Filters








209 Hits in 10.2 sec

From Phonology to Syntax: Unsupervised Linguistic Typology at Different Levels with Language Embeddings [article]

Johannes Bjerva, Isabelle Augenstein
2018 arXiv   pre-print
Additionally, quantitative and qualitative analyses of these language embeddings can tell us how language similarities are encoded in NLP models for tasks at different typological levels.  ...  The representations are learned in an unsupervised manner alongside tasks at three typological levels: phonology (grapheme-to-phoneme prediction, and phoneme reconstruction), morphology (morphological  ...  We choose to investigate three linguistic levels of language: phonology, morphology, and syntax.  ... 
arXiv:1802.09375v1 fatcat:rsmpvxoo6bgyxhmzk7ka3ox56q

From Phonology to Syntax: Unsupervised Linguistic Typology at Different Levels with Language Embeddings

Johannes Bjerva, Isabelle Augenstein
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)  
Additionally, quantitative and qualitative analyses of these language embeddings can tell us how language similarities are encoded in NLP models for tasks at different typological levels.  ...  The representations are learned in an unsupervised manner alongside tasks at three typological levels: phonology (grapheme-to-phoneme prediction, and phoneme reconstruction), morphology (morphological  ...  We choose to investigate three linguistic levels of language: phonology, morphology, and syntax.  ... 
doi:10.18653/v1/n18-1083 dblp:conf/naacl/BjervaA18 fatcat:gzdidkqctncqdjzpx56q6c4e5a

Inflection at the morphology-syntax interface [chapter]

Claudia Marzi, James P. Blevins, Geert Booij, Vito Pirrelli
2020 Word Knowledge and Word Usage  
In doing that, we will embark on an interdisciplinary tour, which will touch upon theoretical, psychological, typological, historical and computational issues in morphology, with a view to looking for  ...  Is it part of language morphology, syntax or both? What are the basic units of inflection and how do speakers acquire and process them? How do they vary across languages?  ...  This makes them extremely powerful processing devices, capturing morphological generalizations at fine-grained levels of generality, ranging from sweeping, exceptionless phonological processes to lexically-conditioned  ... 
doi:10.1515/9783110440577-007 fatcat:d33sxyyhrncgxlh6bpzztpaoqq

Mechanisms for interaction: Syntax as procedures for online interactive meaning building

Ruth Kempson, Stergios Chatzikyriakidis, Ronnie Cann
2016 Behavioral and Brain Sciences  
, as in Dynamic Syntax, in which syntax is the incremental building of semantic representations reflecting real-time parsing dynamics.  ...  We demonstrate that, with such formulation, syntactic, semantic, and morpho-syntactic dependencies are all analysable as grounded in their potential for interaction.  ...  Importantly, prediction allows for top-down information to influence current processing across different levels of linguistic representation, from phonology to discourse, and at different temporal windows  ... 
doi:10.1017/s0140525x15000849 pmid:27562087 fatcat:5fmm7eb6bzfepo4jjjwqoblt5e

What Can Mathematical, Computational, and Robotic Models Tell Us about the Origins of Syntax? [chapter]

2009 Biological Foundations and Origin of Syntax  
Compositionality, along with recursion, is the fundamental 30 feature of human syntax that gives us open-ended expressivity.  ...  Or they will make assumptions about certain aspects of 8 language interaction (such as joint attention or perception) in order to make simulations doable at 9 all.  ...  This recruitment usually concurs with phonological 26 reduction, like the change from "going to" to "gonna".  ... 
doi:10.7551/mitpress/8468.003.0026 fatcat:ctkeihiosnh2fiexb3ow6re7dq

A Probabilistic Generative Model of Linguistic Typology [article]

Johannes Bjerva and Yova Kementchedjhieva and Ryan Cotterell and Isabelle Augenstein
2019 arXiv   pre-print
Furthermore, we show that language embeddings pre-trained on monolingual text allow for generalisation to unobserved languages.  ...  By modelling all languages and features within the same architecture, we show how structural similarities between languages can be exploited to predict typological features with near-perfect accuracy,  ...  The third author acknowledges support from a Facebook Fellowship.  ... 
arXiv:1903.10950v3 fatcat:33ynwnslrjen7g6qz37xuewiom

A Probabilistic Generative Model of Linguistic Typology

Johannes Bjerva, Yova Kementchedjhieva, Ryan Cotterell, Isabelle Augenstein
2019 Proceedings of the 2019 Conference of the North  
Furthermore, we show that language embeddings pre-trained on monolingual text allow for generalisation to unobserved languages.  ...  By modelling all languages and features within the same architecture, we show how structural similarities between languages can be exploited to predict typological features with near-perfect accuracy,  ...  The third author acknowledges support from a Facebook Fellowship.  ... 
doi:10.18653/v1/n19-1156 dblp:conf/naacl/BjervaKCA19 fatcat:knu6nj3gqzg5loth45n3vugjqi

Learning Language Representations for Typology Prediction

Chaitanya Malaviya, Graham Neubig, Patrick Littell
2017 Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing  
When a neural machine translation system learns to translate from one language to another, does it learn the syntax or semantics of the languages?  ...  information missing from typological databases.  ...  linguistic typology from unsupervised neural models.  ... 
doi:10.18653/v1/d17-1268 dblp:conf/emnlp/MalaviyaNL17 fatcat:ucmtp53ksze4jeoofa3rpcquui

Learning Language Representations for Typology Prediction [article]

Chaitanya Malaviya, Graham Neubig, Patrick Littell
2017 arXiv   pre-print
When a neural machine translation system learns to translate from one language to another, does it learn the syntax or semantics of the languages?  ...  information missing from typological databases.  ...  linguistic typology from unsupervised neural models.  ... 
arXiv:1707.09569v1 fatcat:vwtcfspokfedbfvrshxte5b2tm

Modeling Language Variation and Universals: A Survey on Typological Linguistics for Natural Language Processing

Edoardo Maria Ponti, Helen O'Horan, Yevgeni Berzak, Ivan Vulić, Roi Reichart, Thierry Poibeau, Ekaterina Shutova, Anna Korhonen
2019 Computational Linguistics  
Linguistic typology aims to capture structural and semantic variation across the world's languages.  ...  A large-scale typology could provide excellent guidance for multilingual Natural Language Processing (NLP), particularly for languages that suffer from the lack of human labeled resources.  ...  Some other databases only cover features at a specific level of linguistic description.  ... 
doi:10.1162/coli_a_00357 fatcat:cfekqbpmwzegdf6j6atez2rsbe

Modeling Language Variation and Universals: A Survey on Typological Linguistics for Natural Language Processing [article]

Edoardo Maria Ponti, Helen O'Horan, Yevgeni Berzak, Ivan Vulić, Roi Reichart, Thierry Poibeau, Ekaterina Shutova, Anna Korhonen
2020 arXiv   pre-print
Linguistic typology aims to capture structural and semantic variation across the world's languages.  ...  A large-scale typology could provide excellent guidance for multilingual Natural Language Processing (NLP), particularly for languages that suffer from the lack of human labeled resources.  ...  Cross-lingual variation can be found at all levels of linguistic structure.  ... 
arXiv:1807.00914v3 fatcat:3b5vklsb6zfmrlifhtpvlxbk6q

Language Embeddings for Typology and Cross-lingual Transfer Learning [article]

Dian Yu and Taiqi He and Kenji Sagae
2021 arXiv   pre-print
We generate dense embeddings for 29 languages using a denoising autoencoder, and evaluate the embeddings using the World Atlas of Language Structures (WALS) and two extrinsic tasks in a zero-shot setting  ...  We explore whether language representations that capture relationships among languages can be learned and subsequently leveraged in cross-lingual tasks without the use of parallel data.  ...  From phonology to syntax: Unsupervised linguistic typol- ogy at different levels with language embeddings.  ... 
arXiv:2106.02082v1 fatcat:l4sdumpujvg2dbipbq4fz6pj5a

Multi-SimLex: A Large-Scale Evaluation of Multilingual and Cross-Lingual Lexical Semantic Similarity [article]

Ivan Vulić, Simon Baker, Edoardo Maria Ponti, Ulla Petti, Ira Leviant, Kelly Wing, Olga Majewska, Eden Bar, Matt Malone, Thierry Poibeau, Roi Reichart, Anna Korhonen
2020 arXiv   pre-print
(such as fastText, M-BERT and XLM), externally informed lexical representations, as well as fully unsupervised and (weakly) supervised cross-lingual word embeddings.  ...  We introduce Multi-SimLex, a large-scale lexical resource and evaluation benchmark covering datasets for 12 typologically diverse languages, including major languages (e.g., Mandarin Chinese, Spanish,  ...  Acknowledgments This work is supported by the ERC Consolidator Grant LEXICAL: Lexical Acquisition Across Languages (no 648909).  ... 
arXiv:2003.04866v1 fatcat:5mp5s7ehyzdshnywt2zvqverwu

Multi-SimLex: A Large-Scale Evaluation of Multilingual and Cross-Lingual Lexical Semantic Similarity

Ivan Vulić, Edoardo Maria Ponti, Ira Leviant, Olga Majewska, Matt Malone, Roi Reichart, Simon Baker, Ulla Petti, Kelly Wing, Eden Bar, Thierry Poibeau, Anna Korhonen
2020 Computational Linguistics  
(such as fastText, monolingual and multilingual BERT, XLM), externally informed lexical representations, as well as fully unsupervised and (weakly) supervised cross-lingual word embeddings.  ...  We introduce Multi-SimLex, a large-scale lexical resource and evaluation benchmark covering data sets for 12 typologically diverse languages, including major languages (e.g., Mandarin Chinese, Spanish,  ...  Acknowledgments This work is supported by the ERC Consolidator Grant LEXICAL: Lexical Acquisition Across Languages (no 648909).  ... 
doi:10.1162/coli_a_00391 fatcat:42esnmz2gvgs7irdhigl6t7xtm

Typological Features for Multilingual Delexicalised Dependency Parsing

Manon Scholivet, Franck Dary, Alexis Nasr, Benoit Favre, Carlos Ramisch
2019 Proceedings of the 2019 Conference of the North  
The existence of universal models to describe the syntax of languages has been debated for decades.  ...  Our work investigates the use of high-level language descriptions in the form of typological features for multilingual dependency parsing.  ...  The input layer of the MLP corresponds to the concatenation of the embeddings of the different features, with dimensions varying from 396 to 465, depending on the configuration (with or without language  ... 
doi:10.18653/v1/n19-1393 dblp:conf/naacl/ScholivetDNFR19 fatcat:l2i23rhvpvgyjflysavijnvqnu
« Previous Showing results 1 — 15 out of 209 results