Greedy, Joint Syntactic-Semantic Parsing with Stack LSTMs
This addresses parsing efficiency and accuracy for NLP researchers and practitioners, though it is incremental as it builds on existing transition-based and joint parsing approaches.
The authors tackled the problem of joint syntactic-semantic parsing by developing a transition-based parser using stack LSTMs to represent algorithm state, achieving the best published performance on CoNLL 2008-9 English shared tasks among joint models.
We present a transition-based parser that jointly produces syntactic and semantic dependencies. It learns a representation of the entire algorithm state, using stack long short-term memories. Our greedy inference algorithm has linear time, including feature extraction. On the CoNLL 2008--9 English shared tasks, we obtain the best published parsing performance among models that jointly learn syntax and semantics.