mlsanityData to 14 Oct 2026

TAIL: Task-specific Adapters for Imitation Learning with Large Pretrained Models

arXiv 2310.05905 ↗First posted 9 Oct 2023cs.LG cs.AI cs.ROcited by 53

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

Abstract (arXiv)

The full potential of large pretrained models remains largely untapped in control domains like robotics. This is mainly because of the scarcity of data and the computational challenges associated with training or fine-tuning these large models for such applications. Prior work mainly emphasizes either effective pretraining of large models for decision-making or single-task adaptation. But real-world problems will require data-efficient, continual adaptation for new control tasks. Recognizing these constraints, we introduce TAIL (Task-specific Adapters for Imitation Learning), a framework for efficient adaptation to new control tasks. Inspired by recent advancements in parameter-efficient fine-tuning in language domains, we explore efficient fine-tuning techniques -- e.g., Bottleneck Adapters, P-Tuning, and Low-Rank Adaptation (LoRA) -- in TAIL to adapt large pretrained models for new tasks with limited demonstration data. Our extensive experiments in large-scale language-conditioned manipulation tasks comparing prevalent parameter-efficient fine-tuning techniques and adaptation baselines suggest that TAIL with LoRA can achieve the best post-adaptation performance with only 1% of the trainable parameters of full fine-tuning, while avoiding catastrophic forgetting and preserving adaptation plasticity in continual learning settings.

Key facts. Cited by 53 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 58 of its 76 references that are in the map. First posted on arXiv 9 Oct 2023.

Builds on

58 of 76 references.

Closest ancestors · 4

Same direction, newest first

  1. 2022VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Training30 Sep
  2. 2022Behavior Transformers: Cloning k modes with one stone22 Jun
  3. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning4 Feb
  4. 2021AW-Opt: Learning Robotic Skills with Imitation and Reinforcement at Scale9 Nov

Other references · 54

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  6. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  7. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,236
  8. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,915
  9. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  10. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  11. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  12. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,071
38 more
  1. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  2. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  3. 2021Ego4D: Around the World in 3,000 Hours of Egocentric Videocited by 1,556
  4. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,546
  5. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,410
  6. 2018Experience Replay for Continual Learningcited by 1,095
  7. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 898
  8. 2021Towards a Unified View of Parameter-Efficient Transfer Learningcited by 872
  9. 2022A Generalist Agentcited by 808
  10. 2020AdapterFusion: Non-Destructive Task Composition for Transfer Learningcited by 788
  11. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 739
  12. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 725
  13. 2022AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognitioncited by 719
  14. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 694
  15. 2020Intrinsic Dimensionality Explains the Effectiveness of Language Model Fine-Tuningcited by 671
  16. 2021CLIPort: What and Where Pathways for Robotic Manipulationcited by 622
  17. 2020AdapterHub: A Framework for Adapting Transformerscited by 444
  18. 2022VIMA: General Robot Manipulation with Multimodal Promptscited by 407
  19. 2021MT-Opt: Continuous Multi-Task Robotic Reinforcement Learning at Scalecited by 229
  20. 2023Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?cited by 218
  21. 2022Surgical Fine-Tuning Improves Adaptation to Distribution Shiftscited by 205
  22. 2023LIV: Language-Image Representations and Rewards for Robotic Controlcited by 188
  23. 2018Efficient parametrization of multi-domain deep neural networkscited by 168
  24. 2021UniPELT: A Unified Framework for Parameter-Efficient Language Model Tuningcited by 162
  25. 2022Prompting Decision Transformer for Few-Shot Policy Generalizationcited by 146
  26. 2021Actionable Models: Unsupervised Offline Reinforcement Learning of Robotic Skillscited by 125
  27. 2022Understanding and Preventing Capacity Loss in Reinforcement Learningcited by 125
  28. 2019Imitation Learning as f-Divergence Minimizationcited by 123
  29. 2023Constrained Decision Transformer for Offline Safe Reinforcement Learningcited by 69
  30. 2022Reincarnating Reinforcement Learning: Reusing Prior Computation to Accelerate Progresscited by 65
  31. 2019Meta-Q-Learningcited by 62
  32. 2022On Pre-Training for Visuo-Motor Control: Revisiting a Learning-from-Scratch Baselinecited by 56
  33. 2019DisCoRL: Continual Reinforcement Learning via Policy Distillationcited by 36
  34. 2021Context-Aware Safe Reinforcement Learning for Non-Stationary Environmentscited by 33
  35. 2023Hyper-Decision Transformer for Efficient Online Policy Adaptationcited by 33
  36. 2023Lossless Adaptation of Pretrained Vision Models For Robotic Manipulationcited by 29
  37. 2023Learning to Modulate pre-trained Models in RLcited by 28
  38. 2023SPRINT: Scalable Policy Pre-Training via Language Instruction Relabelingcited by 24

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 14; 2025: 18; 2026: 19.

53 papers.

Sort:

By area:

  1. 2024APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasetscited by 198
  2. 2025APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplaycited by 167
  3. 2025Improving Vision-Language-Action Model with Online Reinforcement Learningcited by 116
  4. 2024xLAM: A Family of Large Action Models to Empower AI Agent Systemscited by 111
  5. 2023Bootstrap Your Own Skills: Learning to Solve New Tasks with Large Language Model Guidancecited by 86
  6. 2026Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisonscited by 60
  7. 2024Sparse Diffusion Policy: A Sparse, Reusable, and Flexible Policy for Robot Learningcited by 57
  8. 2024Affordance-based Robot Manipulation with Flow Matchingcited by 54
  9. 2025Robust Finetuning of Vision-Language-Action Robot Policies via Parameter Mergingcited by 22
  10. 2025Dexterous Manipulation through Imitation Learning: A Surveycited by 21
  11. 2024Incremental Learning of Retrievable Skills For Efficient Continual Task Adaptationcited by 19
  12. 2025Reinforced Refinement with Self-Aware Expansion for End-to-End Autonomous Drivingcited by 19
38 more
  1. 2025Parallels Between VLA Model Post-Training and Human Motor Learning: Progress, Challenges, and Trendscited by 17
  2. 2026CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansioncited by 17
  3. 2026Simple Recipe Works: Vision-Language-Action Models are Natural Continual Learners with Reinforcement Learningcited by 17
  4. 2025Mechanistic Finetuning of Vision-Language-Action Models via Few-Shot Demonstrationscited by 15
  5. 2025Skill Expansion and Composition in Parameter Spacecited by 14
  6. 2025Think Small, Act Big: Primitive Prompt Learning for Lifelong Robot Manipulationcited by 14
  7. 2025Continually Evolving Skill Knowledge in Vision Language Action Modelcited by 14
  8. 2024EXTRACT: Efficient Policy Learning by Extracting Transferable Robot Skills from Offline Datacited by 13
  9. 2026Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuningcited by 13
  10. 2025SPECI: Skill Prompts based Hierarchical Continual Imitation Learning for Robot Manipulationcited by 9
  11. 2026CORAL: Scalable Multi-Task Robot Learning via LoRA Expertscited by 9
  12. 2025Behavior Injection: Preparing Language Models for Reinforcement Learningcited by 8
  13. 2025Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learningcited by 8
  14. 2023Scalable and Efficient Continual Learning from Demonstration via a Hypernetwork-generated Stable Dynamics Modelcited by 7
  15. 2024Bridging the Training-Inference Gap in LLMs by Leveraging Self-Generated Tokenscited by 7
  16. 2024Zero-shot Model-based Reinforcement Learning using Large Language Modelscited by 6
  17. 2025Efficient Continual Adaptation of Pretrained Robotic Policy with Online Meta-Learned Adapterscited by 6
  18. 2026World Action Models Enable Continual Imitation Learning with Recurrent Generative Replayscited by 4
  19. 2025Few-Shot Vision-Language Action-Incremental Policy Learningcited by 3
  20. 2025Factorizing Diffusion Policies for Observation Modality Prioritizationcited by 3
  21. 2025Policy Compatible Skill Incremental Learning via Lazy Learning Interfacecited by 3
  22. 2024Continual Robot Skill and Task Learning via Dialoguecited by 2
  23. 2026PHASER: Phase-Aware and Semantic Experience Replay for Vision-Language-Action Modelscited by 2
  24. 2024Task-agnostic Lifelong Robot Learning with Retrieval-based Weighted Local Adaptationcited by 1
  25. 2024Human-like Bots for Tactical Shooters Using Compute-Efficient Sensorscited by 1
  26. 2026A testable framework for AI alignment: Simulation Theology as an engineered worldview for silicon-based agentscited by 1
  27. 2026Lifelong Imitation Learning with Multimodal Latent Replay and Incremental Adjustmentcited by 1
  28. 2026Learning New Tasks via Reusable Skills: Skill-Compositional Experts for Embodied Continual Learningcited by 1
  29. 2026LiMoDE: Rethinking Lifelong Robot Manipulation from a Mixture-of-Dynamic-Experts Perspectivecited by 1
  30. 2024A Methodology-Oriented Study of Catastrophic Forgetting in Incremental Deep Neural Networkscited by 0
  31. 2024OU-CoViT: Copula-Enhanced Bi-Channel Multi-Task Vision Transformers with Dual Adaptation for OU-UWF Imagescited by 0
  32. 2024ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learningcited by 0
  33. 2025Efficient Morphology-Aware Policy Transfer to New Embodimentscited by 0
  34. 2026Revitalizing the Beginning: Avoiding Storage Dependency for Model Merging in Continual Learningcited by 0
  35. 2026Offline Multi-agent Continual Cooperation via Skill Partition and Reusecited by 0
  36. 2026Does Continual Imitation Learning Remain Grounded? A Language-Perturbed Benchmark for Robotic Task Retentioncited by 0
  37. 2026RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulationcited by 0
  38. 2026Dream to Remember: Continual Steering of Generalist Robot Policies with World Modelscited by 0

50 most cited of 53 All 53 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond the Majority: Long-tail Imitation Learning for Robotic Manipulation
  2. 2024Policy Adaptation via Language Optimization: Decomposing Tasks for Few-Shot Imitation
  3. 2020Scalable Multi-Task Imitation Learning with Autonomous Improvement
  4. 2022Learning and Retrieval from Prior Data for Skill-based Imitation Learning
  5. 2026The Gaussian Is Enough: Flow-Matching Priors Do Not Help When Fine-Tuning Large Behavior Models
  6. 2026TAIL-Safe: Task-Agnostic Safety Monitoring for Imitation Learning Policies
  7. 2024Towards Diverse Behaviors: A Benchmark for Imitation Learning with Human Demonstrations
  8. 2023Language-Conditioned Imitation Learning with Base Skill Priors under Unstructured Data