Move-Then-Operate: Behavioral Phasing for Human-Like Robotic Manipulation

TL;DR

The paper introduces Move-Then-Operate, a dual-phase robotic manipulation framework that improves success rates and training efficiency by explicitly separating movement and contact phases with a learnable phase selector.

Contribution

It proposes a novel dual-expert policy architecture with automatic phase labeling, enhancing manipulation performance and data efficiency over monolithic approaches.

Findings

Achieves 68.9% success rate on RoboTwin2 benchmark.

Outperforms monolithic baseline by 24%.

Reaches peak performance in 40% fewer training steps.

Abstract

We present Move-Then-Operate, a Vision language action framework that explicitly decouples robotic manipulation into two distinct behavioral phases: coarse relocation (move) and contact-critical interaction (operate). Unlike monolithic policies that conflate these heterogeneous regimes, our architecture employs a dual-expert policy routed by a learnable phase selector, introducing a structural inductive bias that isolates phase-specific dynamics. Phase labels are automatically generated via an MLLM-based pipeline conditioned on lightweight contextual cues such as end-effector velocity and subtask decomposition to ensure alignment with human motor patterns. Evaluated on the RoboTwin2 benchmark, our method achieves an average success rate of $68.9\%$, outperforming the monolithic $\pi_0$ baseline by $24\%$. It matches or exceeds models trained on $10\times$ more data and reaches peak performance in $40\%$ fewer training steps, demonstrating that architectural disentanglement of move and operate phases is a highly effective and efficient strategy for mastering high-precision manipulation.