arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.02053cs.AIcs.HC
收藏

经济实惠的生成式智能体

Affordable Generative Agents

Tencent(腾讯)

Yangbin Yu, Qin Zhang, Junyou Li, Qiang Fu, Deheng Ye

AI总结:

本文提出AGA框架,通过用学习策略替代重复LLM推理、建模社会关系及压缩对话信息,实现低成本且可信的智能体交互。

摘要:

大型语言模型(LLMs)的出现显著推进了可信交互智能体的模拟。然而,维持智能体长时间交互的高昂成本对部署可信的基于LLM的智能体构成了挑战。因此,在本文中,我们开发了Affordable Generative Agents (AGA)(经济实惠的生成式智能体),这是一个能够在智能体-环境和智能体间两个层面上生成可信且低成本交互的框架。具体而言,对于智能体-环境交互,我们用学习到的策略替代重复的LLM推理;而对于智能体间交互,我们建模智能体之间的社会关系并压缩辅助对话信息。在多个环境上的大量实验表明了我们提出的框架的有效性和效率。此外,我们深入探讨了LLM智能体中涌现的可信行为机制,证明智能体在固定环境中只能生成有限的行为,基于此,我们理解了促进涌现交互行为的方法。我们的代码已在以下网址公开:https://github.com/AffordableGenerativeAgents/Affordable-Generative-Agents。

英文摘要:

The emergence of large language models (LLMs) has significantly advanced the simulation of believable interactive agents. However, the substantial cost on maintaining the prolonged agent interactions poses challenge over the deployment of believable LLM-based agents. Therefore, in this paper, we develop Affordable Generative Agents (AGA), a framework for enabling the generation of believable and low-cost interactions on both agent-environment and inter-agents levels. Specifically, for agent-environment interactions, we substitute repetitive LLM inferences with learned policies; while for inter-agent interactions, we model the social relationships between agents and compress auxiliary dialogue information. Extensive experiments on multiple environments show the effectiveness and efficiency of our proposed framework. Also, we delve into the mechanisms of emergent believable behaviors lying in LLM agents, demonstrating that agents can only generate finite behaviors in fixed environments, based upon which, we understand ways to facilitate emergent interaction behaviors. Our code is publicly available at: https://github.com/AffordableGenerativeAgents/Affordable-Generative-Agents.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑