Semantic Categorization of Social Knowledge for Commonsense Question Answering

Gengyu Wang, Xiaochen Hou, Diyi Yang, Kathleen McKeown, Jing Huang


Abstract
Large pre-trained language models (PLMs) have led to great success on various commonsense question answering (QA) tasks in an end-to-end fashion. However, little attention has been paid to what commonsense knowledge is needed to deeply characterize these QA tasks. In this work, we proposed to categorize the semantics needed for these tasks using the SocialIQA as an example. Building upon our labeled social knowledge categories dataset on top of SocialIQA, we further train neural QA models to incorporate such social knowledge categories and relation information from a knowledge base. Unlike previous work, we observe our models with semantic categorizations of social knowledge can achieve comparable performance with a relatively simple model and smaller size compared to other complex approaches.
Anthology ID:
2021.sustainlp-1.10
Volume:
Proceedings of the Second Workshop on Simple and Efficient Natural Language Processing
Month:
November
Year:
2021
Address:
Virtual
Venues:
EMNLP | sustainlp
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
79–85
Language:
URL:
https://aclanthology.org/2021.sustainlp-1.10
DOI:
10.18653/v1/2021.sustainlp-1.10
Bibkey:
Cite (ACL):
Gengyu Wang, Xiaochen Hou, Diyi Yang, Kathleen McKeown, and Jing Huang. 2021. Semantic Categorization of Social Knowledge for Commonsense Question Answering. In Proceedings of the Second Workshop on Simple and Efficient Natural Language Processing, pages 79–85, Virtual. Association for Computational Linguistics.
Cite (Informal):
Semantic Categorization of Social Knowledge for Commonsense Question Answering (Wang et al., sustainlp 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.sustainlp-1.10.pdf
Code
 posuer/social-commonsense-knolwedge
Data
ARCATOMIC