ProofInfer: Generating Proof via Iterative Hierarchical Inference

Zichu Fei, Qi Zhang, Xin Zhou, Tao Gui, Xuanjing Huang


Abstract
Proof generation focuses on deductive reasoning: given a hypothesis and a set of theories, including some supporting facts and logical rules expressed in natural language, the model generates a proof tree indicating how to deduce the hypothesis from given theories.Current models with state-of-the-art performance employ the stepwise method that adds an individual node to the proof step-by-step.However, these methods actually focus on generating several proof paths rather than a whole tree.During generation, they focus on the most relevant areas of the currently generated node while neglecting the rest of the proof tree. To address this problem, we propose ProofInfer, which generates the proof tree via iterative hierarchical inference.At each step, ProofInfer adds the entire layer to the proof, where all nodes in this layer are generated simultaneously. Since the conventional autoregressive generation architecture cannot simultaneously predict multiple nodes, ProofInfer employs text-to-text paradigm.To this end, we propose a divide-and-conquer algorithm to encode the proof tree as the plain text without losing structure information.Experimental results show that ProofInfer significantly improves performance on several widely-used datasets.In addition, ProofInfer still performs well with data-limited, achieving comparable performance to the state-of-the-art model with about 40% of the training data.
Anthology ID:
2022.emnlp-main.747
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
10883–10892
Language:
URL:
https://aclanthology.org/2022.emnlp-main.747
DOI:
10.18653/v1/2022.emnlp-main.747
Bibkey:
Cite (ACL):
Zichu Fei, Qi Zhang, Xin Zhou, Tao Gui, and Xuanjing Huang. 2022. ProofInfer: Generating Proof via Iterative Hierarchical Inference. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 10883–10892, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
ProofInfer: Generating Proof via Iterative Hierarchical Inference (Fei et al., EMNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.emnlp-main.747.pdf