q2d: Turning Questions into Dialogs to Teach Models How to Search

The Hebrew University of Jerusalem, Google Research
arXiv


q2d is an automatic data generation pipeline that generates information-seeking dialogs from questions. q2d effectively replaces human-annotated data for training query-generation models and creates high-quality training and evaluation data across multiple domains.

q2d's auto-generated dialogs enable query generation models to adapt and improve for specific dialog styles, creating labeled datasets for training and evaluation.
T5 model predictions above/below the line show the impact of fine-tuning on MuSiQue dialogs.



BibTeX

@article{bitton2023q2d,
  title={q2d: Turning Questions into Dialogs to Teach Models How to Search},
  author={Bitton, Yonatan and Cohen-Ganor, Shlomi and Hakimi, Ido and Lewenberg, Yoad and Aharoni, Roee and Weinreb, Enav},
  journal={arXiv preprint arXiv:2304.14318},
  year={2023}
}