Cross-Lingual Transfer of Semantic Roles: From Raw Text to Semantic Roles

We describe a transfer method based on annotation projection to develop a dependency-based semantic role labeling system for languages for which no supervised linguistic information other than parallel data is available. Unlike previous work that presumes the availability of supervised features such as lemmas, part-of-speech tags, and dependency parse trees, we only make use of word and character features. Our deep model considers using character-based representations as well as unsupervised stem embeddings to alleviate the need for supervised features. Our experiments outperform a state-of-the-art method that uses supervised lexico-syntactic features on 6 out of 7 languages in the Universal Proposition Bank.

[1]  Manfred Pinkal,et al.  Towards a Resource for Lexical Semantics: A Large German Corpus with Extensive Semantic Annotation , 2003, ACL.

[2]  Slav Petrov,et al.  A Universal Part-of-Speech Tagset , 2011, LREC.

[3]  Philip Resnik,et al.  Evaluating Translational Correspondence using Annotation Projection , 2002, ACL.

[4]  Luke S. Zettlemoyer,et al.  Deep Semantic Role Labeling: What Works and What’s Next , 2017, ACL.

[5]  Diego Marcheggiani,et al.  A Simple and Accurate Syntax-Agnostic Neural Model for Dependency-based Semantic Role Labeling , 2017, CoNLL.

[6]  Ryan Cotterell,et al.  UniMorph 3.0: Universal Morphology , 2018, LREC.

[7]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[8]  Mirella Lapata,et al.  Cross-lingual Annotation Projection for Semantic Roles , 2009, J. Artif. Intell. Res..

[9]  Ivan Titov,et al.  Bootstrapping Semantic Role Labelers from Parallel Data , 2013, *SEM@NAACL-HLT.

[10]  Hermann Ney,et al.  A Systematic Comparison of Various Statistical Alignment Models , 2003, CL.

[11]  Hai Zhao,et al.  A Full End-to-End Semantic Role Labeler, Syntactic-agnostic Over Syntactic-aware? , 2018, COLING.

[12]  Yunyao Li,et al.  Generating High Quality Proposition Banks for Multilingual Semantic Role Labeling , 2015, ACL.

[13]  Nizar Habash,et al.  Unsupervised Morphology-Based Vocabulary Expansion , 2014, ACL.

[14]  Martha Palmer,et al.  The Revised Arabic PropBank , 2010, Linguistic Annotation Workshop.

[15]  Yunyao Li,et al.  Towards Semi-Automatic Generation of Proposition Banks for Low-Resource Languages , 2016, EMNLP.

[16]  Mirella Lapata,et al.  Neural Semantic Role Labeling with Dependency Path Embeddings , 2016, ACL.

[17]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[18]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[19]  Mirella Lapata,et al.  Cross-linguistic Projection of Role-Semantic Information , 2005, HLT/EMNLP.

[20]  Mikko Kurimo,et al.  Morfessor 2.0: Python Implementation and Extensions for Morfessor Baseline , 2013 .

[21]  Andreas Eisele,et al.  MultiUN: A Multilingual Corpus from United Nation Documents , 2010, LREC.

[22]  Hai Zhao,et al.  Syntax for Semantic Role Labeling, To Be, Or Not To Be , 2018, ACL.

[23]  Michael Collins,et al.  Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with Perceptron Algorithms , 2002, EMNLP.

[24]  Mohammad Sadegh Rasooli,et al.  Transferring Semantic Roles Using Translation and Syntactic Information , 2017, IJCNLP.

[25]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[26]  Pierre Nugues,et al.  Multilingual Semantic Role Labeling , 2009, CoNLL Shared Task.

[27]  Milan Straka,et al.  Tokenizing, POS Tagging, Lemmatizing and Parsing UD 2.0 with UDPipe , 2017, CoNLL.

[28]  Mikko Kurimo,et al.  Morfessor FlatCat: An HMM-Based Method for Unsupervised and Semi-Supervised Learning of Morphology , 2014, COLING.

[29]  Lonneke van der Plas,et al.  Global Methods for Cross-lingual Semantic Role and Predicate Labelling , 2014, COLING.

[30]  Eric P. Xing,et al.  Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , 2014, ACL 2014.

[31]  Wang Ling,et al.  Two/Too Simple Adaptations of Word2Vec for Syntax Problems , 2015, NAACL.

[32]  Ivan Titov,et al.  Cross-lingual Transfer of Semantic Role Labeling Models , 2013, ACL.

[33]  Philipp Koehn,et al.  Europarl: A Parallel Corpus for Statistical Machine Translation , 2005, MTSUMMIT.

[34]  Wei Xu,et al.  End-to-end learning of semantic role labeling using recurrent neural networks , 2015, ACL.

[35]  Noah A. Smith,et al.  Polyglot Semantic Role Labeling , 2018, ACL.

[36]  Kevin Duh,et al.  DyNet: The Dynamic Neural Network Toolkit , 2017, ArXiv.

[37]  Lonneke van der Plas,et al.  Scaling up Automatic Cross-Lingual Semantic Role Annotation , 2011, ACL.

[38]  Fei Xia,et al.  CROWD-IN-THE-LOOP: A Hybrid Approach for Annotating Semantic Roles , 2017, EMNLP.

[39]  Daniel Gildea,et al.  The Proposition Bank: An Annotated Corpus of Semantic Roles , 2005, CL.