UNO-DST: Leveraging Unlabelled Data in Zero-Shot Dialogue State Tracking

arXiv:2310.10492v2 Announce Type: replace
Abstract: Previous zero-shot dialogue state tracking (DST) methods only apply transfer learning, ignoring unlabelled data in the target domain. We transform zero-shot DST into few-shot DST by utilising such unlabelled data via joint and self-training methods. Our method incorporates auxiliary tasks that generate slot types as inverse prompts for main tasks, creating slot values during joint training. Cycle consistency between these two tasks enables the generation and selection of quality samples in unknown target domains for …

abstract apply arxiv cs.cl data dialogue domain few-shot generate self-training state tasks tracking training transfer transfer learning type types via zero-shot