mlsanityData to 7 Oct 2026

Premier-TACO is a Few-Shot Policy Learner: Pretraining Multitask Representation via Temporal Action-Driven Contrastive Loss

arXiv 2402.06187 ↗First posted 9 Feb 2024cs.LG cs.AI cs.ROcited by 10

Not on the map. Why?

Abstract (arXiv)

We present Premier-TACO, a multitask feature representation learning approach designed to improve few-shot policy learning efficiency in sequential decision-making tasks. Premier-TACO leverages a subset of multitask offline datasets for pretraining a general feature representation, which captures critical environmental dynamics and is fine-tuned using minimal expert demonstrations. It advances the temporal action contrastive learning (TACO) objective, known for state-of-the-art results in visual control tasks, by incorporating a novel negative example sampling strategy. This strategy is crucial in significantly boosting TACO's computational efficiency, making large-scale multitask offline pretraining feasible. Our extensive empirical evaluation in a diverse set of continuous control benchmarks including Deepmind Control Suite, MetaWorld, and LIBERO demonstrate Premier-TACO's effectiveness in pretraining visual representations, significantly enhancing few-shot imitation learning of novel tasks. Our code, pretraining data, as well as pretrained model checkpoints will be released at https://github.com/PremierTACO/premier-taco. Our project webpage is at https://premiertaco.github.io.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 55 of its 69 references that are in the map. First posted on arXiv 9 Feb 2024.

Builds on

55 of 69 references.

References · 55

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  4. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  5. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  6. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  7. 2021Align before Fuse: Vision and Language Representation Learning with Momentum Distillationcited by 1,612
  8. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,598
  9. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,528
  10. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  11. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  12. 2021A Minimalist Approach to Offline Reinforcement Learningcited by 882
38 more
  1. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 745
  2. 2022R3M: A Universal Visual Representation for Robot Manipulationcited by 678
  3. 2019Data-Efficient Image Recognition with Contrastive Predictive Codingcited by 672
  4. 2020Contrastive Learning with Hard Negative Samplescited by 619
  5. 2016Training Region-based Object Detectors with Online Hard Example Miningcited by 577
  6. 2020Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixelscited by 573
  7. 2020Learning Invariant Representations for Reinforcement Learning without Reconstructioncited by 424
  8. 2020Hard Negative Mixing for Contrastive Learningcited by 399
  9. 2022VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Trainingcited by 384
  10. 2019Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learningcited by 380
  11. 2022Temporal Difference Learning for Model Predictive Controlcited by 375
  12. 2020Planning to Explore via Self-Supervised World Modelscited by 365
  13. 2021Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learningcited by 359
  14. 2020Data-Efficient Reinforcement Learning with Self-Predictive Representationscited by 309
  15. 2020Decoupling Representation Learning from Reinforcement Learningcited by 263
  16. 2019Libra R-CNN: Towards Balanced Learning for Object Detectioncited by 255
  17. 2023Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?cited by 215
  18. 2021Offline-to-Online Reinforcement Learning via Balanced Replay and Pessimistic Q-Ensemblecited by 211
  19. 2015Learning Multi-Domain Convolutional Neural Networks for Visual Trackingcited by 209
  20. 2021Reinforcement Learning with Prototypical Representationscited by 179
  21. 2022Masked World Models for Visual Controlcited by 169
  22. 2020OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learningcited by 164
  23. 2022The Unsurprising Effectiveness of Pre-Trained Vision Models for Controlcited by 155
  24. 2019Unsupervised State Representation Learning in Ataricited by 144
  25. 2021Discovering and Achieving Goals via World Modelscited by 131
  26. 2019Kinematic State Abstraction and Provably Efficient Rich-Observation Reinforcement Learningcited by 114
  27. 2021Pretraining Representations for Data-Efficient Reinforcement Learningcited by 105
  28. 2020Offline Meta-Reinforcement Learning with Advantage Weightingcited by 92
  29. 2020Deep Reinforcement and InfoMax Learningcited by 79
  30. 2022Pre-Trained Image Encoder for Generalizable Visual Reinforcement Learningcited by 58
  31. 2022Efficient Planning in a Compact Latent Action Spacecited by 57
  32. 2023TACO: Temporal Latent Action-Driven Contrastive Loss for Visual Reinforcement Learningcited by 56
  33. 2022On Pre-Training for Visuo-Motor Control: Revisiting a Learning-from-Scratch Baselinecited by 55
  34. 2023DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimizationcited by 45
  35. 2023SMART: Self-supervised Multi-task pretrAining with contRol Transformerscited by 44
  36. 2023Hyper-Decision Transformer for Efficient Online Policy Adaptationcited by 32
  37. 2023Is Model Ensemble Necessary? Model-based RL via a Single Model with Lipschitz Regularized Value Functioncited by 19
  38. 2022Transfer RL across Observation Feature Spaces via Model-Based Regularizationcited by 18

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 4; 2025: 5; 2026: 1.

10 papers.

  1. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
  2. 2025FLARE: Robot Learning with Implicit World Modelingcited by 105
  3. 2024Learning to Manipulate Anywhere: A Visual Generalizable Framework For Reinforcement Learningcited by 36
  4. 2024PRISE: LLM-Style Sequence Compression for Learning Temporal Action Abstractions in Controlcited by 21
  5. 2025TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment Videoscited by 13
  6. 2024Make-An-Agent: A Generalizable Policy Network Generator with Behavior-Prompted Diffusioncited by 12
  7. 2025TREND: Tri-teaching for Robust Preference-based Reinforcement Learning with Demonstrationscited by 7
  8. 2025Policy Compatible Skill Incremental Learning via Lazy Learning Interfacecited by 2
  9. 2026Saliency-Guided Representation with Consistency Policy Learning for Visual Unsupervised Reinforcement Learningcited by 1
  10. 2025Active Multimodal Distillation for Few-shot Action Recognitioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Learning to Drive by Watching YouTube Videos: Action-Conditioned Contrastive Policy Pretraining
  2. 2022Reinforcement Learning with Action-Free Pre-Training from Videos
  3. 2023Comparing the Efficacy of Fine-Tuning and Meta-Learning for Few-Shot Policy Imitation
  4. 2025ATK: Automatic Task-driven Keypoint Selection for Robust Policy Learning
  5. 2025Learning Long-Context Diffusion Policies via Past-Token Prediction
  6. 2025CDP: Towards Robust Autoregressive Visuomotor Policy Learning via Causal Diffusion
  7. 2023SpawnNet: Learning Generalizable Visuomotor Skills from Pre-trained Networks
  8. 2026Native Video-Action Pretraining for Generalizable Robot Control