Title: Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLP
What are the units of text that we want to model? From bytes to multi-word expressions, text can be analyzed and generated at many granularities. Until recently, most natural language processing (NLP) models operated over words, treating those as discrete and atomic tokens, but starting with byte-pair encoding (BPE), subword-based approaches have become dominant in many areas, enabling small vocabularies while still allowing for fast inference. Is the end of the road character-level model or byte-level processing? In this survey, we connect several lines of work from the pre-neural and neural era, by showing how hybrid approaches of words and characters as well as subword-based approaches based on learned segmentation have been proposed and evaluated. We conclude that there is and likely will never be a silver bullet singular solution for all applications and that thinking seriously about tokenization remains important for many applications. more »« less
"Multilingual neural machine translation (MNMT) jointly trains a shared model for translation with multiple language pairs. However, traditional subword-based MNMT approaches suffer from out-of-vocabulary (OOV) issues and representation bottleneck, which often degrades translation performance on certain language pairs. While byte tokenization is used to tackle the OOV problems in neural machine translation (NMT), until now its capability has not been validated in MNMT. Additionally, existing work has not studied how byte encoding can benefit endangered language translation to our knowledge. We propose a byte-based multilingual neural machine translation system (BMNMT) to alleviate the representation bottleneck and improve translation performance in endangered languages. Furthermore, we design a random byte mapping method with an ensemble prediction to enhance our model robustness. Experimental results show that our BMNMT consistently and significantly outperforms subword/word-based baselines on twelve language pairs up to +18.5 BLEU points, an 840{\%} relative improvement.",
Gaines, Dylan; Vertanen, Keith
(, Association for Computational Linguistics)
Users of Augmentative and Alternative Communication (AAC) may write letter-by-letter via an interface that uses a character language model. However, most state-of-the-art large pretrained language models predict subword tokens of variable length. We investigate how to practically use such models to make accurate and efficient character predictions. Our algorithm for producing character predictions from a subword large language model (LLM) provides more accurate predictions than using a classification layer, a byte-level LLM, or an n-gram model. Additionally, we investigate a domain adaptation procedure based on a large dataset of sentences we curated based on scoring how useful each sentence might be for spoken or written AAC communication. We find our procedure further improves model performance on simple, conversational text.
El-Kishky, Ahmed; Xu, Frank; Zhang, Aston; Han, Jiawei
(, 2019 {IEEE} International Conference on Big Data (Big Data))
Traditionally, many text-mining tasks treat individual word-tokens as the finest meaningful semantic granularity. However, in many languages and specialized corpora, words are composed by concatenating semantically meaningful subword structures. Word-level analysis cannot leverage the semantic information present in such subword structures. With regard to word embedding techniques, this leads to not only poor embeddings for infrequent words in long-tailed text corpora but also weak capabilities for handling out-of-vocabulary words. In this paper we propose MorphMine for unsupervised morpheme segmentation. MorphMine applies a parsimony criterion to hierarchically segment words into the fewest number of morphemes at each level of the hierarchy. This leads to longer shared morphemes at each level of segmentation. Experiments show that MorphMine segments words in a variety of languages into human-verified morphemes. Additionally, we experimentally demonstrate that utilizing MorphMine morphemes to enrich word embeddings consistently improves embedding quality on a variety of of embedding evaluations and a downstream language modeling task.
Schwartz, Lane; Tyers, Francis; Levin, Lori; Kirov, Christo; Littell, Patrick; Lo, Chi-kiu; Prud'hommeaux, Emily; Park, Hyunji Hayley; Steimel, Kenneth; Knowles, Rebecca; et al
(, ArXivorg)
Many techniques in modern computational linguistics and natural language processing (NLP) make the assumption that approaches that work well on English and other widely used European (and sometimes Asian) languages are “language agnostic” – that is that they will also work across the typologically diverse languages of the world. In high-resource languages, especially those that are analytic rather than synthetic, a common approach is to treat morphologically-distinct variants of a common root (such as dog and dogs) as completely independent word types. Doing so relies on two main assumptions: that there exist a limited number of morphological inflections for any given root, and that most or all of those variants will appear in a large enough corpus (conditioned on assumptions about domain, etc.) so that the model can adequately learn statistics about each variant. Approaches like stemming, lemmatization, morphological analysis, subword segmentation, or other normalization techniques are frequently used when either of those assumptions are likely to be violated, particularly in the case of synthetic languages like Czech and Russian that have more inflectional morphology than English. Within the NLP literature, agglutinative languages like Finnish and Turkish are commonly held up as extreme examples of morphological complexity that challenge common modelling assumptions. Yet, when considering all of the world’s languages, Finnish and Turkish are closer to the average case in terms of synthesis. When we consider polysynthetic languages (those at the extreme of morphological complexity), even approaches like stemming, lemmatization, or subword modelling may not suffice. These languages have very high numbers of hapax legomena (words appearing only once in a corpus), underscoring the need for appropriate morphological handling of words, without which there is no hope for a model to capture enough statistical information about those words. Moreover, many of these languages have only very small text corpora, substantially magnifying these challenges. To this end, we examine the current state-of-the-art in language modelling, machine translation, and predictive text completion in the context of four polysynthetic languages: Guaraní, St. Lawrence Island Yupik, Central Alaskan Yup’ik, and Inuktitut. We have a particular focus on Inuit-Yupik, a highly challenging family of endangered polysynthetic languages that ranges geographically from Greenland through northern Canada and Alaska to far eastern Russia. The languages in this family are extraordinarily challenging from a computational perspective, with pervasive use of derivational morphemes in addition to rich sets of inflectional suffixes and phonological challenges at morpheme boundaries. Finally, we propose a novel framework for language modelling that combines knowledge representations from finite-state morphological analyzers with Tensor Product Representations (Smolensky, 1990) in order to enable successful neural language models capable of handling the full linguistic variety of typologically variant languages.
Schwartz, Lane; Tyers, Francis; Levin, Lori; Kirov, Christo; Littell, Patrick; Lo, Chi-kiu; Prud'hommeaux, Emily; Park, Hyunji Hayley; Steimel, Kenneth; Knowles, Rebecca; et al
(, Final Report of the Frederick Jelinek Memorial Summer Workshop)
null
(Ed.)
Many techniques in modern computational linguistics and natural language processing (NLP) make the assumption that approaches that work well on English and other widely used European (and sometimes Asian) languages are “language agnostic” – that is that they will also work across the typologically diverse languages of the world. In high-resource languages, especially those that are analytic rather than synthetic, a common approach is to treat morphologically-distinct variants of a common root (such as dog and dogs) as completely independent word types. Doing so relies on two main assumptions: that there exist a limited number of morphological inflections for any given root, and that most or all of those variants will appear in a large enough corpus (conditioned on assumptions about domain, etc.) so that the model can adequately learn statistics about each variant. Approaches like stemming, lemmatization, morphological analysis, subword segmentation, or other normalization techniques are frequently used when either of those assumptions are likely to be violated, particularly in the case of synthetic languages like Czech and Russian that have more inflectional morphology than English. Within the NLP literature, agglutinative languages like Finnish and Turkish are commonly held up as extreme examples of morphological complexity that challenge common modelling assumptions. Yet, when considering all of the world’s languages, Finnish and Turkish are closer to the average case in terms of synthesis. When we consider polysynthetic languages (those at the extreme of morphological complexity), even approaches like stemming, lemmatization, or subword modelling may not suffice. These languages have very high numbers of hapax legomena (words appearing only once in a corpus), underscoring the need for appropriate morphological handling of words, without which there is no hope for a model to capture enough statistical information about those words. Moreover, many of these languages have only very small text corpora, substantially magnifying these challenges. To this end, we examine the current state-of-the-art in language modelling, machine translation, and predictive text completion in the context of four polysynthetic languages: Guaraní, St. Lawrence Island Yupik, Central Alaskan Yup’ik, and Inuktitut. We have a particular focus on Inuit-Yupik, a highly challenging family of endangered polysynthetic languages that ranges geographically from Greenland through northern Canada and Alaska to far eastern Russia. The languages in this family are extraordinarily challenging from a computational perspective, with pervasive use of derivational morphemes in addition to rich sets of inflectional suffixes and phonological challenges at morpheme boundaries. Finally, we propose a novel framework for language modelling that combines knowledge representations from finite-state morphological analyzers with Tensor Product Representations (Smolensky, 1990) in order to enable successful neural language models capable of handling the full linguistic variety of typologically variant languages.
Sabrina J. Mielke, Zaid Alyafeai, Elizabeth Salesky, Colin Raffel, Manan Dey, Matthias Gallé, Arun Raja, Chenglei Si, Wilson Y. Lee, Benoît Sagot, and Samson Tan. Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLP. Retrieved from https://par.nsf.gov/biblio/10347731. Computing Research Repository (arXiv) .
Sabrina J. Mielke, Zaid Alyafeai, Elizabeth Salesky, Colin Raffel, Manan Dey, Matthias Gallé, Arun Raja, Chenglei Si, Wilson Y. Lee, Benoît Sagot, & Samson Tan. Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLP. Computing Research Repository (arXiv), (). Retrieved from https://par.nsf.gov/biblio/10347731.
Sabrina J. Mielke, Zaid Alyafeai, Elizabeth Salesky, Colin Raffel, Manan Dey, Matthias Gallé, Arun Raja, Chenglei Si, Wilson Y. Lee, Benoît Sagot, and Samson Tan.
"Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLP". Computing Research Repository (arXiv) (). Country unknown/Code not available. https://par.nsf.gov/biblio/10347731.
@article{osti_10347731,
place = {Country unknown/Code not available},
title = {Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLP},
url = {https://par.nsf.gov/biblio/10347731},
abstractNote = {What are the units of text that we want to model? From bytes to multi-word expressions, text can be analyzed and generated at many granularities. Until recently, most natural language processing (NLP) models operated over words, treating those as discrete and atomic tokens, but starting with byte-pair encoding (BPE), subword-based approaches have become dominant in many areas, enabling small vocabularies while still allowing for fast inference. Is the end of the road character-level model or byte-level processing? In this survey, we connect several lines of work from the pre-neural and neural era, by showing how hybrid approaches of words and characters as well as subword-based approaches based on learned segmentation have been proposed and evaluated. We conclude that there is and likely will never be a silver bullet singular solution for all applications and that thinking seriously about tokenization remains important for many applications.},
journal = {Computing Research Repository (arXiv)},
author = {Sabrina J. Mielke and Zaid Alyafeai and Elizabeth Salesky and Colin Raffel and Manan Dey and Matthias Gallé and Arun Raja and Chenglei Si and Wilson Y. Lee and Benoît Sagot and Samson Tan},
}
Warning: Leaving National Science Foundation Website
You are now leaving the National Science Foundation website to go to a non-government website.
Website:
NSF takes no responsibility for and exercises no control over the views expressed or the accuracy of
the information contained on this site. Also be aware that NSF's privacy policy does not apply to this site.
An official website of the United States government
