Multi-Granularity Information Interaction Framework for Incomplete Utterance Rewriting
This work improves dialogue systems by enabling more accurate rewriting of incomplete utterances, though it appears incremental as it builds on existing IUR methods.
The paper tackles the problem of Incomplete Utterance Rewriting by addressing failures in capturing important word sources and avoiding irrelevant words, resulting in a framework that outperforms state-of-the-art models on benchmark datasets Restoration-200K and CANAND.
Recent approaches in Incomplete Utterance Rewriting (IUR) fail to capture the source of important words, which is crucial to edit the incomplete utterance, and introduce words from irrelevant utterances. We propose a novel and effective multi-task information interaction framework including context selection, edit matrix construction, and relevance merging to capture the multi-granularity of semantic information. Benefiting from fetching the relevant utterance and figuring out the important words, our approach outperforms existing state-of-the-art models on two benchmark datasets Restoration-200K and CANAND in this field. Code will be provided on \url{https://github.com/yanmenxue/QR}.