DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors

• URL: http://arxiv.org/abs/2604.25050v1
• Date: Mon, 27 Apr 2026 23:04:03 GMT
• Title: DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors
• Abstract summary: We propose DiscreteRTC, which replaces external corrections with native unmasking.<n>DiscreteRTC is simpler to implement with 0 lines of code for async inpainting, faster at inference with only 0.7x computation compared with generating actions from scratch, and better at execution with 50% higher success rate in real-world dynamic pick task compared with flow-based RTC.
• Score: 57.944744187489185
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Unlike chatbots, physical AI must act while the world keeps evolving. Therefore, the inter-chunk pause of synchronous executors are fatal for dynamic tasks regardless of how fast the inference is. Asynchronous execution -- thinking while acting -- is therefore a structural requirement, and real-time chunking (RTC) makes it viable by recasting chunk transitions as inpainting: freezing committed actions and consistently generating the remainder. However, RTC with flow-matching policy is structurally suboptimal: its inpainting comes from inference-time corrections rather than the base policy, yielding little pre-training benefit, specific fine-tuning, heuristic guidance, and extra computation that inflates the latency. In this work, we observe that discrete diffusion policies, which generate actions by iteratively unmasking, are natural asynchronous executors that resolve all limitations at once: they are fine-tuning free since inpainting is their native operation, while early stopping further provides adaptive guidance and reduces inference cost. We propose DiscreteRTC, which replaces external corrections with native unmasking, and show on dynamic simulated benchmarks and real-world dynamic manipulation tasks that it achieves higher success rates than continuous RTC and other baselines. In summary, DiscreteRTC is simpler to implement with 0 lines of code for async inpainting, faster at inference with only 0.7x computation compared with generating actions from scratch, and better at execution with 50% higher success rate in real-world dynamic pick task compared with flow-matching-based RTC. More visualizations are on https://outsider86.github.io/DiscreteRTCSite/.

Related papers

• SmoothRL: Online Reinforcement Learning During Asynchronous Execution [12.24698147742807]
  Deploying robot policies in the physical world requires satisfying two fundamental desiderata: reliability and smooth real-time execution.<n>We present SmoothRL, an online reinforcement learning framework that fine-tunes a pretrained policy within an asynchronous inference loop.<n>We evaluate SmoothRL on real-world robotic tasks requiring high precision, as well as highly dynamic tasks that necessitate asynchronous execution.
  arXiv  Detail & Related papers  (2026-08-30T12:56:09Z)
• FutureRTC: Real-Time Robot Execution with Anticipatory-Conditioned Action Chunking [51.97839311685636]
  FutureRTC is a plug-and-play adaptation framework that predicts execution-time observations and states for asynchronous VLA control without modifying the underlying policy.<n>We show that FutureRTC achieves robustness to inference delays, resulting in smoother trajectories, faster execution, and consistently higher task success rates.
  arXiv  Detail & Related papers  (2026-07-27T05:10:16Z)
• Start Right, Arrive Right: Asynchronous Execution via Initial Noise Selection [7.357958909658845]
  Action chunking enables robot policies to produce temporally coherent behavior.<n> generating multi-step action sequences with flow-based policies incurs latency that is incompatible with real-time control.<n>We introduce textbfPAINT, a training-free method that finds this noise via backward inversion and constructs the final chunk through a repainting rule.
  arXiv  Detail & Related papers  (2026-06-18T04:14:11Z)
• Action-Prior Denoising for Smooth Real-Time Chunking [12.956533987402054]
  Real-time chunking (RTC) lets chunked action policies operate under inference delay by conditioning a newly generated action chunk on actions already committed by the previous chunk.<n>Training-time RTC simulates this delay during learning and avoids expensive runtime guidance at deployment.<n>We propose Soft RTC, a training-time RTC generalization based on action-prior denoising.
  arXiv  Detail & Related papers  (2026-05-25T07:49:25Z)
• Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy [52.106797722292896]
  We present DCDP, a Dynamic Closed-Loop Diffusion Policy framework that integrates chunk-based action generation with real-time correction.<n>In dynamic PushT simulations, DCDP improves adaptability by 19% without retraining while requiring only 5% additional computation.
  arXiv  Detail & Related papers  (2026-03-02T15:04:18Z)
• Unbiased Gradient Estimation for Event Binning via Functional Backpropagation [64.88399635309918]
  We propose a novel framework for unbiased gradient estimation of arbitrary binning functions by synthesizing weak derivatives during backpropagation.<n>We achieve 9.4% lower EPE in self-supervised optical flow, and 5.1% lower RMS error in SLAM, demonstrating broad benefits for event-based visual perception.
  arXiv  Detail & Related papers  (2026-02-13T04:05:03Z)
• Real-Time Robot Execution with Masked Action Chunking [38.37108371991901]
  Real-time execution is essential for cyber-physical systems such as robots.<n>Asynchronous inference has recently emerged as a system-level paradigm for real-time robot manipulation.<n>We propose REMAC, which learns corrective adjustments on the pretrained policy through masked action chunking.
  arXiv  Detail & Related papers  (2026-01-27T23:48:32Z)
• Training-Time Action Conditioning for Efficient Real-Time Chunking [45.91723746443767]
  We propose simulating inference delay at training time and conditioning on action prefixes directly, eliminating inference-time overhead.<n>In simulated experiments, we find that training-time RTC outperforms inference-time RTC at higher inference delays.<n>Our results suggest that training-time action conditioning is a practical drop-in replacement for inference-time inpainting in real-time robot control.
  arXiv  Detail & Related papers  (2025-12-05T18:57:28Z)
• AsyncHZP: Hierarchical ZeRO Parallelism with Asynchronous Scheduling for Scalable LLM Training [4.643969942380424]
  We propose a novel asynchronous variant of ZeRO to achieve superior performance while maintaining simplicity and memory efficiency.<n>Unlike traditional ZeRO, which employs over-fine-grained sharding that can lead to inefficient communication, AsyncHZP adaptively reshards parameters, gradients, and states across different replica groups.<n>AsyncHZP consistently outperforms classic ND parallelism, achieving state-of-the-art performance without complex strategic tuning.
  arXiv  Detail & Related papers  (2025-10-23T01:29:35Z)
• Real-Time Execution of Action Chunking Flow Policies [49.1574468325115]
  This paper presents a novel inference-time algorithm that enables asynchronous execution of action interacting systems.<n>It is applicable to any diffusion- or VLA-based systems executing out of the box with no re-training.<n>Results show that RTC is fast, performant, and uniquely robust to inference manipulation.
  arXiv  Detail & Related papers  (2025-06-09T01:01:59Z)
• COrAL: Order-Agnostic Language Modeling for Efficient Iterative Refinement [80.18490952057125]
  Iterative refinement has emerged as an effective paradigm for enhancing the capabilities of large language models (LLMs) on complex tasks. We propose Context-Wise Order-Agnostic Language Modeling (COrAL) to overcome these challenges. Our approach models multiple token dependencies within manageable context windows, enabling the model to perform iterative refinement internally.
  arXiv  Detail & Related papers  (2024-10-12T23:56:19Z)
• AEGNN: Asynchronous Event-based Graph Neural Networks [54.528926463775946]
  Event-based Graph Neural Networks generalize standard GNNs to process events as "evolving"-temporal graphs. AEGNNs are easily trained on synchronous inputs and can be converted to efficient, "asynchronous" networks at test time.
  arXiv  Detail & Related papers  (2022-03-31T16:21:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.