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 2408.10455cs.CLcs.AI
收藏

IDEA:通过归纳、演绎和溯因增强大语言模型Agent的规则学习能力

IDEA: Enhancing the Rule Learning Ability of Large Language Model Agent through Induction, Deduction, and Abduction

Department of Computer Science University of Texas at Dallas(计算机科学系德克萨斯大学达拉斯分校)

Kaiyu He, Mian Zhang, Shuo Yan, Peilin Wu, Zhiyu Zoey Chen

摘要:

虽然大语言模型(LLM)的演绎和归纳推理已经得到了彻底的评估,他们在交互式环境中的整体规则学习的熟练程度仍然很少探索。我们引入RULEARN,一个新的基准来评估规则学习能力的LLM代理在交互式设置。在RULEARN中,智能体与模拟环境进行策略性交互,以收集观察结果,识别模式并解决复杂问题。为了提高LLM代理的规则学习能力,我们提出了IDEA,一种新的推理框架,集成了归纳,演绎和溯因的过程。IDEA代理通过溯因从有限的观察中生成初始假设,设计计划来验证这些假设或通过演绎利用它们来解决问题,并通过归纳改进先前的假设,动态建立和应用模仿人类规则学习行为的规则。我们对IDEA框架的评估,其中涉及五个有代表性的LLM,表明了基线的显着改进。此外,我们对人类参与者的研究揭示了人类和LLM之间规则学习行为的显着差异。我们相信我们的基准将作为一个有价值的和具有挑战性的资源,IDEA将为LLM代理的开发提供关键的见解,能够在现实世界的场景中进行类似人类的规则学习。我们的代码和数据是公开的。

英文摘要:

While large language models (LLMs) have been thoroughly evaluated for deductive and inductive reasoning, their proficiency in holistic rule learning in interactive environments remains less explored. We introduce RULEARN, a novel benchmark to assess the rule-learning abilities of LLM agents in interactive settings. In RULEARN, agents strategically interact with simulated environments to gather observations, discern patterns, and solve complex problems. To enhance the rule-learning capabilities for LLM agents, we propose IDEA, a novel reasoning framework that integrates the process of Induction, Deduction, and Abduction. The IDEA agent generates initial hypotheses from limited observations through abduction, devises plans to validate these hypotheses or leverages them to solve problems via deduction, and refines previous hypotheses through induction, dynamically establishing and applying rules that mimic human rule-learning behaviors. Our evaluation of the IDEA framework, which involves five representative LLMs, demonstrates significant improvements over the baseline. Furthermore, our study with human participants reveals notable discrepancies in rule-learning behaviors between humans and LLMs. We believe our benchmark will serve as a valuable and challenging resource, and IDEA will provide crucial insights for the development of LLM agents capable of human-like rule learning in real-world scenarios. Our code and data is publicly available.

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。

↑