mlsanityData to 14 Oct 2026

DynaGuide: Steering Diffusion Polices with Active Dynamic Guidance

arXiv 2506.13922 ↗First posted 16 Jun 2025cs.ROcited by 44

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

Abstract (arXiv)

Deploying large, complex policies in the real world requires the ability to steer them to fit the needs of a situation. Most common steering approaches, like goal-conditioning, require training the robot policy with a distribution of test-time objectives in mind. To overcome this limitation, we present DynaGuide, a steering method for diffusion policies using guidance from an external dynamics model during the diffusion denoising process. DynaGuide separates the dynamics model from the base policy, which gives it multiple advantages, including the ability to steer towards multiple objectives, enhance underrepresented base policy behaviors, and maintain robustness on low-quality objectives. The separate guidance signal also allows DynaGuide to work with off-the-shelf pretrained diffusion policies. We demonstrate the performance and features of DynaGuide against other steering approaches in a series of simulated and real experiments, showing an average steering success of 70% on a set of articulated CALVIN tasks and outperforming goal-conditioning by 5.4x when steered with low-quality objectives. We also successfully steer an off-the-shelf real robot policy to express preference for particular objects and even create novel behavior. Videos and more can be found on the project website: https://dynaguide.github.io

Key facts. Cited by 44 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 53 of its 59 references that are in the map. First posted on arXiv 16 Jun 2025.

Builds on

53 of 59 references.

Closest ancestors · 6

Same direction, newest first

  1. 2024Dynamics-Guided Diffusion Model for Sensor-less Robot Manipulator Design23 Feb
  2. 2023Zero-Shot Robotic Manipulation with Pretrained Image-Editing Diffusion Models16 Oct
  3. 2023Generative Skill Chaining: Long-Horizon Skill Planning with Diffusion Models13 Oct
  4. 2023Goal-Conditioned Imitation Learning using Score-based Diffusion Policies5 Apr
  5. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar
  6. 2022Is Conditional Generative Modeling all you need for Decision-Making?28 Nov

Other references · 47

  1. 2020Denoising Diffusion Implicit Modelscited by 9,781
  2. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  4. 2022Classifier-Free Diffusion Guidancecited by 5,847
  5. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  6. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,407
  7. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  8. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,653
  9. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  10. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  11. 2017Hindsight Experience Replaycited by 1,136
  12. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,061
35 more
  1. 2020Mastering Atari with Discrete World Modelscited by 1,007
  2. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 725
  3. 2021CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Taskscited by 584
  4. 2022BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learningcited by 523
  5. 2022DayDreamer: World Models for Physical Robot Learningcited by 400
  6. 2022Temporal Difference Learning for Model Predictive Controlcited by 397
  7. 2024Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robotscited by 394
  8. 2024DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planningcited by 386
  9. 2022Transformers are Sample-Efficient World Modelscited by 328
  10. 2019Learning Latent Plans from Playcited by 316
  11. 2016Deep Visual Foresight for Planning Robot Motioncited by 285
  12. 2018Visual Reinforcement Learning with Imagined Goalscited by 251
  13. 2019Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Modelcited by 231
  14. 2018Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Controlcited by 230
  15. 2025Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Modelscited by 221
  16. 2025Unified Video Action Modelcited by 205
  17. 2019Goal-conditioned Imitation Learningcited by 162
  18. 2024Data Scaling Laws in Imitation Learning for Robotic Manipulationcited by 145
  19. 2023Transformer-based World Models Are Happy With 100k Interactionscited by 139
  20. 2024ALOHA Unleashed: A Simple Recipe for Robot Dexteritycited by 134
  21. 2024Yell At Your Robot: Improving On-the-Fly from Language Correctionscited by 124
  22. 2018Zero-Shot Visual Imitationcited by 117
  23. 2024Steering Your Generalists: Improving Robotic Foundation Models via Value Guidancecited by 106
  24. 2020Offline Reinforcement Learning from Images with Latent Space Modelscited by 95
  25. 2025Inference-Time Alignment in Diffusion Models with Reward-Guided Generation: Tutorial and Reviewcited by 88
  26. 2023"No, to the Right" -- Online Language Corrections for Robotic Manipulation via Shared Autonomycited by 60
  27. 2024Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Samplingcited by 52
  28. 2025From Foresight to Forethought: VLM-In-the-Loop Policy Steering via Latent Alignmentcited by 52
  29. 2025Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysiscited by 45
  30. 2024Inference-Time Policy Steering through Human Interactionscited by 38
  31. 2024Multi-Task Interactive Robot Fleet Learning with Visual World Modelscited by 36
  32. 2024CLIP-RT: Learning Language-Conditioned Robotic Policies from Natural Language Supervisioncited by 16
  33. 2025LUMOS: Language-Conditioned Imitation Learning with World Modelscited by 14
  34. 2024To Err is Robotic: Rapid Value-Based Trial-and-Error during Deploymentcited by 7
  35. 2023Learning Sparse Control Tasks from Pixels by Latent Nearest-Neighbor-Guided Explorationscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 40.

44 papers.

Sort:

By area:

  1. 2026DreamDojo: A Generalist Robot World Model from Large-Scale Human Videoscited by 135
  2. 2026PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulationcited by 82
  3. 2026Do World Action Models Generalize Better than VLAs? A Robustness Studycited by 36
  4. 2025Seeing to Act, Prompting to Specify: A Bayesian Factorization of Vision Language Action Policycited by 18
  5. 2026VLS: Steering Pretrained Robot Policies via Vision-Language Modelscited by 15
  6. 2025MoE-DP: An MoE-Enhanced Diffusion Policy for Robust Long-Horizon Robotic Manipulation with Skill Decomposition and Failure Recoverycited by 10
  7. 2026RFS: Reinforcement Learning with Residual Flow Steering for Dexterous Manipulationcited by 10
  8. 2025Compose Your Policies! Improving Diffusion-based or Flow-based Robot Policies via Test-time Distribution-level Compositioncited by 9
  9. 2026Feedback World Model Enables Precise Guidance of Diffusion Policycited by 9
  10. 2026OmniGuide: Universal Guidance Fields for Enhancing Generalist Robot Policiescited by 6
  11. 2026Breaking Lock-In: Preserving Steerability under Low-Data VLA Post-Trainingcited by 6
  12. 2026You've Got a Golden Ticket: Improving Generative Robot Policies With A Single Noise Vectorcited by 5
32 more
  1. 2026Seeing Farther and Smarter: Value-Guided Multi-Path Reflection for VLM Policy Optimizationcited by 4
  2. 2026Spatial Memory for Out-of-Vision Manipulation in Vision-Language-Actioncited by 4
  3. 2025EVE: A Generator-Verifier System for Generative Policiescited by 3
  4. 2026Dreaming the Unseen: World Model-regularized Diffusion Policy for Out-of-Distribution Robustnesscited by 3
  5. 2026QPILOTS: Efficient Test-Time Q-Steering for Flow Policiescited by 3
  6. 2026Compositional Visual Planning via Inference-Time Diffusion Scalingcited by 2
  7. 2026PPGuide: Steering Diffusion Policies with Performance Predictive Guidancecited by 2
  8. 2026Lifting Embodied World Models for Planning and Controlcited by 2
  9. 2026Beyond Action Residuals: Real-World Robot Policy Steering via Bottleneck Latent Reinforcement Learningcited by 2
  10. 2026Reinforcement Learning for Flow-Matching Policies with Density Transportcited by 2
  11. 2026Robot Self-Improvement via Human-Video Dynamics Modelscited by 2
  12. 2026Grounding Generative Policies in Physics: Optimization-Guided Diffusion for Robot Controlcited by 2
  13. 2026DEFLECT: Temporal Counterfactual Preference Learning for Delay-Robust Asynchronous VLAscited by 1
  14. 2026SAPS: Shared Autonomy for Policy Steering by Blending Teleoperation with a Pretrained VLAcited by 1
  15. 2026Behavior Uncloning: Distilling Mode Redirection into Policy Weights without Inference-Time Steeringcited by 1
  16. 2026Beyond Noise Steering: Dual-Latent Space Reinforcement Learning for Generative Robot Policycited by 1
  17. 2026TraceFlow: Guiding Frozen Flow-Matching Robot Policies with Success and Failure Tracescited by 1
  18. 2026Conflict-Aware Additive Guidance for Flow Models under Compositional Rewardscited by 0
  19. 2026Training-Free Imitation Learning with Closed-Form Diffusion Policiescited by 0
  20. 2026Flow Control: Steering Vision-Language-Action Models with Simple Real-Time Inputscited by 0
  21. 2026Steering Autoregressive Vision-Language-Action Policies via Action Token Interventioncited by 0
  22. 2026ReGuide: From Test-Time Guidance to Self-Improving Diffusion Policiescited by 0
  23. 2026Optimal Transport Q-Learning for Flow Policy Steering and Accelerationcited by 0
  24. 2026Source-Lifted Flow Matching for Intervenable Multimodal Imitationcited by 0
  25. 2026GRACE: Gradient-Free Robot Action Generation via Combined Diffusion-MPPI Posterior Mean Estimationcited by 0
  26. 2026Skills in Weights, Memory in Code: Hybrid Learning for Memory-Dependent Robot Manipulationcited by 0
  27. 2026Zero-Shot Reactive Obstacle Avoidance for Generative Robot Policiescited by 0
  28. 2026Correcting WHERE, Preserving HOW: Compositional Generalization for Vision-Language-Action Models via Referential Guidancecited by 0
  29. 2026Cue the Flow: Steering Flow-Matching Policies for Open-World Delivery Manipulationcited by 0
  30. 2026Robot Learning with Visual Predicted Forcecited by 0
  31. 2026Encoded but Not in Control: Revealing the Grounding Gap in Vision-Language Robot Policiescited by 0
  32. 2026Steertuning Policies from a Few Human Videoscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Diffusion Policy Policy Optimization
  2. 2025DDAT: Diffusion Policies Enforcing Dynamically Admissible Robot Trajectories
  3. 2024One-Step Diffusion Policy: Fast Visuomotor Policies via Diffusion Distillation
  4. 2025D3P: Dynamic Denoising Diffusion Policy via Reinforcement Learning
  5. 2025Real-Time Iteration Scheme for Diffusion Policy
  6. 2026Fisher-Preserving Guidance: Training-Free Manifold Constraints for Safe Diffusion Control
  7. 2025Constraint-Aware Diffusion Guidance for Robotics: Real-Time Obstacle Avoidance for Autonomous Racing
  8. 2025RA-DP: Rapid Adaptive Diffusion Policy for Training-Free High-frequency Robotics Replanning