arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.14548cs.AI
收藏

超越效用的LLM智能体:一种开放式视角

LLM Agents Beyond Utility: An Open-Ended Perspective

ETH Zurich(苏黎世联邦理工学院)

Asen Nachkov, Xi Wang, Luc Van Gool

AI总结:

本文采用开放式实验设置,增强预训练LLM智能体以自主生成任务、积累知识,定性评估其能力,发现其能遵循复杂指令并自创任务,但存在提示敏感、重复生成及缺乏自我表征等局限,为未来智能体训练指明方向。

摘要:

最近的LLM智能体在思维链推理和函数调用方面取得了巨大进展。随着其能力的增长,一个重要的问题出现了:这种软件能否不仅代表一个智能的问题解决工具,而且代表一个具有自身权利的实体,能够规划、设计即时任务,并朝着更广泛、更模糊的目标进行推理?为了研究这个问题,我们采用了一种开放式的实验设置,在该设置中,我们增强了一个预训练的LLM智能体,使其能够生成自己的任务、积累知识,并与其环境进行广泛交互。我们定性地研究了由此产生的开放式智能体。它能够可靠地遵循复杂的多步骤指令,跨运行存储和重用信息,并提出和解决自己的任务,尽管它仍然对提示设计敏感,容易产生重复的任务生成,并且无法形成自我表征。这些发现既展示了将预训练的LLM适应于开放性的前景,也展示了其当前的局限性,并为未来训练智能体管理记忆、高效探索和追求抽象长期目标指明了方向。

英文摘要:

Recent LLM agents have made great use of chain of thought reasoning and function calling. As their capabilities grow, an important question arises: can this software represent not only a smart problem-solving tool, but an entity in its own right, that can plan, design immediate tasks, and reason toward broader, more ambiguous goals? To study this question, we adopt an open-ended experimental setting where we augment a pretrained LLM agent with the ability to generate its own tasks, accumulate knowledge, and interact extensively with its environment. We study the resulting open-ended agent qualitatively. It can reliably follow complex multi-step instructions, store and reuse information across runs, and propose and solve its own tasks, though it remains sensitive to prompt design, prone to repetitive task generation, and unable to form self-representations. These findings illustrate both the promise and current limits of adapting pretrained LLMs toward open-endedness, and point to future directions for training agents to manage memory, explore productively, and pursue abstract long-term goals.

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。

↑