Skill-SLM: Agent Skill-driven Small Language Models for Reliable Robot Operation

What happened

arXiv:2610.10812v1 Announce Type: new Abstract: Small language models (SLMs) have been increasingly adopted for onboard robot operation because they enable intelligent decision-making. This paper proposes Skill-SLM, a framework that reformulates SLM-driven robot operation as a task-decomposition and skill-composition problem.

Given a natural language task instruction, Skill-SLM decomposes the task into subtasks, selects appropriate skills from the skill library, and orchestrates the selected skills into executable robot operations. Experiments on UAV operation tasks indicate that Skill-SLM substantially outperforms distillation-oriented baselines, especially on unseen tasks that require generalization of capabilities.

However, existing approaches are mainly distillation-oriented and rely on enumerating representative task-solution pairs. This makes dataset construction difficult and limits generalization to diverse robot tasks whose possible forms grow rapidly.