mlsanityData to 14 Oct 2026

WarmPrior: Straightening Flow-Matching Policies with Temporal Priors

arXiv 2605.13959 ↗First posted 13 May 2026cs.LG cs.AI cs.ROcited by 4

On the map: Diffusion models → Diffusion and flow policies for robotics

Abstract (arXiv)

Generative policies based on diffusion and flow matching have become a dominant paradigm for visuomotor robotic control. We show that replacing the standard Gaussian source distribution with WarmPrior, a simple temporally grounded prior constructed from readily available recent action history, consistently improves success rates on robotic manipulation tasks. We trace this gain to markedly straighter probability paths, echoing the effect of optimal-transport couplings in Rectified Flow. Beyond standard behavior cloning, WarmPrior also reshapes the exploration distribution in prior-space reinforcement learning, improving both sample efficiency and final performance. Collectively, these results identify the source distribution as an important and underexplored design axis in generative robot control.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 39 of its 42 references that are in the map. First posted on arXiv 13 May 2026.

Builds on

39 of 42 references.

Closest ancestors · 10

Same direction, newest first

  1. 2026STEP: Warm-Started Visuomotor Policies with Spatiotemporal Consistency Prediction9 Feb
  2. 2026Action-to-Action Flow Matching7 Feb
  3. 2025Steering Your Diffusion Policy with Latent Space Reinforcement Learning18 Jun
  4. 2024One-Step Diffusion Policy: Fast Visuomotor Policies via Diffusion Distillation28 Oct
  5. 2024Learning Robotic Manipulation Policies from Point Clouds with Conditional Flow Matching11 Sep
  6. 2024ManiCM: Real-time 3D Diffusion Policy via Consistency Model for Robotic Manipulation3 Jun
  7. 2024Consistency Policy: Accelerated Visuomotor Policies via Consistency Distillation13 May
  8. 2024Riemannian Flow Matching Policy for Robot Motion Learning15 Mar
  9. 2024Don't Start from Scratch: Behavioral Refinement via Interpolant-based Policy Diffusion25 Feb
  10. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar

Other references · 29

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2020Denoising Diffusion Implicit Modelscited by 9,781
  5. 2025Qwen3 Technical Reportcited by 8,820
  6. 2022Scalable Diffusion Models with Transformerscited by 6,054
  7. 2022Flow Matching for Generative Modelingcited by 5,702
  8. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  9. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  10. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  11. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,407
  12. 2025π0.5: a Vision-Language-Action Model with Open-World Generalizationcited by 1,761
17 more
  1. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,568
  2. 2018Group Normalizationcited by 1,524
  3. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,225
  4. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,061
  5. 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 939
  6. 2023Improving and generalizing flow-based generative models with minibatch optimal transportcited by 879
  7. 2024DROID: A Large-Scale In-The-Wild Robot Manipulation Datasetcited by 852
  8. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 725
  9. 2023MimicGen: A Data Generation System for Scalable Robot Learning using Human Demonstrationscited by 327
  10. 2023Multisample Flow Matching: Straightening Flows with Minibatch Couplingscited by 278
  11. 2025Real-Time Execution of Action Chunking Flow Policiescited by 216
  12. 2023Diffusion Schr"odinger Bridge Matchingcited by 153
  13. 2025A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulationcited by 129
  14. 2023Simulation-free Schr"odinger bridges via score and flow matchingcited by 83
  15. 2025Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Modelscited by 64
  16. 2024AdaFlow: Imitation Learning with Variance-Adaptive Flow-Based Policiescited by 50
  17. 2025HAMLET: Switch your Vision-Language-Action Model into a History-Aware Policycited by 46

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026Frequency-Conditioned Flow Matching for Vision-Language-Action Modelscited by 0
  2. 2026Object-Centric Conditioning for Visuomotor Flow Matchingcited by 0
  3. 2026Optimizing H-Graph Hybridization for Diffusion-Guided RRTcited by 0
  4. 2026Toward Real-Time VLAs: Stage-Aware Two-Step Flow Denoising and System-Level Evaluationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Where Should Action Generation Begin? A Learnable Source Prior for Generative Robot Policies
  2. 2026Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control
  3. 2026Temporal Policy: History-Initialized Action Generation for Robotic Learning from Demonstration
  4. 2025Much Ado About Noising: Dispelling the Myths of Generative Robotic Control
  5. 2025Generative Predictive Control: Flow Matching Policies for Dynamic and Difficult-to-Demonstrate Tasks
  6. 2026FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space
  7. 2026One-Step Flow Policy: Self-Distillation for Fast Visuomotor Policies
  8. 2026FLASH: Efficient Visuomotor Policy via Sparse Sampling