ActionFlow: A Pipelined Action Acceleration for Vision Language Models on Edge

• URL: http://arxiv.org/abs/2512.20276v1
• Date: Tue, 23 Dec 2025 11:29:03 GMT
• Title: ActionFlow: A Pipelined Action Acceleration for Vision Language Models on Edge
• Authors: Yuntao Dai, Hang Gu, Teng Wang, Qianyu Cheng, Yifei Zheng, Zhiyong Qiu, Lei Gong, Wenqi Lou, Xuehai Zhou,
• Abstract summary: Vision-Language-Action (VLA) models have emerged as a unified paradigm for robotic perception and control.<n>Current VLA models operate at only 3-5 Hz on edge devices due to the memory bound nature of autoregressive decoding.<n>We introduce ActionFlow, a system-level inference framework tailored for resource-constrained edge plat forms.
• Score: 11.016302257907936
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision-Language-Action (VLA) models have emerged as a unified paradigm for robotic perception and control, enabling emergent generalization and long-horizon task execution. However, their deployment in dynamic, real-world environments is severely hin dered by high inference latency. While smooth robotic interaction requires control frequencies of 20 to 30 Hz, current VLA models typi cally operate at only 3-5 Hz on edge devices due to the memory bound nature of autoregressive decoding. Existing optimizations often require extensive retraining or compromise model accuracy. To bridge this gap, we introduce ActionFlow, a system-level inference framework tailored for resource-constrained edge plat forms. At the core of ActionFlow is a Cross-Request Pipelin ing strategy, a novel scheduler that redefines VLA inference as a macro-pipeline of micro-requests. The strategy intelligently batches memory-bound Decode phases with compute-bound Prefill phases across continuous time steps to maximize hardware utilization. Furthermore, to support this scheduling, we propose a Cross Request State Packed Forward operator and a Unified KV Ring Buffer, which fuse fragmented memory operations into efficient dense computations. Experimental results demonstrate that ActionFlow achieves a 2.55x improvement in FPS on the OpenVLA-7B model without retraining, enabling real-time dy namic manipulation on edge hardware. Our work is available at https://anonymous.4open.science/r/ActionFlow-1D47.

Related papers

• LiteVLA-Edge: Quantized On-Device Multimodal Control for Embedded Robotics [0.6119773373677944]
  We present LiteVLA-Edge, a deployment-oriented VLA pipeline for fully on-device inference on Jetson Orin-class hardware.<n>Our approach combines supervised image-to-action fine-tuning in FP32 with post-training 4-bit GGUF quantization and GPU-accelerated inference.<n>Under our configuration, LiteVLA-Edge achieves a mean end-to-end runtime of 150.5,ms (approximately 6.6,Hz) while operating entirely offline.
  arXiv  Detail & Related papers  (2026-03-03T03:20:52Z)
• HybridFlow: A Two-Step Generative Policy for Robotic Manipulation [2.2200541495683996]
  MeanFlow, as a one-step variant of flow matching, has shown strong potential in image generation.<n>HybridFlow balances inference speed and generation quality by leveraging the rapid advantage of MeanFlow one-step generation.<n>We envision HybridFlow as a practical low-latency method to enhance real-world interaction capabilities of robotic manipulation policies.
  arXiv  Detail & Related papers  (2026-02-14T10:50:23Z)
• AsyncVLA: An Asynchronous VLA for Fast and Robust Navigation on the Edge [49.66156306240961]
  High latency breaks the control loop, rendering powerful models unsafe for real-time deployment.<n>We propose AsyncVLA, an asynchronous control framework that decouples semantic reasoning from reactive execution.<n>AsyncVLA achieves a 40% higher success rate than state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-02-13T21:31:19Z)
• TIDAL: Temporally Interleaved Diffusion and Action Loop for High-Frequency VLA Control [15.534182843429043]
  Large-scale Vision-Language-Action (VLA) models offer semantic generalization but suffer from high inference latency.<n>We propose TIDAL, a hierarchical framework that decouples semantic reasoning from high-frequency actuation.<n> TIDAL operates as a backbone-agnostic module for diffusion-basedVLAs, using a dual-frequency architecture.
  arXiv  Detail & Related papers  (2026-01-21T12:43:11Z)
• Asynchronous Fast-Slow Vision-Language-Action Policies for Whole-Body Robotic Manipulation [10.09057399213028]
  Vision-Language-Action (VLA) systems integrate a Vision-Language Model (VLM) for semantic reasoning with an action expert generating continuous action signals.<n>We introduce a truly asynchronous Fast-Slow VLA framework (DuoCore-FS) to organize the system into a fast pathway for action generation and a slow pathway for rich VLM reasoning.
  arXiv  Detail & Related papers  (2025-12-23T09:28:20Z)
• ARMFlow: AutoRegressive MeanFlow for Online 3D Human Reaction Generation [48.716675019745885]
  3D human reaction generation faces three main challenges: high motion fidelity, real-time inference, and autoregressive adaptability for online scenarios.<n>We propose ARMFlow, a MeanFlow-based autoregressive framework that models temporal dependencies between motions and velocity.<n>Our single-step online generation surpasses existing methods on InterHuman and InterX by over 40% in FID, while matching offline state-of-the-art performance despite using only partial sequence conditions.
  arXiv  Detail & Related papers  (2025-12-18T06:28:42Z)
• Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach [78.4812458793128]
  We propose textbfTACO, a test-time-scaling framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks.<n>Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits.
  arXiv  Detail & Related papers  (2025-12-02T14:42:54Z)
• OmniSAT: Compact Action Token, Faster Auto Regression [70.70037017501357]
  We introduce an Omni Swift Action Tokenizer, which learns a compact, transferable action representation.<n>The resulting discrete tokenization shortens the training sequence by 6.8$times$, and lowers the target entropy.
  arXiv  Detail & Related papers  (2025-10-08T03:55:24Z)
• SuperFlow++: Enhanced Spatiotemporal Consistency for Cross-Modal Data Pretraining [62.433137130087445]
  SuperFlow++ is a novel framework that integrates pretraining and downstream tasks using consecutive camera pairs.<n>We show that SuperFlow++ outperforms state-of-the-art methods across diverse tasks and driving conditions.<n>With strong generalizability and computational efficiency, SuperFlow++ establishes a new benchmark for data-efficient LiDAR-based perception in autonomous driving.
  arXiv  Detail & Related papers  (2025-03-25T17:59:57Z)
• FAST: Efficient Action Tokenization for Vision-Language-Action Models [98.15494168962563]
  We propose a new compression-based tokenization scheme for robot actions, based on the discrete cosine transform.<n>Based on FAST, we release FAST+, a universal robot action tokenizer, trained on 1M real robot action trajectories.
  arXiv  Detail & Related papers  (2025-01-16T18:57:04Z)
• ScaleFlow++: Robust and Accurate Estimation of 3D Motion from Video [26.01796507893086]
  This paper proposes a 3D motion perception method called ScaleFlow++ that is easy to generalize. With just a pair of RGB images, ScaleFlow++ can robustly estimate optical flow and motion-in-depth (MID) On KITTI, ScaleFlow++ achieved the best monocular scene flow estimation performance, reducing SF-all from 6.21 to 5.79.
  arXiv  Detail & Related papers  (2024-09-16T11:59:27Z)
• ActionFlow: Equivariant, Accurate, and Efficient Policies with Spatially Symmetric Flow Matching [20.20511152176522]
  ActionFlow is a policy class that integrates spatial symmetry inductive biases. On the representation level, ActionFlow introduces an SE(3) Invariant Transformer architecture. For action generation, ActionFlow leverages Flow Matching, a state-of-the-art deep generative model.
  arXiv  Detail & Related papers  (2024-09-06T19:30:36Z)
• ScaleFlow++: Robust and Accurate Estimation of 3D Motion from Video [15.629496237910999]
  This paper proposes a 3D motion perception method called ScaleFlow++ that is easy to generalize. With just a pair of RGB images, ScaleFlow++ can robustly estimate optical flow and motion-in-depth (MID) On KITTI, ScaleFlow++ achieved the best monocular scene flow estimation performance, reducing SF-all from 6.21 to 5.79.
  arXiv  Detail & Related papers  (2024-07-13T07:58:48Z)
• GMFlow: Learning Optical Flow via Global Matching [124.57850500778277]
  We propose a GMFlow framework for learning optical flow estimation. It consists of three main components: a customized Transformer for feature enhancement, a correlation and softmax layer for global feature matching, and a self-attention layer for flow propagation. Our new framework outperforms 32-iteration RAFT's performance on the challenging Sintel benchmark.
  arXiv  Detail & Related papers  (2021-11-26T18:59:56Z)

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.