Skip to content
Preprint

Why Does Action Chunking Improve Behavioral Cloning Performance in Robotic Control?

Aug 2026 · 13 citations · ⚡ 2 influential · 63 references
Computer Science

TL;DR

In simulated and real-world robotic control settings, it is shown that in simulated and real-world robotic control settings, one can match the performance of action chunking without action chunking---by deploying an action chunking policy as an ensemble of policies with randomized delays.

Abstract

Action chunking---predicting and executing multiple actions instead of a single action---has proven to be a critical component for learning effective robotic control policies. However, our precise understanding of why action chunking improves performance has remained limited. In this work we seek to close this gap. Through rigorous experimental evaluations in both simulated and real-world settings, we show that existing hypotheses for the success of action chunking---temporal consistency, horizon reduction, and representation learning---fail to explain the success of action chunking. Instead, we find that action chunking benefits from greater non-Markovian expressivity and reduced compounding error compared to Markovian policies, but, in many settings of interest, these effects can be fully captured by delayed policies, which at each step predict a single action based on the observation $k$ steps in the past. We then show that there exists an additional benefit of action chunking that we refer to as implicit ensembling. In particular, by learning a diversity of temporal relationships (that is, $a_t | o_t, a_t | o_{t-1}, \ldots$), action-chunked policies exhibit behavior matching that of a model ensemble, increasing their robustness and generalization ability over policies that only learn a single temporal relationship. Building on these insights, we show that in simulated and real-world robotic control settings, we can match the performance of action chunking without action chunking---by deploying an action chunking policy as an ensemble of policies with randomized delays. Furthermore, we propose a policy class that amplifies the benefits of action chunking by explicitly instantiating an ensemble, and which we show significantly improves over the performance of action chunking in many domains.

View source

Similar papers

#machine learning Preprint Aug 2026

Three Steps at a Time: Learning Representations from Action Sequences in Contrastive RL

This work extends contrastive reinforcement learning (CRL), a prototypical self-supervised method, to operate over action chunks, and finds that this results in large, pervasive gains across established offline and online benchmarks: +31.7% and +93.1% across 18 and 11 environments respectively.

Michal Korniak, Kamil Dybek, Benjamin Eysenbach et al. · 3 citations · ⚡1
#machine learning Preprint Sep 2026

Action Chunking Proximal Policy Optimization with Feedback Correction

Action chunking provides temporal abstraction in reinforcement learning by selecting short action sequences instead of individual actions, but many existing approaches face two limitations in high-dimensional robotic control. First, many rely on value functions over action chunks, which can be difficult to learn as act...

Sang J. Hahn, Jonghyun Choi · 0 citations
Preprint Aug 2026

Revisiting Open-Loop Execution in Robotics: Toward Reactive, Higher-Performing Policies

Action chunking --- the practice of predicting a sequence of actions and executing a prefix open-loop --- has emerged as a key enabler of recent progress in imitation learning for robotic manipulation. However, executing long open-loop prefixes reduces reactivity, limiting policies'ability to correct for errors. Furthe...

Michael Zeng, Abhinav Agarwal, Ajay Bati et al. · 4 citations · ⚡1
Preprint Oct 2026

When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models

Vision-Language-Action (VLA) models serve as unified policies for robotic manipulation, yet their expensive inference forces robots to pause between policy calls, resulting in stop-and-go execution that interrupts smooth motion and prolongs task completion. Extending the action chunk reduces policy calls and hence thes...

Seonghoon Yu, Dongwon Kim, Hyungrok Jung et al. · 1 citation
#artificial intelligence Preprint Oct 2026

Measuring the Stability Assumption Behind Action Chunking

Action chunking improves the performance of policies learned by behavioural cloning, and several mechanisms have been proposed to explain why, including temporal consistency, horizon reduction, representation learning, and reduced error compounding. We instead study what happens to an action error once it enters the sy...

Aryan Goyal · 0 citations

Achieve What You Imagined: Learning to Align Actions with Visual Plans

This work uses a frozen action-conditioned world model to predict action-conditioned consequences and construct feedback based on consistency between the two future predictions and alignment with the terminal goal to show that cross-model prediction discrepancy can provide useful feedback for improving robot policies u...

Yu-Heng Qiao, Zi-Ran Wei, Xiao-Hang Wang et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.