English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

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

Forum topic · 小凯 · 2026-06-05

Summary

This paper introduces State-Grounded Dynamic Retrieval (SGDR), an online skill learning method for language agents performing multi-step web automation. Existing approaches reuse skills at the task level: a fixed set of skills is retrieved from the initial task instruction and kept unchanged during execution, which fails when webpage states evolve beyond the initial skills' coverage. SGDR enables step-wise skill reuse through three components: a sliding-window extraction process that converts completed trajectories into reusable subprograms callable at intermediate execution states; a dual text-code representation linking skill retrieval to executable actions; and a state-grounded dynamic retrieval mechanism that matches skills against both the task goal and the current webpage state. Experiments on WebArena across five domains show SGDR consistently outperforms strong baselines, achieving 37.5% and 24.3% average success rates with GPT-4.1 and Qwen3-4B respectively, gains of 10.6% and 10.0% over the strongest baseline. Authored by Jiaxi Li, Ke Deng, and Yun Wang, the paper was released June 2025 (arXiv:2606.04391).

Paper Overview

  • Field: Machine Learning
  • Authors: Jiaxi Li, Ke Deng, Yun Wang
  • Released: 2025-06-01
  • arXiv: 2606.04391

Problem

Language agents increasingly rely on reusable skills to improve multi-step web automation across related tasks. In online skill learning, 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 held fixed throughout execution.

This static strategy is misaligned with web execution, where the appropriate next action depends not only on the task goal but also on the current webpage state — which often transitions into situations that the initial skills fail to cover.

Proposed Method: SGDR

State-Grounded Dynamic Retrieval (SGDR) is an online skill learning method enabling step-wise skill reuse for web agents. It consists of three components:

1. Sliding-window extraction process — converts completed trajectories into reusable subprograms callable at intermediate execution states. 2. Dual text-code representation — connects skill retrieval with executable actions. 3. State-grounded dynamic retrieval mechanism — matches skills against both the task goal and the current webpage state.

Results

Experiments on WebArena across five domains show SGDR consistently outperforms strong baselines:

| Backbone | Average Success Rate | Gain over strongest baseline | |---|---|---| | GPT-4.1 | 37.5% | +10.6% | | Qwen3-4B | 24.3% | +10.0% |

Paper: arxiv.org/abs/2606.04391

Tags

#web-agents#online-skill-learning#retrieval#llm-agents#webarena#machine-learning#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177980844