Understanding Behavior Cloning with Action Quantization

ArXi:2603.20538v1 Announce Type: new Behavior cloning is a fundamental paradigm in machine learning, enabling policy learning from expert nstrations across robotics, autonomous driving, and generative models. Autoregressive models like transformer have proven remarkably effective, from large language models (LLMs) to vision-language-action systems (VLAs). However, applying autoregressive models to continuous control requires discretizing actions through quantization, a practice widely adopted yet poorly understood theoretically. This paper provides theoretical foundations for this practice.