JEF-Hinter: Leveraging Offline Knowledge for Improving Web Agents Adaptation
Abstract
Large language model (LLM) agents perform well in sequential decision-making tasks, but improving them on unfamiliar domains often requires costly online interactions or fine-tuning on large expert datasets. These strategies are impractical for closed-source models and expensive for open-source ones, with risks of catastrophic forgetting. Offline trajectories offer reusable knowledge, yet demonstration-based methods struggle because raw traces are long, noisy, and tied to specific tasks. We present Just-in-time Episodic Feedback Hinter (JEF-Hinter), an agentic system that distills offline traces into compact, context-aware hints. A zooming mechanism highlights decisive steps in long trajectories, capturing both strategies and pitfalls. Unlike prior methods, JEF-Hinter leverages both successful and failed trajectories, extracting guidance even when only failure data is available, while supporting parallelized hint generation and benchmark-independent prompting. At inference, a retriever selects relevant hints for the current state, providing targeted guidance with transparency and traceability. Experiments on MiniWoB++, WorkArena-L1, and WebArena-Lite show that JEF-Hinter consistently outperforms strong baselines, including human- and document-based hints.
Cite
@article{arxiv.2510.04373,
title = {JEF-Hinter: Leveraging Offline Knowledge for Improving Web Agents Adaptation},
author = {Hadi Nekoei and Aman Jaiswal and Patrice Bechard and Oleh Shliazhko and Orlando Marquez Ayala and Mathieu Reymond and Massimo Caccia and Alexandre Drouin and Sarath Chandar and Alexandre Lacoste},
journal= {arXiv preprint arXiv:2510.04373},
year = {2026}
}