AI-SOVEREIGN
Home / AI Research / Teaching Vision-Language-Action Models to Stop Safely
TEAI Research

Teaching Vision-Language-Action Models to Stop Safely

by DM2-ND · Zheyuan Liu, Chuhan Wang, Jinduo Guo, Yue Huang, Xiangchi Yuan, Ningshan Ma

Vision-language-action (VLA) models enable robots to perform diverse tasks from visual observations and language instructions. While prior work has explored task suppression and refusal, stopping prohibited behavior once execution is underway and controlling the resulting physical response remain challenging. Our evaluation reveals that policies which suppress a task from its initial state can still complete it when taking over partway through execution. Even when the task remains unfinished, the robot may drop a held object, making non-completion an incomplete measure of safety. To address these challenges, we introduce State-Aware Fallback Imitation (SAFI) for tasks where holding…

Upvotes7
Comments3
Published18 Nov 2026