Differences

This shows you the differences between two versions of the page.

--- nlp:dependency_parsing [2021/01/07 09:59] – jmflanig
+++ nlp:dependency_parsing [2023/06/15 07:36] (current) – external edit 127.0.0.1
@@ Line 1: / Line 1: @@
 ====== Dependency Parsing ======
+See also [[http://nlpprogress.com/english/dependency_parsing.html|NLP Progress - Dependency Parsing]]
 ===== Graph-Based Dependency Parsing =====
 ==== Early papers (pre-neural): ====
   * The paper that started graph-based dependency parsing: [[https://www.aclweb.org/anthology/H05-1066.pdf|McDonald et al 2005 - Non-projective Dependency Parsing using Spanning Tree Algorithms]]
+  * [[https://www.aclweb.org/anthology/P10-1001.pdf|Koo & Collins 2010 - Efficient Third-order Dependency Parsers]]
 ==== Neural graph-based dependency parsing ====
-  * [[https://www.aclweb.org/anthology/Q16-1023.pdf|Kiperwasser & Goldberg 2016 - Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations]]  First neural graph-based parser.  One of the early papers to advocate using BiLSTMs as features in NLP models.
+  * [[https://aclanthology.org/P15-1031.pdf|Pei et al 2015 - An Effective Neural Network Model for Graph-based Dependency Parsing]]
+  * [[https://www.aclweb.org/anthology/Q16-1023.pdf|Kiperwasser & Goldberg 2016 - Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations]]  An early neural graph-based parser.  One of the early papers to advocate using BiLSTMs as features in NLP models.
+  * [[https://arxiv.org/pdf/1606.01280.pdf|Zhang et al 2016 - Dependency Parsing as Head Selection]] Applies MST and Eisner's algorithm to a neural network trained to predict the head of each word.
   * [[https://www.aclweb.org/anthology/P16-1218.pdf|Wang & Chang 2016 - Graph-based Dependency Parsing with Bidirectional LSTM]] Similar and concurrent (but slightly later) than Kiperwasser & Goldberg
   * [[https://arxiv.org/pdf/1611.01734.pdf|Dozat & Manning 2017 - Deep Biaffine Attention for Neural Dependency Parsing]]  Popular neural graph-based parser, often used as a baseline model.
+  * [[https://arxiv.org/pdf/1911.03875.pdf|Mrini et al 2019 - Rethinking Self-Attention: Towards Interpretability in Neural Parsing]]
+  * [[https://arxiv.org/pdf/1807.01745.pdf|Gómez-Rodríguez et al 2018 - Global Transition-based Non-projective Dependency Parsing]]
+  * **[[https://arxiv.org/pdf/2010.02550.pdf|Zmigrod et al 2020 - Please Mind the Root: Decoding Arborescences for Dependency Parsing]]**
 === Neural graph-based models with higher-order features ===
   * [[https://www.aclweb.org/anthology/P19-1237.pdf|Ji et al 2019 - Graph-based Dependency Parsing with Graph Neural Networks]]
+  * [[https://arxiv.org/pdf/2005.00975.pdf|Zhang et al 2020 - Efficient Second-Order TreeCRF for Neural Dependency Parsing]]
 ===== Transition-Based Dependency Parsing =====
-  * [[https://www.aclweb.org/anthology/D14-1082.pdf|Chen & Manning 2014 - A Fast and Accurate Dependency Parser using Neural Networks]]  The first neural dependency parser
+  * [[https://www.aclweb.org/anthology/D14-1082.pdf|Chen & Manning 2014 - A Fast and Accurate Dependency Parser using Neural Networks]]  The first neural dependency parser.
+  * [[https://arxiv.org/pdf/1805.01087.pdf|Ma et al 2018 - Stack-Pointer Networks for Dependency Parsing]]  Has a good overview of existing parsers at the time.
+  * [[https://arxiv.org/pdf/1804.06004.pdf|Keith et al 2018 - Monte Carlo Syntax Marginals for Exploring and Using Dependency Parses]]
+===== Unsupervised Dependency Parsing ======
+  * **Overviews**
+    * Concise summary of prior work in [[https://www.aclweb.org/anthology/2020.tacl-1.15.pdf|Nishida 2020]].
+    * [[https://arxiv.org/pdf/2010.01535.pdf|Han et al 2020 - A Survey of Unsupervised Dependency Parsing]]
+  * **Key papers**
+    * [[https://www.aclweb.org/anthology/P04-1061.pdf|Klein & Manning 2001 - Corpus-Based Induction of Syntactic Structure: Models of Dependency and Constituency]] Dependency Model with Valence (DMV). What made Dan Klein famous
+    * [[https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/938/smith.2sp.thesis06.pdf?sequence=1&isAllowed=y|Noah Smith 2006 - Novel Estimation Methods for Unsupervised Discovery of Latent Structure in Natural Language Text]] What made Noah Smith famous
+    * [[https://www.aclweb.org/anthology/N09-1009.pdf|Cohen & Smith 2009 - Shared Logistic Normal Distributions for Soft Parameter Tying in Unsupervised Grammar Induction]] What made Shay Cohen famous
+===== Software =====
+  * [[https://stanfordnlp.github.io/stanza/|Stanza]] This parser is very good, better than Stanford Core NLP, NLTK, or other tools
+===== People =====
+  * [[https://scholar.google.com/citations?user=faXAgZQAAAAJ&hl=en|Zhenghua Li]]
+===== Related Pages =====
+  * [[Constituency Parsing]]
+  * [[Semantic Dependencies]] (Semantic Dependency Parsing)