Action-Prior Denoising for Smooth Real-Time Chunking

• URL: http://arxiv.org/abs/2605.25537v1
• Date: Mon, 25 May 2026 07:49:25 GMT
• Title: Action-Prior Denoising for Smooth Real-Time Chunking
• Abstract summary: 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.
• Score: 12.956533987402054
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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. Training-time RTC simulates this delay during learning and avoids expensive guidance at deployment, but its binary prefix mask treats all non-prefix tokens as fully unconstrained. This under-models asynchronous execution: early overlap actions are fixed, while later overlap actions remain editable but should still stay close to the previous plan. We propose Soft RTC, a training-time RTC generalization based on action-prior denoising. Soft RTC constructs corrupted overlap tokens from partially denoised states instead of pure noise and injects the aligned previous chunk as the same prior during inference through a lightweight token-wise blending rule. On the 12 released large Kinetix levels, a short soft window nearly matches hard training-time RTC in overall solve rate (0.809 vs. 0.815), while a medium window reduces high-delay action delta and jerk by 9.1% and 9.6% relative to hard RTC. Both variants keep near-naive runtime, unlike inference-time RTC baselines. A small preliminary real-robot sorting study provides additional evidence that training-time RTC can improve completion and that Soft RTC gives the lowest commanded-action finite-difference metrics among the tested policies.

Related papers

• ChunkFlow: Towards Continuity-Consistent Chunked Policy Learning [5.727608625370017]
  ChunkFlow is a seam-aware training-and-execution framework for chunked policies.<n>It partitions each chunk into frozen, editable, and future zones, applies deterministic overlap blending at execution, and trains raw predictions with seam and first- and second-order continuity losses.<n> Experiments on CALVIN, LIBERO, and real robots show an improved success-stability trade-off with low-latency inference.
  arXiv  Detail & Related papers  (2026-07-14T17:43:25Z)
• SEAM: Smooth Execution of Action-Chunked Motion for Vision-Language-Action Policies [26.372187124257362]
  Vision-Language-Action (VLA) policies that execute fixed-length action chunks can exhibit multimodal bifurcation.<n>We propose SEAM (Smooth Execution of Action-Chunked Motion), a training-free inference-time method for flow matching VLAs.<n>On LIBERO-10 with pi_0.5, SEAM reduces boundary jerk by 28%, reduces chunk transition discontinuity by 27%, preserves baseline-level task success, and keeps denoising-loop cost near the unguided baseline.
  arXiv  Detail & Related papers  (2026-07-06T02:32:10Z)
• DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors [57.944744187489185]
  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.
  arXiv  Detail & Related papers  (2026-04-27T23:04:03Z)
• Learning Native Continuation for Action Chunking Flow Policies [40.56048312294812]
  Action chunking enables Vision Language Action (VLA) models to run in real time, but naive chunked execution often exhibits discontinuities at chunk boundaries.<n>We propose Legato, a training-time continuation method for action-chunked flow-based VLA policies.<n>Legato produces smoother trajectories and reduces spurious multimodal switching during execution, leading to less hesitation and shorter task completion time.
  arXiv  Detail & Related papers  (2026-02-13T14:56:06Z)
• 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)
• 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)
• Training Long-Context LLMs Efficiently via Chunk-wise Optimization [60.05884946552877]
  We present textitSequential Chunk-wise Optimization (SeCO), a memory-efficient training paradigm that partitions lengthy inputs into manageable chunks.<n>We also introduce textitSparse Chunk-wise Optimization (SpaCO), which reduces computational overhead by selectively propagating gradients to specific chunks.<n>SpaCO decouples the computational cost of backpropagation from the context length, enabling training time to gradually converge to inference time as sequences become longer.
  arXiv  Detail & Related papers  (2025-05-22T14:11:34Z)
• Efficient local linearity regularization to overcome catastrophic overfitting [59.463867084204566]
  Catastrophic overfitting (CO) in single-step adversarial training results in abrupt drops in the adversarial test accuracy (even down to 0%) We introduce a regularization term, called ELLE, to mitigate CO effectively and efficiently in classical AT evaluations.
  arXiv  Detail & Related papers  (2024-01-21T22:55:26Z)
• Non-stationary Delayed Online Convex Optimization: From Full-information to Bandit Setting [71.82716109461967]
  We propose an algorithm called Mild-OGD for the full-information case, where delayed gradients are available.<n>We show that the dynamic regret of Mild-OGD can be automatically bounded by $O(sqrtbardT(P_T+1))$ under the in-order assumption.<n>We also develop a bandit variant of Mild-OGD for a more challenging case with only delayed loss values.
  arXiv  Detail & Related papers  (2023-05-20T07:54:07Z)
• Relaxing the Conditional Independence Assumption of CTC-based ASR by Conditioning on Intermediate Predictions [14.376418789524783]
  We train a CTC-based ASR model with auxiliary CTC losses in intermediate layers in addition to the original CTC loss in the last layer. Our method is easy to implement and retains the merits of CTC-based ASR: a simple model architecture and fast decoding speed.
  arXiv  Detail & Related papers  (2021-04-06T18:00:03Z)

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.