Supertagging With LSTMs
نویسندگان
چکیده
In this paper we present new state-of-the-art performance on CCG supertagging and parsing. Our model outperforms existing approaches by an absolute gain of 1.5%. We analyze the performance of several neural models and demonstrate that while feed-forward architectures can compete with bidirectional LSTMs on POS tagging, models that encode the complete sentence are necessary for the long range syntactic information encoded in supertags.
منابع مشابه
An Empirical Exploration of Skip Connections for Sequential Tagging
In this paper, we empirically explore the effects of various kinds of skip connections in stacked bidirectional LSTMs for sequential tagging. We investigate three kinds of skip connections connecting to LSTM cells: (a) skip connections to the gates, (b) skip connections to the internal states and (c) skip connections to the cell outputs. We present comprehensive experiments showing that skip co...
متن کاملStacking or Supertagging for Dependency Parsing - What's the Difference?
Supertagging was recently proposed to provide syntactic features for statistical dependency parsing, contrary to its traditional use as a disambiguation step. We conduct a broad range of controlled experiments to compare this specific application of supertagging with another method for providing syntactic features, namely stacking. We find that in this context supertagging is a form of stacking...
متن کاملCCG Supertagging with a Recurrent Neural Network
Recent work on supertagging using a feedforward neural network achieved significant improvements for CCG supertagging and parsing (Lewis and Steedman, 2014). However, their architecture is limited to considering local contexts and does not naturally model sequences of arbitrary length. In this paper, we show how directly capturing sequence information using a recurrent neural network leads to f...
متن کاملHPSG Supertagging: A Sequence Labeling View
Supertagging is a widely used speed-up technique for deep parsing. In another aspect, supertagging has been exploited in other NLP tasks than parsing for utilizing the rich syntactic information given by the supertags. However, the performance of supertagger is still a bottleneck for such applications. In this paper, we investigated the relationship between supertagging and parsing, not just to...
متن کاملTOWARDS EFFICIENT STATISTICAL PARSING USING LEXICALIZED GRAMMATICAL INFORMATION by
For a long time, the goal of wide-coverage natural language parsers had remained elusive. Much progress has been made recently, however, with the development of lexicalized statistical models of natural language parsing. Although lexicalized tree adjoining grammar (TAG) is a lexicalized grammatical formalism whose development predates these recent advances, its application in lexicalized statis...
متن کامل