mlsanityData to 7 Oct 2026

RAPAC-DP: Response-Aligned Pending-Action Compensation for Diffusion Policies under Delayed Execution

arXiv 2608.15924 ↗First posted 16 Aug 2026cs.ROcited by 0

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

Abstract (arXiv)

Cloud-side inference gives imitation-learning policies access to greater computational resources, but communication and computation delays can degrade control performance. To compensate for these delays, we propose RAPAC-DP, a response-aligned pending-action compensation framework designed for both diffusion- and flow-based action generators. RAPAC-DP encodes the actions already scheduled for execution before the cloud response arrives into a pending-action sequence that serves as the conditioning input to a parameter-efficient compensation pathway. When delay effects are negligible, bypassing this pathway exactly recovers the frozen base policy. For training, RAPAC-DP constructs delay-conditioned samples from delay-free demonstrations, requiring neither explicit system dynamics nor additional delayed demonstrations. At the largest fixed delay tested on Kinetix, RAPAC-DP retained 81.4% of its overall delay-free performance. At the largest fixed delay tested on each RoboMimic task, it achieved a mean success rate of 0.633 across the three tasks. These results demonstrate the effectiveness of pending-action compensation for cloud-deployed imitation-learning policies.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 33 of its 36 references that are in the map. First posted on arXiv 16 Aug 2026.

Builds on

33 of 36 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024FlowPolicy: Enabling Fast and Robust 3D Flow-based Policy via Consistency Flow Matching for Robot Manipulation6 Dec
  2. 2024Consistency Policy: Accelerated Visuomotor Policies via Consistency Distillation13 May
  3. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusion7 Mar
  4. 2022Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning12 Aug

Other references · 29

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  3. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  4. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 2,993
  5. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,777
  6. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,037
  7. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,007
  8. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  9. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  10. 2021What Matters in Learning from Offline Human Demonstrations for Robot Manipulationcited by 699
  11. 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Roboticscited by 507
  12. 2022Behavior Transformers: Cloning k modes with one stonecited by 299
17 more
  1. 2025Real-Time Execution of Action Chunking Flow Policiescited by 199
  2. 2020Reinforcement Learning with Random Delayscited by 56
  3. 2025VLASH: Real-Time VLAs via Future-State-Aware Asynchronous Inferencecited by 54
  4. 2025Leave No Observation Behind: Real-time Correction for VLA Action Chunkscited by 45
  5. 2024Kinetix: Investigating the Training of General Agents through Open-Ended Physics-Based Control Taskscited by 40
  6. 2020Delay-Aware Model-Based Reinforcement Learning for Continuous Controlcited by 33
  7. 2021Revisiting State Augmentation methods for Reinforcement Learning with Stochastic Delayscited by 20
  8. 2022Asynchronous Reinforcement Learning for Real-Time Control of Physical Robotscited by 19
  9. 2025VLA-RAIL: A Real-Time Asynchronous Inference Linker for VLA Models and Robotscited by 18
  10. 2019Network Offloading Policies for Cloud Robotics: a Learning-based Approachcited by 17
  11. 2026Real-Time Robot Execution with Masked Action Chunkingcited by 16
  12. 2024Reinforcement Learning from Delayed Observations via World Modelscited by 13
  13. 2025Handling Delay in Real-Time Reinforcement Learningcited by 7
  14. 2022FogROS2: An Adaptive Platform for Cloud and Fog Robotics Using ROS 2cited by 6
  15. 2024DEER: A Delay-Resilient Framework for Reinforcement Learning with Variable Delayscited by 6
  16. 2026Adaptive Capacity Allocation for Vision Language Action Fine-tuningcited by 6
  17. 2024FogROS2-PLR: Probabilistic Latency-Reliability For Cloud Roboticscited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Delay-Aware Diffusion Policy: Bridging the Observation-Execution Gap in Dynamic Tasks
  2. 2024Streaming Diffusion Policy: Fast Policy Synthesis with Variable Noise Diffusion Models
  3. 2026JEPA Policy: Diffusion-Free Imitation Learning via Paired Action and Future Representation Prediction
  4. 2025Masked Generative Policy for Robotic Control
  5. 2025Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling
  6. 2025Improving Generative Behavior Cloning via Self-Guidance and Adaptive Chunking
  7. 2026CoLA-Flow Policy: Temporally Coherent Imitation Learning via Continuous Latent Action Flow Matching for Robotic Manipulation
  8. 2026Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry