arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.11473cs.CLcs.AIcs.SC
收藏

像程序执行者一样推理

Reasoning Like Program Executors

University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Sea AI Lab(Sea人工智能实验室);Microsoft Research Asia(微软亚洲研究院);Microsoft Azure AI(微软Azure人工智能部门)

Xinyu Pi, Qian Liu, Bei Chen, Morteza Ziyadi, Zeqi Lin, Qiang Fu, Yan Gao, Jian-Guang Lou, Weizhu Chen

摘要:

基于自然语言的推理是研究界长期以来的目标。然而,研究表明,现有的语言模型在推理方面存在不足。为了解决这个问题,我们提出了POET,一种新的推理预训练范式。POET通过对语言模型进行程序及其执行结果的预训练,使语言模型能够通过数据驱动的方法获取程序执行者所拥有的推理知识。POET在概念上很简单,可以由不同类型的程序执行器实例化。在本文中,我们展示了两个简单的实例POET-Math和POET-Logic,以及一个复杂的实例POET-SQL。在6个基准测试上的实验结果表明,POET可以显著提高自然语言推理中的模型性能,如数值推理、逻辑推理和多跳推理。POET为推理增强预训练打开了一扇新的大门,我们希望我们的分析能为未来的程序执行器推理研究提供启示。

英文摘要:

Reasoning over natural language is a long-standing goal for the research community. However, studies have shown that existing language models are inadequate in reasoning. To address the issue, we present POET, a novel reasoning pre-training paradigm. Through pre-training language models with programs and their execution results, POET empowers language models to harvest the reasoning knowledge possessed by program executors via a data-driven approach. POET is conceptually simple and can be instantiated by different kinds of program executors. In this paper, we showcase two simple instances POET-Math and POET-Logic, in addition to a complex instance, POET-SQL. Experimental results on six benchmarks demonstrate that POET can significantly boost model performance in natural language reasoning, such as numerical reasoning, logical reasoning, and multi-hop reasoning. POET opens a new gate on reasoning-enhancement pre-training, and we hope our analysis would shed light on the future research of reasoning like program executors.

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。

↑