Improved Inference for Unlexicalized Parsing

By Slav Petrov, Daniel Klein. 2007

722 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, background): “lts on PTB x23 (D=discriminative, G=generative, S=semisupervised). Model type F 1 Henderson (2004) D 89.4 Socher et al. (2013a) D 90.4 Zhu et al. (2013) D 90.4 Vinyals et al. (2015) – WSJ only D 90.5 Petrov and Klein (2007) G 90.1 Bod (2003) G 90.7 Shindo et al. (2012) – single G 91.1 Shindo et al. ”
Unbounded Dependency Recovery for Parser Evaluation (2009, background): “Statistical parsers are now obtaining Parseval scores of over 90% on the WSJ section of the Penn Treebank (Bod, 2003; Petrov and Klein, 2007; Huang, 2008; Carreras et al., 2008).”
What Do Recurrent Neural Network Grammars Learn About Syntax? (2016, background): “erminals, and non-terminal symbols; our action history is quite close to this, with each NT(X) capturing a left parenthesis and X nonterminal, and each REDUCE capturing a right parenthesis. Model F 1 Petrov and Klein (2007) 90.1 Shindo et al. (2012) 91.1 Zhu et al. (2013)y 91.3 McClosky et al. (2006”
EU-BRIDGE MT: Combined Machine Translation (2014, methodology): “A string-to-tree system trained with target-side syntactic annotation obtained with the German Berkeley Parser (Petrov and Klein, 2007; Petrov and Klein, 2008).”
Self-Training PCFG Grammars with Latent Annotations Across Languages (2009, background): “Take Chinese for example; there have been several attempts to develop accurate parsers for Chinese (Bikel and Chiang, 2000; Levy and Manning, 2003; Petrov and Klein, 2007), but the state-of-the-art performance, around 83% F measure on Penn Chinese Treebank (achieved by the Berkeley parser (Petrov an”
A Latent Variable Model for Generative Dependency Parsing (2007, methodology): “Recently several latent variable models for constituent parsing have been proposed [23, 28, 41, 43, 39, 25, 40].”
Word Segmentation, Unknown-word Resolution, and Morphological Agreement in a Hebrew Parsing System (2013, methodology): “Their work triggered investigations in automatic grammar reﬁnement and state-splitting (Matsuzaki, Miyao, and Tsujii 2005; Prescher 2005), which was then perfected in work by Petrov and colleagues (Petrov et al. 2006; Petrov and Klein 2007; Petrov 2009).”
Syntactic Processing Using the Generalized Perceptron and Beam Search (2011, background): “Petrov and Klein (2007) reported LR and LP of 85 .”
Edinburgh’s Syntax-Based Systems at WMT 2014 (2014, methodology): “Contrastive systems were built with other parsers: BitPar (Schmid, 2004), the German Stanford Parser (Rafferty and Manning, 2008), and the German Berkeley Parser (Petrov and Klein, 2007; Petrov and Klein, 2008).”
Bleaching Text: Abstract Features for Cross-lingual Gender Prediction (2018, methodology): “They are mostly motivated by intuition and inspired by prior work, like the use of shape features from NER and parsing (Petrov and Klein, 2007; Schnabel and Schütze, 2014; Plank et al., 2016; Limsopatham and Collier, 2016): • Frequency Each word is presented as its binned frequency in the training…”
From symbolic to sub-symbolic information in question classification (2011, methodology): “We use the Berkeley Parser (Petrov and Klein 2007) for the purpose of parsing questions, trained on the QuestionBank (Judge et al. 2006), a treebank of 4,000 parse-annotated questions.”
Self-Training with Products of Latent Variable Grammars (2010, methodology): “8 in F score) than the product model in Petrov (2010). This is made possible by self-training on the output of a high quality product model.”
Collective Generation of Natural Image Descriptions (2012, positive): “All content types are pre-computed on the million database photos, and caption parsing is performed using the Berkeley PCFG parser (Petrov et al., 2006; Petrov and Klein, 2007).”
A Minimal Span-Based Neural Constituency Parser (2017, background): “Many early successful approaches to constituency parsing focused on rich modeling of correlations in the output space, typically by engineering proabilistic context-free grammars with state spaces enriched to capture long-distance dependencies and lexical phenomena (Collins, 2003; Klein and Manning,”
Inducing Tree-Substitution Grammars (2010, methodology): “Unlike Context Free Grammars (CFGs) a 3.”
Improved Neural Machine Translation with a Syntax-Aware Encoder and Decoder (2017, methodology): “parse the Chinese sentences with the Berkeley Parser5 (Petrov and Klein, 2007) and binarize the resulting trees following Zhang and Clark (2009).”