Fast and Accurate Shift-Reduce Constituent Parsing

By Muhua Zhu, Wenliang Chen, Min Zhang, Jingbo Zhu. 2013

221 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Recurrent Neural Network Grammars (2016, methodology): “…operators are capable of directly generating arbitrary tree structures rather than, e.g., assuming binarized trees, as is the case in much prior work that has used transition-based algorithms to produce phrase-structure trees (Sagae and Lavie, 2005; Zhang and Clark, 2011; Zhu et al., 2013).”
Grammar as a Foreign Language (2014, methodology): “Relatedly, [13] present excellent parsing results with a single left-to-right pass, but require a stack to explicitly delay making decisions and a parsing-specific transition strategy in order to achieve good parsing accuracies.”
Learning to Reason: End-to-End Module Networks for Visual Question Answering (2017, background): “The earliest work on NMNs [3] used fixed rule-based layouts generated from dependency parses [27].”
Recent Trends in Deep Learning Based Natural Language Processing (2017, background): “[140] Feature-based transition parsing 91.”
Modeling Relationships in Referential Expressions with Compositional Modular Networks (2016, methodology): “In our full model, the language representation is end-to-end optimized with other parts, while it is hard to jointly optimize an external language parser like [33] for this task.”
Parsing as Pretraining (2020, methodology): “Transition-based systems (Zhu et al. 2013) used templates of hand-crafted features suiting the task at hand.”
Feature Optimization for Constituent Parsing via Neural Networks (2015, methodology): “In the last few years, however, with the use of effective parsing strategies, approximate inference algorithms, and more efﬁcient training methods, discriminative models began to surpass the generative models (Carreras et al., 2008; Zhu et al., 2013; Wang and Xue, 2014).”
Exploiting Syntactic Features in a Parsed Tree to Improve End-to-End TTS (2019, unknown): “In recent years, research on parsing technologies has been greatly advanced and many new parsing algorithms are available , like Probabilistic Context Free Grammar (PCFG) [14], factored parser [15], Shift-Reduce Parser [16] and Straight to the Tree [17].”
Neural CRF Parsing (2015, methodology): “We further compare to the shift-reduce parser of Zhu et al. (2013), which uses unlabeled data in the form of Brown clusters.”
Recent Trends in Deep Learning Based Natural Language Processing [Review Article] (2018, background): “[134] fEATuRE-BASED TRANSiTioN PARSiNg 91.”
In-Order Transition-based Constituent Parsing (2017, unknown): “(2013) 90.4 Zhu et al. (2013) 90.4 Vinyals et al. (2015) 90.7 Watanabe and Sumita (2015) 90.7 Shindo et al. (2012) 91.1 Durrett and Klein (2015) 91 2016; Watanabe and Sumita, 2015) and the state-of-the-art hybrid parsers (Durrett and Klein, 2015; Liu and Zhang, 2017), achieving the state-of-the-art…”
Joint POS Tagging and Transition-based Constituent Parsing in Chinese with Non-local Features (2014, methodology): “To resolve this conﬂict, we integrate the debinariza-tion process into the parsing process, i.e., when a (Collins and Koo, 2005) Semi-supervised word cluster features have been successfully applied to many NLP tasks (Miller et al., 2004; Koo et al., 2008; Zhu et al., 2013).”
Parsing as Language Modeling (2016, methodology): “To obtain silver trees, we parse the entire section of the New York Times (NYT) of the ﬁfth Gigaword (Parker et al., 2011) with a product of eight Berkeley parsers (Petrov, 2010) 2 and ZPar (Zhu et al., 2013) and select 24 million trees on which both parsers agree (Li et al., 2014).”
Improving Neural Parsing by Disentangling Model Combination and Reranking Effects (2017, methodology): “This silver dataset was obtained by parsing the entire New York Times section of the ﬁfth Gigaword corpus using a product of eight Berkeley parsers (Petrov, 2010) and ZPar (Zhu et al., 2013), then retaining 24 million sentences on which both parsers agreed.”
Mandarin Prosodic Phrase Prediction based on Syntactic Trees (2016, methodology): “As the shift-reduce constituency parser is much faster than the previous one with competitive accuracy, we select the shift-reduce parser and train it using the same data set-up as [19].”