Dec, 2023

LLM 强化的分层语言代理用于实时的人工智能与人类协调

TL;DRAI agents using Large Language Models (LLMs) have revolutionized human-AI coordination in various tasks, but their high inference latency makes them unsuitable for real-time applications; this paper proposes a Hierarchical Language Agent (HLA) that combines a proficient LLM, a lightweight LLM, and a reactive policy to provide strong reasoning abilities in real-time execution, showcasing superior cooperation abilities, faster responses, and more consistent language communications compared to other agents.