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

ArXi:2606.04391v1 Announce Type: new Language agents increasingly rely on reusable skills to improve multi-step web automation across related tasks. A growing line of work studies online skill learning, where agents continually induce skills from previous task trajectories and reuse them in future tasks on the fly. However, existing methods mainly reuse skills at the task-level: a fixed set of skills is retrieved based on the initial task instruction and then held fixed throughout execution.