Back to AIBriefs
AnalysisAI AgentsAI Models

Online Skill Learning for Web Agents via State-Grounded Dynamic Retrieval

The paper proposes a method for web agents to learn reusable skills from past task trajectories using state-grounded dynamic retrieval. This approach improves multi-step web automation by enabling skill induction and reuse across related tasks.

·
7 days ago
Online Skill Learning for Web Agents via State-Grounded Dynamic Retrieval — AIBriefs