@InProceedings{fried-stern-klein:2017:Short,
  author    = {Fried, Daniel  and  Stern, Mitchell  and  Klein, Dan},
  title     = {Improving Neural Parsing by Disentangling Model Combination and Reranking Effects},
  booktitle = {Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)},
  month     = {July},
  year      = {2017},
  address   = {Vancouver, Canada},
  publisher = {Association for Computational Linguistics},
  pages     = {161--166},
  abstract  = {Recent work has proposed several generative neural models for constituency
	parsing that achieve state-of-the-art results. Since direct search in these
	generative models is difficult, they have primarily been used to rescore
	candidate outputs from base parsers in which decoding is more straightforward. 
	We first present an algorithm for direct search in these generative models.  We
	then demonstrate that the rescoring results are at least partly due to implicit
	model combination rather than reranking effects.  Finally, we show that
	explicit model combination can improve performance even further, resulting in
	new state-of-the-art numbers on the PTB of 94.25 F1 when training only on gold
	data and 94.66 F1 when using external data.},
  url       = {http://aclweb.org/anthology/P17-2025}
}