Sluice resolution without hand-crafted features over brittle syntax trees

Abstract

Sluice resolution in English is the problem of finding antecedents of wh-fronted ellipses. Previous work has relied on handcrafted features over syntax trees that scale poorly to other languages and domains; in particular, to dialogue, which is one of the most interesting applications of sluice resolution. Syntactic information is arguably important for sluice resolution, but we show that multi-task learning with partial parsing as auxiliary tasks effectively closes the gap and buys us an additional 9% error reduction over previous work. Since we are not directly relying on features from partial parsers, our system is more robust to domain shifts, giving a 26% error reduction on embedded sluices in dialogue.

BibTeX

@inproceedings{ronning2018sluice,
  title={{Sluice Resolution without Hand-Crafted Features over Brittle Syntax Trees}},
  author={R{\o}nning, Ola and Hardt, Daniel and S{\o}gaard, Anders},
  booktitle={Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)},
  pages={236--241},
  year={2018}
}