Self-Training PCFG Grammars with Latent Annotations Across Languages

We investigate the effectiveness of self-training PCFG grammars with latent annotations (PCFG-LA) for parsing languages with different amounts of labeled training data. Compared to Charniak's lexicalized parser, the PCFG-LA parser was more effectively adapted to a language for which parsing has been less well developed (i.e., Chinese) and benefited more from self-training. We show for the first time that self-training is able to significantly improve the performance of the PCFG-LA parser, a single generative parser, on both small and large amounts of labeled training data. Our approach achieves state-of-the-art parsing accuracies for a single parser on both English (91.5%) and Chinese (85.2%).

• Publication year

  2009

• Venue

  Conference on Empirical Methods in Natural Language Processing

• Publication date

  2009-08-06

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.3115/1699571.1699621
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Chinese Statistical Parsing

  2009cited by this paper
• Optimizing Chinese Word Segmentation for Machine Translation Performance

  2008cited by this paper
• Semi-Supervised Convex Training for Dependency Parsing

  2008cited by this paper
• Simple Semi-supervised Dependency Parsing

  2008cited by this paper
• Forest Reranking: Discriminative Parsing with Non-Local Features

  2008cited by this paper
• Sparse Multi-Scale Grammars for Discriminative Latent Variable Parsing

  2008cited by this paper
• When is Self-Training Effective for Parsing?

  2008cited by this paper
• Mandarin Part-of-Speech Tagging and Discriminative Reranking

  2007cited by this paper
• Improved Inference for Unlexicalized Parsing

  2007influential reference
• Self-Training for Enhancement and Domain Adaptation of Statistical Parsers Trained on Small Datasets

  2007cited by this paper
• Effective Self-Training for Parsing

  2006influential reference
• Learning Accurate, Compact, and Interpretable Tree Annotation

  2006influential reference
• Tregex and Tsurgeon: tools for querying and manipulating tree data structures

  2006cited by this paper
• Coarse-to-Fine n-Best Parsing and MaxEnt Discriminative Reranking

  2005cited by this paper
• Probabilistic CFG with Latent Annotations

  2005cited by this paper
• The Penn Chinese TreeBank: Phrase structure annotation of a large corpus

  2005influential reference
• Bootstrapping statistical parsers from small datasets

  2003cited by this paper
• Head-Driven Statistical Models for Natural Language Parsing

  2003cited by this paper
• Is it Harder to Parse Chinese, or the Chinese Treebank?

  2003influential reference
• A Maximum-Entropy-Inspired Parser

  2000cited by this paper
• Two Statistical Parsing Models Applied to the Chinese Treebank

  2000cited by this paper
• Statistical Parsing with a Context-Free Grammar and Word Statistics

  1997cited by this paper

• Transforming Human-Machine Interaction: Generative AI Virtual Asst

  2024cites this paper
• Curriculum-Style Fine-Grained Adaption for Unsupervised Cross-Lingual Dependency Transfer

  2023cites this paper
• Out-of-Domain Discourse Dependency Parsing via Bootstrapping: An Empirical Analysis on Its Effectiveness and Limitation

  2022cites this paper
• Improving Low-resource RRG Parsing with Cross-lingual Self-training

  2022cites this paper
• Self-training For Pre-training Language Models

  2021cites this paper
• FlipDA: Effective and Robust Data Augmentation for Few-Shot Learning

  2021cites this paper
• Exploiting Cloze-Questions for Few-Shot Text Classification and Natural Language Inference

  2020cites this paper
• Reciprocal Supervised Learning Improves Neural Machine Translation

  2020cites this paper
• Predictions For Pre-training Language Models

  2020cites this paper
• Towards Scalable Image Classifier Learning with Noisy Labels via Domain Adaptation

  2020cites this paper
• slimIPL: Language-Model-Free Iterative Pseudo-Labeling

  2020cites this paper
• Zero-shot Text Classification via Reinforced Self-training

  2020cites this paper
• Capturing document context inside sentence-level neural machine translation models with self-training

  2020cites this paper
• RBN: enhancement in language attribute prediction using global representation of natural language transfer learning technology like Google BERT

  2019cites this paper
• Revisiting Self-Training for Neural Sequence Generation

  2019cites this paper
• Dynamic Self-training Framework for Graph Convolutional Networks

  2019cites this paper
• Smooth Methods in Head-Driven Statistical Models for Parsing

  2019cites this paper
• Two Local Models for Neural Constituent Parsing

  2018influential citation
• Learning How to Self-Learn: Enhancing Self-Training Using Neural Reinforcement Learning

  2018cites this paper
• Adaptive Semantic Segmentation with a Strategic Curriculum of Proxy Labels

  2018cites this paper
• The Devil is in the Details: Parsing Unknown German Words

  2017cites this paper
• Attention is All you Need

  2017cites this paper
• Latent-Variable PCFGs: Background and Applications

  2017cites this paper
• Building a Treebank for Vietnamese Syntactic Parsing

  2017cites this paper
• Improving Shift‐Reduce Phrase‐Structure Parsing with Constituent Boundary Information

  2017cites this paper
• Shift-Reduce Constituent Parsing with Neural Lookahead Features

  2016cites this paper
• Recurrent Neural Network Grammars

  2016cites this paper
• Enhancing Shift-Reduce Constituent Parsing with Action N-Gram Model

  2016cites this paper
• The Role of Intended Audience in Determining Modality Type : A Study in Relation to the Iranian Constitution

  2016cites this paper
• Irish dependency treebanking and parsing

  2016cites this paper
• Building the Vietnamese Phrase Treebank by Improved Probabilistic Context-Free Grammars

  2016cites this paper
• Iterative parameter mixing for distributed large-margin training of structured predictors for natural language processing

  2015cites this paper
• POS tagging of Chinese Buddhist texts using Recurrent Neural Networks

  2015cites this paper
• Modeling Reportable Events as Turning Points in Narrative

  2015cites this paper
• Parser self-training for syntax-based machine translation

  2015cites this paper
• The role of syntax and semantics in machine translation and quality estimation of machine-translated user-generated content

  2015cites this paper
• Graph-Based Lexicon Regularization for PCFG With Latent Annotations

  2015cites this paper
• Effective Use of Cross-Domain Parsing in Automatic Speech Recognition and Error Detection

  2015cites this paper
• Mapping Unseen Words to Task-Trained Embedding Spaces

  2015cites this paper
• Training with Auto-parsed Whole Trees

  2015cites this paper
• Structured Training for Neural Network Transition-Based Parsing

  2015cites this paper
• Domain adaptation for parsing in automatic speech recognition

  2014cites this paper
• Chinese Unknown Word Recognition for PCFG-LA Parsing

  2014cites this paper
• Efficient Latent-variable Grammars : Learning and Inference

  2014cites this paper
• A word clustering approach to domain adaptation: Robust parsing of source and target domains

  2014cites this paper
• Parsing low-resource languages using Gibbs sampling for PCFGs with latent annotations

  2014cites this paper
• Grammar as a Foreign Language

  2014influential citation
• Joint POS Tagging and Transition-based Constituent Parsing in Chinese with Non-local Features

  2014influential citation
• Learning latent variable grammars from complementary perspectives

  2014cites this paper
• Lexicon expansion for latent variable grammars

  2014cites this paper
• Ambiguity-aware Ensemble Training for Semi-supervised Dependency Parsing

  2014influential citation
• Exploiting limited data for parsing

  2014cites this paper
• Fast and Accurate Shift-Reduce Constituent Parsing

  2013cites this paper
• Iterative Transformation of Annotation Guidelines for Constituency Parsing

  2013cites this paper
• Word Segmentation, Unknown-word Resolution, and Morphological Agreement in a Hebrew Parsing System

  2013influential citation
• Semi-Supervised Learning and Domain Adaptation in Natural Language Processing

  2013cites this paper
• Augmented Parsing of Unknown Word by Graph-Based Semi-Supervised Learning

  2013cites this paper
• Factored Soft Source Syntactic Constraints for Hierarchical Machine Translation

  2013cites this paper
• Working with a small dataset - semi-supervised dependency parsing for Irish

  2013cites this paper
• Improving shift-reduce constituency parsing with large-scale unlabeled data

  2013cites this paper
• Embedding epistemic modals in English: A corpus-based study

  2012cites this paper
• Practical and efficient incorporation of syntactic features into statistical language models

  2012cites this paper
• Self-Training Tree Substitution Grammars for Domain Adaptation

  2012influential citation
• Phrase Parses Reranking Based on Higher-Order Lexical Dependencies: Phrase Parses Reranking Based on Higher-Order Lexical Dependencies

  2012influential citation
• SELECTIVE LEARNING IN THE ACQUISITION OF KANNADA

  2012cites this paper
• Revisiting the Case for Explicit Syntactic Information in Language Models

  2012cites this paper
• NAACL-HLT 2012 WLM 2012: Will We Ever Really Replace the N-gram Model? On the Future of Language Modeling for HLT Workshop Notes

  2012cites this paper
• Exploiting Lexical Dependencies from Large-Scale Data for Better Shift-Reduce Constituency Parsing

  2012influential citation
• Proceedings of the NAACL-HLT 2012 Workshop: Will We Ever Really Replace the N-gram Model? On the Future of Language Modeling for HLT - Workshop Notes

  2012cites this paper
• Intégration de ressources lexicales riches dans un analyseur syntaxique probabiliste. (Integration of lexical resources in a probabilistic parser)

  2012cites this paper
• Learning sub-word units and exploiting contextual information for open vocabulary speech recognition

  2011cites this paper
• Joint Hebrew Segmentation and Parsing using a PCFGLA Lattice Parser

  2011influential citation
• Efficient discriminative training of long-span language models

  2011cites this paper
• Parse Reranking Based on Higher-Order Lexical Dependencies

  2011cites this paper
• Generalized Interpolation in Decision Tree LM

  2011cites this paper
• Coarse-to-Fine Natural Language Processing

  2011influential citation
• Data point selection for self-training

  2011cites this paper
• The Second Workshop on Statistical Parsing of Morphologically Rich Languages ( SPMRL

  2011cites this paper
• From News to Comment: Resources and Benchmarks for Parsing the Language of Web 2.0

  2011cites this paper
• Feature-Rich Log-Linear Lexical Model for Latent Variable PCFG Grammars

  2011influential citation
• Modeling Dependencies in Natural Languages with Latent Variables

  2011cites this paper
• OOV Sensitive Named-Entity Recognition in Speech

  2011cites this paper
• Comparing the Use of Edited and Unedited Text in Parser Self-Training

  2011cites this paper
• Selective learning in the acquisition of Kannada ditransitives

  2011cites this paper
• Syntactic Decision Tree LMs: Random Selection or Intelligent Design?

  2011cites this paper
• Improving Part-of-speech Tagging for Context-free Parsing

  2011cites this paper
• Self-Training with Products of Latent Variable Grammars

  2010influential citation
• Contextual Information Improves OOV Detection in Speech

  2010cites this paper
• Handling Unknown Words in Statistical Latent-Variable Parsing Models for Arabic, English and French

  2010cites this paper
• Appropriately Handled Prosodic Breaks Help PCFG Parsing

  2010cites this paper
• Proceedings of the NAACL HLT 2010 Workshop on Computational Linguistics and Writing: Writing Process

  2010cites this paper
• Soft Syntactic Constraints for Hierarchical Phrase-Based Translation Using Latent Syntactic Distributions

  2010cites this paper
• Learning Simple Wikipedia: A Cogitation in Ascertaining Abecedarian Language

  2010cites this paper
• Uptraining for Accurate Deterministic Question Parsing

  2010cites this paper
• Better Arabic Parsing: Baselines, Evaluations, and Analysis

  2010cites this paper
• Treebank Conversion based Self-training Strategy for Parsing

  2010cites this paper
• Workshop on Computational Linguistics and Writing: Writing Processes and Authoring Aids (cl&w 2010) Grammaticality Judgement in a Word Completion Task Exploring Individual Differences in Student Writing with a Narrative Composition Support Environment Workshop Program Exploring Individual Difference

  2010cites this paper
• Products of Random Latent Variable Grammars

  2010cites this paper
• Phrase Structure Parsing with Dependency Structure

  2010cites this paper
• A Parse-and-Trim Approach with Information Significance for Chinese Sentence Compression

  2009cites this paper