMoDE: Rethinking Lifelong Robot Manipulation from a Mixture-of-Dynamic-Experts Perspectivecited by 1
  4. 2025Joint Flashback Adaptation for Forgetting-Resistant Instruction Tuningcited by 0
  5. 2026ContiGuard: A Framework for Continual Toxicity Detection Against Evolving Evasive Perturbationscited by 0
  6. 2026Efficient Skill Grounding via Code Refactoring with Small Language Modelscited by 0
  7. 2026Does Continual Imitation Learning Remain Grounded? A Language-Perturbed Benchmark for Robotic Task Retentioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025T2S: Tokenized Skill Scaling for Lifelong Imitation Learning
  2. 2026Benchmarking World Models for Continual Learning on Compositional Tasks
  3. 2020Scalable Multi-Task Imitation Learning with Autonomous Improvement
  4. 2025Learning a Thousand Tasks in a Day
  5. 2026Can Vision-Language-Action Models Learn from Real-World Data Continually without Forgetting?
  6. 2024Continual Task Learning through Adaptive Policy Self-Composition
  7. 2025Continually Evolving Skill Knowledge in Vision Language Action Model
  8. 2024M2Distill: Multi-Modal Distillation for Lifelong Imitation Learning