Who Speaks Like a Style of Vitamin: Towards Syntax-Aware Dialogue Summarization Using Multi-Task Learning
Seolhwa Lee, Kisu Yang, Chanjun Park, João Sedoc, Heuiseok Lim · IEEE Access · 2021
Abstractive dialogue summarization is a challenging task for several reasons. First, most of the important pieces of information in a conversation are scattered across utterances through multi-party interactions with different textual styles. Second, dialogues are often informal structures, wherein different individuals express personal perspectives, unlike text summarization, tasks that usually target formal documents such as news articles. To address these issues, we focused on the association between utterances from individual speakers and unique syntactic structures. Speakers have unique textual styles that can contain linguistic information, such as voiceprint. To do this, we approach ad-hoc analysis to explore speakers’ text styles and constructed a syntax-aware model by leveraging linguistic information (i.e., POS tagging), which alleviates the above issues by inherently distinguishing sentences uttered from individual speakers. Our approach allows for both data and model-centric investigation. Also, we employed multi-task learning of both syntax-aware information and dialogue summarization. To the best of our knowledge, our approach is the first method to apply multi-task learning to the dialogue summarization task. Experiments on a SAMSum corpus (a large-scale dialogue summarization corpus) demonstrated that our method improved upon the vanilla model. Consequently, we found that our efforts of syntax-aware approach have been reflected by the model.