Language models have been shown to perform remarkably well on a wide range of natural language processing tasks. In this paper, we propose LEAP, a novel system that uses language models to perform multi-step logical reasoning and incorporates explicit planning into the inference procedure. Explicit planning enables the system to make more informed reasoning decisions at each step by looking ahead into their future effects. Moreover, we propose a training strategy that safeguards the planning process from being led astray by spurious features. Our full system significantly outperforms other competing methods on multiple standard datasets. When using small T5 models as its core selection and deduction components, our system performs competitively compared to GPT-3 despite having only about 1B parameters (i.e., 175 times smaller than GPT-3). When using GPT-3.5, it significantly outperforms chain-of-thought prompting on the challenging PrOntoQA dataset. We have conducted extensive empirical studies to demonstrate that explicit planning plays a crucial role in the system's performance.
@article{arxiv.2303.15714,
title = {Explicit Planning Helps Language Models in Logical Reasoning},
author = {Hongyu Zhao and Kangrui Wang and Mo Yu and Hongyuan Mei},
journal= {arXiv preprint arXiv:2303.15714},
year = {2023}
}
Comments
EMNLP 2023 camera-ready; updated results on PrOntoQA with code bugs fixed