Great~Truths~are ~Always ~Simple: A Rather Simple Knowledge Encoder for Enhancing the Commonsense Reasoning Capacity of Pre-Trained Models
Jinhao Jiang, Kun Zhou, Ji-Rong Wen, Xin Zhao
Abstract
Commonsense reasoning in natural language is a desired ability of artificial intelligent systems. For solving complex commonsense reasoning tasks, a typical solution is to enhance pre-trained language models (PTMs) with a knowledge-aware graph neural network (GNN) encoder that models a commonsense knowledge graph (CSKG).Despite the effectiveness, these approaches are built on heavy architectures, and can’t clearly explain how external knowledge resources improve the reasoning capacity of PTMs. Considering this issue, we conduct a deep empirical analysis, and find that it is indeed relation features from CSKGs (but not node features) that mainly contribute to the performance improvement of PTMs. Based on this finding, we design a simple MLP-based knowledge encoder that utilizes statistical relation paths as features. Extensive experiments conducted on five benchmarks demonstrate the effectiveness of our approach, which also largely reduces the parameters for encoding CSKGs.Our codes and data are publicly available at https://github.com/RUCAIBox/SAFE.
BibTeX
@inproceedings{jiang-etal-2022-great,
title = "$Great~Truths~are ~Always ~Simple:$ A Rather Simple Knowledge Encoder for Enhancing the Commonsense Reasoning Capacity of Pre-Trained Models",
author = "Jiang, Jinhao and
Zhou, Kun and
Wen, Ji-Rong and
Zhao, Xin",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2022",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-naacl.131/",
doi = "10.18653/v1/2022.findings-naacl.131",
pages = "1730--1741"
}