paper-with-me

Papers

Sparse ActionGen: Accelerating Diffusion Policy with Real-time Pruning

2026-01-19 · Kangye Ji, Jianbo Zhou, Yuan Meng, Ye Li, Hanyun Cui, Zhi Wang arxiv

Diffusion Policy has dominated action generation due to its strong capabilities for modeling multi-modal action distributions, but its multi-step denoising processes make it impractical for real-time visuomotor control. Existing caching-based acceleration methods typically rely on $\textit{static}$ schedules that fail to adapt to the $\textit{dynamics}$ of robot-environment interactions, thereby leading to suboptimal performance. In this paper, we propose $\underline{\textbf{S}}$parse $\underline{\textbf{A}}$ction$\underline{\textbf{G}}$en ($\textbf{SAG}$) for extremely sparse action generation. To accommodate the iterative interactions, SAG customizes a rollout-adaptive prune-then-reuse mechanism that first identifies prunable computations globally and then reuses cached activations to substitute them during action diffusion. To capture the rollout dynamics, SAG parameterizes an observation-conditioned diffusion pruner for environment-aware adaptation and instantiates it with a highly parameter- and inference-efficient design for real-time prediction. Furthermore, SAG introduces a one-for-all reusing strategy that reuses activations across both timesteps and blocks in a zig-zag manner, minimizing the global redundancy. Extensive experiments on multiple robotic benchmarks demonstrate that SAG achieves up to 4$\times$ generation speedup without sacrificing performance. Project Page: https://sparse-actiongen.github.io.

📄 PDF Abstract BibTeX arXiv:2601.12894

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

SPADE: An Input-Adaptive Sparse Attention Engine for Fast Video Diffusion Models Inference

2026-08-04 · Shanghao Liu, Renze Chen, Size Zheng, Yuanqiang Liu 외 arxiv

Video diffusion transformers (vDiTs) generate high quality but pay quadratic self-attention cost, making inference prohibitive at video-token scales. The challenge is input-adaptive sparsity: selecting critical Q/K/V tok…

Video Generation

Accelerating Residual Reinforcement Learning with Uncertainty Estimation

2025-06-21 · Lakshita Dodeja, Karl Schmeckpeper, Shivam Vats, Thomas Weng 외

Residual Reinforcement Learning (RL) is a popular approach for adapting pretrained policies by learning a lightweight residual policy that provides corrective actions. While Residual RL is more sample-efficient than fine…

D4RLreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Sparse Diffusion Policy: A Sparse, Reusable, and Flexible Policy for Robot Learning

2024-07-01 · Yixiao Wang, Yifei Zhang, Mingxiao Huo, Ran Tian 외

The increasing complexity of tasks in robotics demands efficient strategies for multitask and continual learning. Traditional models typically rely on a universal policy for all tasks, facing challenges such as high comp…

Continual LearningMixture-of-Experts

AID-DTI: Accelerating High-fidelity Diffusion Tensor Imaging with Detail-Preserving Model-based Deep Learning

2024-01-03 · Wenxin Fan, Jian Cheng, Cheng Li, Xinrui Ma 외

Deep learning has shown great potential in accelerating diffusion tensor imaging (DTI). Nevertheless, existing methods tend to suffer from Rician noise and detail loss in reconstructing the DTI-derived parametric maps es…

PulseCol: Periodically Refreshed Column-Sparse Attention for Accelerating Diffusion Language Models

2026-05-20 · Yanyi Lyu, Letian Chen, Futing Sun, Miao Zhang 외 arxiv

Inference in diffusion large language models (dLLMs) is computationally expensive, as full self-attention must be repeatedly executed at each step of the denoising process without KV cache. Recent sparse attention method…

Computational Efficiency