elBERto: Self-supervised commonsense learning for question answering

• Published in: Knowledge-based systems 2022-12, Vol.258, p.109964, Article 109964
• Main Authors: Zhan, Xunlin, Li, Yuan, Dong, Xiao, Liang, Xiaodan, Hu, Zhiting, Carin, Lawrence
• Format: Article
• Language: English
• Subjects: Commonsense reasoning; Question answering; Self-supervised learning
• ISSN: 0950-7051; 1872-7409
• DOI: 10.1016/j.knosys.2022.109964

Commonsense question answering requires reasoning about everyday situations and causes and effects implicit in context. Typically, existing approaches first retrieve external evidence and then perform commonsense reasoning using these evidence. In this paper, we propose a Self-supervised Bidirectional Encoder Representation Learning of Commonsense (elBERto) framework, which is compatible with off-the-shelf QA model architectures. The framework comprises five self-supervised tasks to force the model to fully exploit the additional training signals from contexts containing rich commonsense. The tasks include a novel Contrastive Relation Learning task to encourage the model to distinguish between logically contrastive contexts, a new Jigsaw Puzzle task that requires the model to infer logical chains in long contexts, and three classic self-supervised learning(SSL) tasks to maintain pre-trained models’ language encoding ability. On the representative WIQA, CosmosQA, and ReClor datasets, elBERto outperforms all other methods using the same backbones and the same training set, including those utilizing explicit graph reasoning and external knowledge retrieval. Moreover, elBERto achieves substantial improvements on out-of-paragraph and no-effect questions where simple lexical similarity comparison does not help, indicating that it successfully learns commonsense and is able to leverage it when given dynamic context. •A self-supervised learning method is proposed to solve commonsense QA task.•A novel CRL task and a JP task are presented to facilitate commonsense learning.•Experiments verify that elBERTo significantly outperforms the existing algorithms.