LASER: LLM Agent with State-Space Exploration for Web Navigation

arXiv:2309.08172v2 Announce Type: replace
Abstract: Large language models (LLMs) have been successfully adapted for interactive decision-making tasks like web navigation. While achieving decent performance, previous methods implicitly assume a forward-only execution mode for the model, where they only provide oracle trajectories as in-context examples to guide the model on how to reason in the environment. Consequently, the model could not handle more challenging scenarios not covered in the in-context examples, e.g., mistakes, leading to sub-optimal performance. To address this issue, …

abstract agent arxiv context cs.cl decision examples exploration guide interactive language language models large language large language models llm llms making navigation oracle performance space state tasks type web web navigation