PolySkill: Learning Generalizable Skills Through Polymorphic Abstraction

Researchers introduce PolySkill, a framework that enables AI agents to learn generalizable skills by separating abstract goals from concrete implementations, inspired by software engineering polymorphism. The method improves skill reuse by 1.7x and boosts success rates by up to 13.9% on web navigation tasks while reducing execution steps by over 20%.

• →PolySkill framework decouples skill goals from their implementation to create more generalizable AI agent capabilities.
• →The method improves skill reuse by 1.7x on familiar websites and increases success rates up to 13.9% on new sites.
• →Agents using PolySkill require 20% fewer steps to complete tasks compared to baseline methods.
• →The framework enables self-exploration learning without specified tasks, improving autonomous skill acquisition.
• →Research provides a practical approach toward building continuously learning agents for web environments.