SOURCE-LINKED INTELLIGENCE

Learning Beyond What Humans Can Demonstrate

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Behavior cloning for robot manipulation relies on expert demonstrations. However, for tasks that require dynamic stability, precise contact timing, or dexterous coordination, human operators may find it hard or even impossible to collect data. We study this infeasible-demonstration regime and propose GLIDE: Guardrails for Learning from Infeasible Demonstrations Efficiently, a framework that infers task-specific failure modes and converts them into executable guardrails for data collection and policy deployment. Given a task description and the conditioning teleoperation code, GLIDE writes guar

Read original source ↗ Open in workspace

recordType

paper

region

Global