DialogUSR

홈페이지 · 논문 2편

DialogUSR dataset covers 23 domains with a multi-step crowd-sourcing procedure. It comprises 36.7 Chinese characters by assembling 3.6 single-intent queries (including initial and follow-up queries) and is designed for dialogue utterance splitting and reformulation task. Source: DialogUSR: Complex Dialogue Utterance Splitting and Reformulation for Multiple Intent Detection Image Source: https://arxiv.org/pdf/2210.11279v1.pdf