Words in some natural languages can have a composite structure. Elements of this structure include the root (that could also be composite), prefixes and suffixes with which various nuances and relations to other words can be expressed. Thus, in order to build a proper word representation one must take into account its internal structure. From a corpus of texts we extract a set of frequent subwords and from the latter set we select patterns, i.e. subwords which encapsulate information on character $n$-gram regularities. The selection is made using the pattern-based Conditional Random Field model with $l_1$ regularization. Further, for every word we construct a new sequence over an alphabet of patterns. The new alphabet's symbols confine a local statistical context stronger than the characters, therefore they allow better representations in ${\mathbb{R}}^n$ and are better building blocks for word representation. In the task of subword-aware language modeling, pattern-based models outperform character-based analogues by 2-20 perplexity points. Also, a recurrent neural network in which a word is represented as a sum of embeddings of its patterns is on par with a competitive and significantly more sophisticated character-based convolutional architecture.
This paper introduces Dynamic Programming Encoding (DPE), a new segmentation algorithm for tokenizing sentences into subword units. We view the subword segmentation of output sentences as a latent variable that should be marginalized out for learning and inference. A mixed character-subword transformer is proposed, which enables exact log marginal likelihood estimation and exact MAP inference to find target segmentations with maximum posterior probability. DPE uses a lightweight mixed character-subword transformer as a means of pre-processing parallel data to segment output sentences using dyn…
State-of-the-art models in natural language processing rely on separate rigid subword tokenization algorithms, which limit their generalization ability and adaptation to new settings. In this paper, we propose a new model inductive bias that learns a subword tokenization end-to-end as part of the model. To this end, we introduce a soft gradient-based subword tokenization module (GBST) that automatically learns latent subword representations from characters in a data-driven fashion. Concretely, GBST enumerates candidate subword blocks and learns to score them in a position-wise fashion using a …
Center for Language and Cognition（语言与认知中心）；University of Groningen（格罗宁根大学）
Subword-level models have been the dominant paradigm in NLP. However, character-level models have the benefit of seeing each character individually, providing the model with more detailed information that ultimately could lead to better models. Recent works have shown character-level models to be competitive with subword models, but costly in terms of time and computation. Character-level models with a downsampling component alleviate this, but at the cost of quality, particularly for machine translation. This work analyzes the problems of previous downsampling methods and introduces a novel d…
University of Maryland（马里兰大学）；University of Colorado（科罗拉多大学）；Google Research（谷歌研究院）
Text classification must sometimes be applied in a low-resource language with no labeled training data. However, training data may be available in a related language. We investigate whether character-level knowledge transfer from a related language helps text classification. We present a cross-lingual document classification framework (CACO) that exploits cross-lingual subword similarity by jointly training a character-based embedder and a word-based classifier. The embedder derives vector representations for input words from their written forms, and the classifier makes predictions based on t…
Ibaraki University（茨城大学）；Tokyo University of Agriculture and Technology（东京农工大学）
This study extracted and analyzed the linguistic speech patterns that characterize Japanese anime or game characters. Conventional morphological analyzers, such as MeCab, segment words with high performance, but they are unable to segment broken expressions or utterance endings that are not listed in the dictionary, which often appears in lines of anime or game characters. To overcome this challenge, we propose segmenting lines of Japanese anime or game characters using subword units that were proposed mainly for deep learning, and extracting frequently occurring strings to obtain expressions …
Tel Aviv University（特拉维夫大学）；Bar Ilan University（巴伊兰大学）
Large pretrained language models (PLMs) typically tokenize the input string into contiguous subwords before any pretraining or inference. However, previous studies have claimed that this form of subword tokenization is inadequate for processing morphologically-rich languages (MRLs). We revisit this hypothesis by pretraining a BERT-style masked language model over character sequences instead of word-pieces. We compare the resulting model, dubbed TavBERT, against contemporary PLMs based on subwords for three highly complex and ambiguous MRLs (Hebrew, Turkish, and Arabic), testing them on both mo…
University of Houston（休斯顿大学）；Salesforce Research（Salesforce研究院）
Byte-pair encoding (BPE) is a ubiquitous algorithm in the subword tokenization process of language models as it provides multiple benefits. However, this process is solely based on pre-training data statistics, making it hard for the tokenizer to handle infrequent spellings. On the other hand, though robust to misspellings, pure character-level models often lead to unreasonably long sequences and make it harder for the model to learn meaningful words. To alleviate these challenges, we propose a character-based subword module (char2subword) that learns the subword embedding table in pre-trained…
We approach the problem of generalizing pre-trained word embeddings beyond fixed-size vocabularies without using additional contextual information. We propose a subword-level word vector generation model that views words as bags of character $n$-grams. The model is simple, fast to train and provides good vectors for rare or unseen words. Experiments show that our model achieves state-of-the-art performances in English word similarity task and in joint prediction of part-of-speech tag and morphosyntactic attributes in 23 languages, suggesting our model's ability in capturing the relationship be…
Representation learning is the foundation of machine reading comprehension and inference. In state-of-the-art models, character-level representations have been broadly adopted to alleviate the problem of effectively representing rare or complex words. However, character itself is not a natural minimal linguistic unit for representation or word embedding composing due to ignoring the linguistic coherence of consecutive characters inside word. This paper presents a general subword-augmented embedding framework for learning and composing computationally-derived subword-level representations. We s…
Representation learning is the foundation of machine reading comprehension. In state-of-the-art models, deep learning methods broadly use word and character level representations. However, character is not naturally the minimal linguistic unit. In addition, with a simple concatenation of character and word embedding, previous models actually give suboptimal solution. In this paper, we propose to use subword rather than character for word embedding enhancement. We also empirically explore different augmentation strategies on subword-augmented embedding to enhance the cloze-style reading compreh…
Center for Information and Speech Processing（信息与语音处理中心）；Ludwig Maximilian University of Munich（慕尼黑大学）
Applying the Transformer architecture on the character level usually requires very deep architectures that are difficult and slow to train. These problems can be partially overcome by incorporating a segmentation into tokens in the model. We show that by initially training a subword model and then finetuning it on characters, we can obtain a neural machine translation model that works at the character level without requiring token segmentation. We use only the vanilla 6-layer Transformer Base architecture. Our character-level models better capture morphological phenomena and show more robustne…
Simplified Chinese to Traditional Chinese character conversion is a common preprocessing step in Chinese NLP. Despite this, current approaches have poor performance because they do not take into account that a simplified Chinese character can correspond to multiple traditional characters. Here, we propose a model that can disambiguate between mappings and convert between the two scripts. The model is based on subword segmentation, two language models, as well as a method for mapping between subword sequences. We further construct benchmark datasets for topic classification and script conversio…
Shanghai Jiao Tong University（上海交通大学）；Tencent AI Lab（腾讯AI实验室）
Embedding from Language Models (ELMo) has shown to be effective for improving many natural language processing (NLP) tasks, and ELMo takes character information to compose word representation to train language models.However, the character is an insufficient and unnatural linguistic unit for word representation.Thus we introduce Embedding from Subword-aware Language Models (ESuLMo) which learns word representation from subwords using unsupervised segmentation over words.We show that ESuLMo can enhance four benchmark NLP tasks more effectively than ELMo, including syntactic dependency parsing, …
Current neural query auto-completion (QAC) systems rely on character-level language models, but they slow down when queries are long. We present how to utilize subword language models for the fast and accurate generation of query completion candidates. Representing queries with subwords shorten a decoding length significantly. To deal with issues coming from introducing subword language model, we develop a retrace algorithm and a reranking method by approximate marginalization. As a result, our model achieves up to 2.5 times faster while maintaining a similar quality of generated results compa…
Facebook AI Research（脸书人工智能研究院）；New York University（纽约大学）
Almost all existing machine translation models are built on top of character-based vocabularies: characters, subwords or words. Rare characters from noisy text or character-rich languages such as Japanese and Chinese however can unnecessarily take up vocabulary slots and limit its compactness. Representing text at the level of bytes and using the 256 byte set as vocabulary is a potential solution to this issue. High computational cost has however prevented it from being widely deployed or used in practice. In this paper, we investigate byte-level subwords, specifically byte-level BPE (BBPE), w…
Tsinghua University（清华大学）；Shanghai Liulishuo Information Technology Co., Ltd（上海流利说信息技术有限公司）
In this paper, we present an end-to-end automatic speech recognition system, which successfully employs subword units in a hybrid CTC-Attention based system. The subword units are obtained by the byte-pair encoding (BPE) compression algorithm. Compared to using words as modeling units, using characters or subword units does not suffer from the out-of-vocabulary (OOV) problem. Furthermore, using subword units further offers a capability in modeling longer context than using characters. We evaluate different systems over the LibriSpeech 1000h dataset. The subword-based hybrid CTC-Attention syste…
Continuous word representations, trained on large unlabeled corpora are useful for many natural language processing tasks. Popular models that learn such representations ignore the morphology of words, by assigning a distinct vector to each word. This is a limitation, especially for languages with large vocabularies and many rare words. In this paper, we propose a new approach based on the skipgram model, where each word is represented as a bag of character $n$-grams. A vector representation is associated to each character $n$-gram; words being represented as the sum of these representations. …
The use of subword-level information (e.g., characters, character n-grams, morphemes) has become ubiquitous in modern word representation learning. Its importance is attested especially for morphologically rich languages which generate a large number of rare words. Despite a steadily increasing interest in such subword-informed word representations, their systematic comparative analysis across typologically diverse languages and different tasks is still missing. In this work, we deliver such a study focusing on the variation of two crucial components required for subword-level integration into…
Most existing machine translation systems operate at the level of words, relying on explicit segmentation to extract tokens. We introduce a neural machine translation (NMT) model that maps a source character sequence to a target character sequence without any segmentation. We employ a character-level convolutional network with max-pooling at the encoder to reduce the length of source representation, allowing the model to be trained at a speed comparable to subword-level models while capturing local regularities. Our character-to-character model outperforms a recently proposed baseline with a s…
National Laboratory of Pattern Recognition, CASIA（中国科学院自动化研究所模式识别国家重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；CAS Center for Excellence in Brain Science and Intelligence Technology（中国科学院脑科学与智能技术卓越创新中心）
Neural machine translation (NMT), a new approach to machine translation, has been proved to outperform conventional statistical machine translation (SMT) across a variety of language pairs. Translation is an open-vocabulary problem, but most existing NMT systems operate with a fixed vocabulary, which causes the incapability of translating rare words. This problem can be alleviated by using different translation granularities, such as character, subword and hybrid word-character. Translation involving Chinese is one of the most difficult tasks in machine translation, however, to the best of our…
Singapore University of Technology and Design（新加坡科技设计大学）
We investigate a lattice LSTM network for Chinese word segmentation (CWS) to utilize words or subwords. It integrates the character sequence features with all subsequences information matched from a lexicon. The matched subsequences serve as information shortcut tunnels which link their start and end characters directly. Gated units are used to control the contribution of multiple input links. Through formula derivation and comparison, we show that the lattice LSTM is an extension of the standard LSTM with the ability to take multiple inputs. Previous lattice LSTM model takes word embeddings a…
RIKEN AIP（理化学研究所先进智能项目中心）；Tohoku University（东北大学）；Heidelberg Institute for Theoretical Studies gGmbH（海德堡理论研究所）
Pretrained contextual and non-contextual subword embeddings have become available in over 250 languages, allowing massively multilingual NLP. However, while there is no dearth of pretrained embeddings, the distinct lack of systematic evaluations makes it difficult for practitioners to choose between them. In this work, we conduct an extensive evaluation comparing non-contextual subword embeddings, namely FastText and BPEmb, and a contextual representation method, namely BERT, on multilingual named entity recognition and part-of-speech tagging. We find that overall, a combination of BERT, BPEmb…
Enhancing Sindhi Word Segmentation using Subword Representation Learning and Position-aware Self-attention
Institute of Business Management（商业管理学院）；Dalhousie University（达尔豪斯大学）；Heidelberg University（海德堡大学）；Bahria University（巴里亚大学）；College of Computer Science and Engineering（计算机科学与工程学院）；Taibah University（塔伊巴大学）；Faculty of Computing and Information Technology（计算机与信息技术学院）；King Abdulaziz University（阿卜杜勒阿齐兹国王大学）；School of Computer Science and Technology（计算机科学与技术学院）；Harbin Institute of Technology（哈尔滨工业大学）
This paper describes the UMDSub system that participated in Task 2 of SemEval-2018. We developed a system that predicts an emoji given the raw text in a English tweet. The system is a Multi-channel Convolutional Neural Network based on subword embeddings for the representation of tweets. This model improves on character or word based methods by about 2\%. Our system placed 21st of 48 participating systems in the official evaluation.
For different language pairs, word-level neural machine translation (NMT) models with a fixed-size vocabulary suffer from the same problem of representing out-of-vocabulary (OOV) words. The common practice usually replaces all these rare or unknown words with a <UNK> token, which limits the translation performance to some extent. Most of recent work handled such a problem by splitting words into characters or other specially extracted subword units to enable open-vocabulary translation. Byte pair encoding (BPE) is one of the successful attempts that has been shown extremely competitive by prov…
School of Informatics, University of Edinburgh（爱丁堡大学信息学院）；Samsung Electronics Polska sp. z o.o. - Samsung R&D Institute Poland（三星电子波兰有限公司——三星波兰研发中心）
Neural machine translation (NMT) models typically operate with a fixed vocabulary, but translation is an open-vocabulary problem. Previous work addresses the translation of out-of-vocabulary words by backing off to a dictionary. In this paper, we introduce a simpler and more effective approach, making the NMT model capable of open-vocabulary translation by encoding rare and unknown words as sequences of subword units. This is based on the intuition that various word classes are translatable via smaller units than words, for instance names (via character copying or transliteration), compounds (…
Standard pretrained language models operate on sequences of subword tokens without direct access to the characters that compose each token's string representation. We probe the embedding layer of pretrained language models and show that models learn the internal character composition of whole word and subword tokens to a surprising extent, without ever seeing the characters coupled with the tokens. Our results show that the embedding layer of RoBERTa holds enough information to accurately spell up to a third of the vocabulary and reach high average character ngram overlap on all token types. W…
University of Antwerp（安特卫普大学）；Computational Linguistics and Psycholinguistics Research Center（计算语言学与心理语言学研究中心）；Antwerp Centre for Digital Humanities and Literary Criticism（安特卫普数字人文与文学批评中心）
Neural machine translation (NMT) is nowadays commonly applied at the subword level, using byte-pair encoding. A promising alternative approach focuses on character-level translation, which simplifies processing pipelines in NMT considerably. This approach, however, must consider relatively longer sequences, rendering the training process prohibitively expensive. In this paper, we discuss a novel, Transformer-based approach, that we compare, both in speed and in quality to the Transformer at subword and character levels, as well as previously developed character-level models. We evaluate our mo…
Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Carnegie Mellon University（卡内基梅隆大学）
This paper proposes a novel approach to create an unit set for CTC based speech recognition systems. By using Byte Pair Encoding we learn an unit set of an arbitrary size on a given training text. In contrast to using characters or words as units this allows us to find a good trade-off between the size of our unit set and the available training data. We evaluate both Crossword units, that may span multiple word, and Subword units. By combining this approach with decoding methods using a separate language model we are able to achieve state of the art results for grapheme based CTC systems.
Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）；Amazon Alexa（亚马逊Alexa）
Named entity recognition (NER) is a vital task in spoken language understanding, which aims to identify mentions of named entities in text e.g., from transcribed speech. Existing neural models for NER rely mostly on dedicated word-level representations, which suffer from two main shortcomings. First, the vocabulary size is large, yielding large memory requirements and training time. Second, these models are not able to learn morphological or phonological representations. To remedy the above shortcomings, we adopt a neural solution based on bidirectional LSTMs and conditional random fields, whe…
We propose the first character-level white-box adversarial attack method against transformer models. The intuition of our method comes from the observation that words are split into subtokens before being fed into the transformer models and the substitution between two close subtokens has a similar effect to the character modification. Our method mainly contains three steps. First, a gradient-based method is adopted to find the most vulnerable words in the sentence. Then we split the selected words into subtokens to replace the origin tokenization result from the transformer tokenizer. Finally…
Université de Montréal（蒙特利尔大学）；New York University（纽约大学）
The existing machine translation systems, whether phrase-based or neural, have relied almost exclusively on word-level modelling with explicit segmentation. In this paper, we ask a fundamental question: can neural machine translation generate a character sequence without any explicit segmentation? To answer this question, we evaluate an attention-based encoder-decoder with a subword-level encoder and a character-level decoder on four language pairs--En-Cs, En-De, En-Ru and En-Fi-- using the parallel corpora from WMT'15. Our experiments show that the models with a character-level decoder outper…
In spoken Keyword Search, the query may contain out-of-vocabulary (OOV) words not observed when training the speech recognition system. Using subword language models (LMs) in the first-pass recognition makes it possible to recognize the OOV words, but even the subword n-gram LMs suffer from data sparsity. Recurrent Neural Network (RNN) LMs alleviate the sparsity problems but are not suitable for first-pass recognition as such. One way to solve this is to approximate the RNNLMs by back-off n-gram models. In this paper, we propose to interpolate the conventional n-gram models and the RNNLM appro…
We propose several ways of reusing subword embeddings and other weights in subword-aware neural language models. The proposed techniques do not benefit a competitive character-aware model, but some of them improve the performance of syllable- and morpheme-aware models while showing significant reductions in model sizes. We discover a simple hands-on principle: in a multi-layer input embedding model, layers should be tied consecutively bottom-up if reused at output. Our best morpheme-aware model with properly reused weights beats the competitive word-level model by a large margin across multipl…
In Neural Machine Translation (NMT) the usage of subwords and characters as source and target units offers a simple and flexible solution for translation of rare and unseen words. However, selecting the optimal subword segmentation involves a trade-off between expressiveness and flexibility, and is language and dataset-dependent. We present Block Multitask Learning (BMTL), a novel NMT architecture that predicts multiple targets of different granularities simultaneously, removing the need to search for the optimal segmentation strategy. Our multi-task model exhibits improvements of up to 1.7 BL…
University of Edinburgh（爱丁堡大学）；Pontificia Universidad Católica del Perú（秘鲁天主教大学）
Language modelling is regularly analysed at word, subword or character units, but syllables are seldom used. Syllables provide shorter sequences than characters, they can be extracted with rules, and their segmentation typically requires less specialised effort than identifying morphemes. We reconsider syllables for an open-vocabulary generation task in 20 languages. We use rule-based syllabification methods for five languages and address the rest with a hyphenation tool, which behaviour as syllable proxy is validated. With a comparable perplexity, we show that syllables outperform characters,…
Department of Computer Science and Engineering（计算机科学与工程系）；Ohio State University（俄亥俄州立大学）
Sentence pair modeling is critical for many NLP tasks, such as paraphrase identification, semantic textual similarity, and natural language inference. Most state-of-the-art neural models for these tasks rely on pretrained word embedding and compose sentence-level semantics in varied ways; however, few works have attempted to verify whether we really need pretrained embeddings in these tasks. In this paper, we study how effective subword-level (character and character n-gram) representations are in sentence pair modeling. Though it is well-known that subword models are effective in tasks with s…
Word embeddings improve generalization over lexical features by placing each word in a lower-dimensional space, using distributional information obtained from unlabeled data. However, the effectiveness of word embeddings for downstream NLP tasks is limited by out-of-vocabulary (OOV) words, for which embeddings do not exist. In this paper, we present MIMICK, an approach to generating OOV word embeddings compositionally, by learning a function from spellings to distributional embeddings. Unlike prior work, MIMICK does not require re-training on the original word embedding corpus; instead, learni…
In this work we tackle the problem of sentence boundary detection applied to French as a binary classification task ("sentence boundary" or "not sentence boundary"). We combine convolutional neural networks with subword-level information vectors, which are word embedding representations learned from Wikipedia that take advantage of the words morphology; so each word is represented as a bag of their character n-grams. We decide to use a big written dataset (French Gigaword) instead of standard size transcriptions to train and evaluate the proposed architectures with the intention of using the t…
Word vector representations enable machines to encode human language for spoken language understanding and processing. Confusion2vec, motivated from human speech production and perception, is a word vector representation which encodes ambiguities present in human spoken language in addition to semantics and syntactic information. Confusion2vec provides a robust spoken language representation by considering inherent human language ambiguities. In this paper, we propose a novel word vector space estimation by unsupervised learning on lattices output by an automatic speech recognition (ASR) syste…
Budapest University of Technology and Economics（布达佩斯技术与经济大学）；Institute for Computer Science and Control（计算机科学与控制研究所）
We examine the role of character patterns in three tasks: morphological analysis, lemmatization and copy. We use a modified version of the standard sequence-to-sequence model, where the encoder is a pattern matching network. Each pattern scores all possible N character long subwords (substrings) on the source side, and the highest scoring subword's score is used to initialize the decoder as well as the input to the attention mechanism. This method allows learning which subwords of the input are important for generating the output. By training the models on the same source but different target,…
The Blavatnik School of Computer Science, Tel Aviv University（特拉维夫大学布拉瓦特尼克计算机科学学院）；Facebook AI Research（Facebook人工智能研究院）
Many NLP models operate over sequences of subword tokens produced by hand-crafted tokenization rules and heuristic subword induction algorithms. A simple universal alternative is to represent every computerized text as a sequence of bytes via UTF-8, obviating the need for an embedding layer since there are fewer token types (256) than dimensions. Surprisingly, replacing the ubiquitous embedding layer with one-hot representations of each byte does not hurt performance; experiments on byte-to-byte machine translation from English to 10 different languages show a consistent improvement in BLEU, r…
T-FREE: Subword Tokenizer-Free Generative LLMs via Sparse Representations for Memory-Efficient Embeddings
Aleph Alpha；Technical University Darmstadt（达姆施塔特工业大学）；Hessian Center for Artificial Intelligence (hessian.AI)（黑森人工智能中心）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）
iFLYTEK Research（科大讯飞研究）；State Key Laboratory of Cognitive Intelligence（认知智能国家重点实验室）；Harbin Institute of Technology（哈尔滨工业大学）；Research Center for Social Computing and Information Retrieval (SCIR)（社会计算与信息检索研究中心（SCIR））；iFLYTEK AI Research (Hebei)（科大讯飞AI研究（河北））；University of Maryland（马里兰大学）
Most pre-trained language models (PLMs) construct word representations at subword level with Byte-Pair Encoding (BPE) or its variations, by which OOV (out-of-vocab) words are almost avoidable. However, those methods split a word into subword units and make the representation incomplete and fragile. In this paper, we propose a character-aware pre-trained language model named CharBERT improving on the previous methods (such as BERT, RoBERTa) to tackle these problems. We first construct the contextual word embedding for each token from the sequential character representations, then fuse the repre…
University of Edinburgh（爱丁堡大学）；Pontificia Universidad Católica del Perú（秘鲁天主教教皇大学）
Language modelling and machine translation tasks mostly use subword or character inputs, but syllables are seldom used. Syllables provide shorter sequences than characters, require less-specialised extracting rules than morphemes, and their segmentation is not impacted by the corpus size. In this study, we first explore the potential of syllables for open-vocabulary language modelling in 21 languages. We use rule-based syllabification methods for six languages and address the rest with hyphenation, which works as a syllabification proxy. With a comparable perplexity, we show that syllables out…
Subword tokenization is a commonly used input pre-processing step in most recent NLP models. However, it limits the models' ability to leverage end-to-end task learning. Its frequency-based vocabulary creation compromises tokenization in low-resource languages, leading models to produce suboptimal representations. Additionally, the dependency on a fixed vocabulary limits the subword models' adaptability across languages and domains. In this work, we propose a vocabulary-free neural tokenizer by distilling segmentation information from heuristic-based subword tokenization. We pre-train our char…
