Back to the Future: Bidirectional Information Decoupling Network for Multi-turn Dialogue Modeling
This work addresses the challenge of understanding human dialogues for machines, with incremental improvements in encoding methods for dialogue-related applications.
The paper tackles the problem of multi-turn dialogue modeling by proposing a bidirectional information decoupling network (BiDeN) to better capture temporal characteristics, achieving improved performance across various downstream tasks.
Multi-turn dialogue modeling as a challenging branch of natural language understanding (NLU), aims to build representations for machines to understand human dialogues, which provides a solid foundation for multiple downstream tasks. Recent studies of dialogue modeling commonly employ pre-trained language models (PrLMs) to encode the dialogue history as successive tokens, which is insufficient in capturing the temporal characteristics of dialogues. Therefore, we propose Bidirectional Information Decoupling Network (BiDeN) as a universal dialogue encoder, which explicitly incorporates both the past and future contexts and can be generalized to a wide range of dialogue-related tasks. Experimental results on datasets of different downstream tasks demonstrate the universality and effectiveness of our BiDeN.