Can Explicit Physical Feasibility Benefit VLA Learning? An Empirical Study

TL;DR

This paper investigates whether explicit physical feasibility supervision can improve vision-language-action models for robot control, showing that it enhances reliability, efficiency, and task performance.

Contribution

The study introduces a geometry-grounded feasibility objective into VLA training and demonstrates its benefits through obstacle-aware manipulation experiments.

Findings

Feasibility supervision improves physical reliability of VLA policies.

Augmenting training with feasibility signals enhances task performance.

Explicit feasibility guidance accelerates learning in low-data scenarios.

Abstract

Vision-Language-Action (VLA) models map multimodal inputs directly to robot actions and are typically trained through large-scale imitation learning. While this paradigm has shown strong performance, prevailing VLA training procedures do not explicitly supervise hard physical constraints such as obstacle avoidance or kinematic feasibility. As a result, the geometric structure underlying physically feasible behavior must be inferred only implicitly from demonstrations. In this paper, we study whether introducing explicit feasibility supervision can provide effective structured guidance for VLA policies. We formulate a simple geometry-grounded feasibility objective and integrate it into the training stage of a diffusion-based VLA policy. To evaluate this idea systematically, we use obstacle-aware manipulation as a controlled probe of geometry-dependent physical feasibility. Empirical results show that augmenting VLA training with feasibility supervision improves both physical reliability and overall task performance, while also enhancing learning efficiency in the low-data regime. These findings indicate that explicit feasibility signals can effectively complement imitation-based VLA learning, highlighting their potential for developing more reliable VLA policies.