Joint RNN-Based Greedy Parsing and Word Composition
This work addresses parsing efficiency for NLP applications, but it is incremental as it builds on existing greedy and neural methods.
The paper tackles syntactic parsing by introducing a greedy neural parser that jointly learns parsing and word composition, achieving F1 performance comparable to existing parsers while being faster.
This paper introduces a greedy parser based on neural networks, which leverages a new compositional sub-tree representation. The greedy parser and the compositional procedure are jointly trained, and tightly depends on each-other. The composition procedure outputs a vector representation which summarizes syntactically (parsing tags) and semantically (words) sub-trees. Composition and tagging is achieved over continuous (word or tag) representations, and recurrent neural networks. We reach F1 performance on par with well-known existing parsers, while having the advantage of speed, thanks to the greedy nature of the parser. We provide a fully functional implementation of the method described in this paper.