cs.RONov 6, 2026

ODEAction: From Predicting Chunks to Evolving Actions

Authors: Qishen Wang, Pengxiang Ding, Shifeng Bao, Mingyang Sun, Hengtao Li, Yi Gao, Baohua Yin, Yang Liu, 

Organizations: Symbiosis Robotics · University of Science and Technology of China · Zhejiang University · Westlake University · 

Abstract

Vision-language-action (VLA) models typically generate actions in chunks, dividing continuous robot execution into separately predicted action chunks, which can introduce discontinuities at chunk boundaries. We propose ODEAction, a chunk-free action generation paradigm that replaces action chunk prediction with Continuous Action Evolution. We use an ordinary differential equation (ODE) to describe the continuous evolution of actions, and define the vector field of this ODE as the Action Transition Field to characterize how actions evolve with respect to evolution time. During inference, the model takes the current action as the initial condition and numerically solves the ODE defined by the learned Action Transition Field, thereby recursively generating subsequent actions with explicit dependencies. ODEAction is implemented as a modular action head that can be integrated with different vision-language backbones. Experiments on LIBERO, EBench, RoboTwin, and real-robot tasks show that, compared with action-chunk-based baselines, ODEAction improves both task performance and trajectory quality, with particularly pronounced gains on fine manipulation tasks.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. E0: Expressive Fine-Grained Discrete Action Prediction for Vision-Language-Action Models via Tweedie Discrete Diffusion

    Nov 26, 2025Zhihao Zhan, Jiaying Zhou, Likui Zhang +10Vision-Language-Action ModelsDiffusion-Based VLA Models

  2. ODE Policy: Robot Learning via Action Dynamics, Not Actions

    Nov 1, 2026Jindou Jia, Jingliang Li, Kuangji Zuo +9Robot Policy LearningRobotic RL

  3. DriftingVLA: Native One-Step Vision-Language-Action Generation via Per-Dimension Temporal Drifting

    Aug 30, 2026Yuxuan Gao, Shiqi Zhang, Yedong Shen +6Efficient VLA Model InferenceVisuomotor Policy Learning