Representation Learning for Text-level Discourse Parsing

Abstract

Text-level discourse parsing is notoriously difficult, as distinctions between discourse relations require subtle semantic judgments that are not easily captured using standard features. In this paper, we present a representation learning approach, in which we transform surface features into a latent space that facilitates RST discourse parsing. By combining the machinery of large-margin transition-based structured prediction with representation learning, our method jointly learns to parse discourse while at the same time learning a discourse-driven projection of surface features. The resulting shift-reduce discourse parser obtains substantial improvements over the previous state-of-the-art in predicting relations and nuclearity on the RST Treebank.

Cross-lingual RST Discourse Parsing

“…a Scores reported from (Li et al, 2014), and DPLP (Ji and Eisenstein, 2014). b For Brazilian Portuguese, inter-annotator agreement scores are only available for the CST-news corpus ; For Spanish, only precision scores are reported ; For Basque, the scores reported are different (Iruskieta et al, 2015).…”

“…Note that we do not use all the words in the EDUs as features, contrary to (Li et al, 2014;Ji and Eisenstein, 2014). Our only word features are the words in the head set and at the boundaries, thus 7 words per EDU.…”

“…Rhetorical Structure Theory (RST) (Mann and Thompson, 1988) is a prominent linguistic theory of discourse structures, in which texts are analyzed as constituency trees, such as the one in Figure 1. This theory guided the annotation of the RST Discourse Treebank (RST-DT) for English, from which several textlevel discourse parsers have been induced (Hernault et al, 2010;Joty et al, 2012;Feng and Hirst, 2014;Li et al, 2014;Ji and Eisenstein, 2014). Such parsers have proven to be useful for various downstream applications (Daumé III and Marcu, 2009;Burstein et al, 2003;Higgins et al, 2004;Thione et al, 2004;Sporleder and Lapata, 2005;Taboada and Mann, 2006;Louis et al, 2010;Bhatia et al, 2015).…”

Discourse Mode Identification in Essays

“…One of the advantages is the ability of automatic representation learning. Representing words or relations with continuous vectors (Mikolov et al, 2013;Ji and Eisenstein, 2014) embeds semantics in the same space, which benefits alleviating the data sparseness problem and enables end-to-end and multi-task learning. Recurrent neural networks (RNNs) (Graves, 2012) and the variants like Long Short-Term Memory (LSTM) (Hochreiter and Schmidhuber, 1997) and Gated Recurrent (GRU) (Cho et al, 2014) neural networks show good performance for capturing long distance dependencies on tasks like Named Entity Recognition (NER) (Chiu and Nichols, 2016;Ma and Hovy, 2016), dependency parsing (Dyer et al, 2015) and semantic composition of documents (Tang et al, 2015).…”

Does syntax help discourse segmentation? Not so much

“…Most work on discourse parsing has focused on English and on the RST-DT (Ji and Eisenstein, 2014;Feng and Hirst, 2014;Li et al, 2014;Joty et al, 2013), and so discourse segmentation (Xuan Bach et al, 2012;Fisher and Roark, 2007;Subba and Di Eugenio, 2007). And while discourse parsing is a document level task, discourse segmentation is done at the sentence level, assuming that sentence boundaries are known.…”