In this work, we represent CMV-BERT, which improves the pretraining of a language model via two ingredients: (a) contrastive learning, which is well studied in the area of computer vision; (b) multiple vocabularies, one of which is fine-grained and the other is coarse-grained. The two methods both provide different views of an original sentence, and both are shown to be beneficial. Downstream tasks demonstrate our proposed CMV-BERT are effective in improving the pretrained language models.
Despite the development of pre-trained language models (PLMs) significantly raise the performances of various Chinese natural language processing (NLP) tasks, the vocabulary for these Chinese PLMs remain to be the one provided by Google Chinese Bert \cite{devlin2018bert}, which is based on Chinese characters. Second, the masked language model pre-training is based on a single vocabulary, which limits its downstream task performances. In this work, we first propose a novel method, \emph{seg\_tok}, to form the vocabulary of Chinese BERT, with the help of Chinese word segmentation (CWS) and subwo…
University of Pittsburgh（匹兹堡大学）；Google Research（谷歌研究院）；Kuaishou Technology（快手科技）；Google Brain（谷歌大脑）
Pretrained language models like BERT have achieved good results on NLP tasks, but are impractical on resource-limited devices due to memory footprint. A large fraction of this footprint comes from the input embeddings with large input vocabulary and embedding dimensions. Existing knowledge distillation methods used for model compression cannot be directly applied to train student models with reduced vocabulary sizes. To this end, we propose a distillation method to align the teacher and student embeddings via mixed-vocabulary training. Our method compresses BERT-LARGE to a task-agnostic model …
We present IndoBERTweet, the first large-scale pretrained model for Indonesian Twitter that is trained by extending a monolingually-trained Indonesian BERT model with additive domain-specific vocabulary. We focus in particular on efficient model adaptation under vocabulary mismatch, and benchmark different ways of initializing the BERT embedding layer for new word types. We find that initializing with the average BERT subword embedding makes pretraining five times faster, and is more effective than proposed methods for vocabulary adaptation in terms of extrinsic evaluation over seven Twitter-b…
LTCI（通信与信息处理实验室）；Télécom Paris（巴黎电信学院）；Institut Polytechnique de Paris（巴黎理工学院）；Soda（Soda实验室）；Inria Saclay（法国国家信息与自动化研究所萨克莱中心）；CEA（法国原子能和替代能源委员会）；Université Paris-Saclay（巴黎萨克雷大学）
State-of-the-art NLP systems represent inputs with word embeddings, but these are brittle when faced with Out-of-Vocabulary (OOV) words. To address this issue, we follow the principle of mimick-like models to generate vectors for unseen words, by learning the behavior of pre-trained embeddings using only the surface form of words. We present a simple contrastive learning framework, LOVE, which extends the word representation of an existing pre-trained language model (such as BERT), and makes it robust to OOV with few additional parameters. Extensive evaluations demonstrate that our lightweight…
Can pretrained language models (PLMs) generate derivationally complex words? We present the first study investigating this question, taking BERT as the example PLM. We examine BERT's derivational capabilities in different settings, ranging from using the unmodified pretrained model to full finetuning. Our best model, DagoBERT (Derivationally and generatively optimized BERT), clearly outperforms the previous state of the art in derivation generation (DG). Furthermore, our experiments show that the input segmentation crucially impacts BERT's derivational knowledge, suggesting that the performanc…
How does the input segmentation of pretrained language models (PLMs) affect their interpretations of complex words? We present the first study investigating this question, taking BERT as the example PLM and focusing on its semantic representations of English derivatives. We show that PLMs can be interpreted as serial dual-route models, i.e., the meanings of complex words are either stored or else need to be computed from the subwords, which implies that maximally meaningful input tokens should allow for the best generalization on new words. This hypothesis is confirmed by a series of semantic …
Center for Information and Language Processing (CIS), LMU Munich（慕尼黑大学语言与信息处理中心（CIS））
Recent research investigates factual knowledge stored in large pretrained language models (PLMs). Instead of structural knowledge base (KB) queries, masked sentences such as "Paris is the capital of [MASK]" are used as probes. The good performance on this analysis task has been interpreted as PLMs becoming potential repositories of factual knowledge. In experiments across ten linguistically diverse languages, we study knowledge contained in static embeddings. We show that, when restricting the output space to a candidate set, simple nearest neighbor matching using static embeddings performs be…
Sofia University “St. Kliment Ohridski”（圣克莱门特·奥里季斯基索菲亚大学）；Qatar Computing Research Institute, HBKU（卡塔尔计算研究所（哈马德·本·哈利法大学））
Detecting the user's intent and finding the corresponding slots among the utterance's words are important tasks in natural language understanding. Their interconnected nature makes their joint modeling a standard part of training such models. Moreover, data scarceness and specialized vocabularies pose additional challenges. Recently, the advances in pre-trained language models, namely contextualized models such as ELMo and BERT have revolutionized the field by tapping the potential of training very large models with just a few steps of fine-tuning on a task-specific dataset. Here, we leverage …
Large pre-trained models have achieved great success in many natural language processing tasks. However, when they are applied in specific domains, these models suffer from domain shift and bring challenges in fine-tuning and online serving for latency and capacity constraints. In this paper, we present a general approach to developing small, fast and effective pre-trained models for specific domains. This is achieved by adapting the off-the-shelf general pre-trained models and performing task-agnostic knowledge distillation in target domains. Specifically, we propose domain-specific vocabular…
School of Foreign Languages and Cultures, Xiamen University（厦门大学外文学院）；DeepAI Tech Co., Ltd（深湃信息科技（深圳）有限公司）
The automatic classification is a process of automatically assigning text documents to predefined categories. An accurate automatic patent classifier is crucial to patent inventors and patent examiners in terms of intellectual property protection, patent management, and patent information retrieval. We present BERT-CNN, a hierarchical patent classifier based on pre-trained language model by training the national patent application documents collected from the State Information Center, China. The experimental results show that BERT-CNN achieves 84.3% accuracy, which is far better than the two c…
The pre-trained language models have achieved great successes in various natural language understanding (NLU) tasks due to its capacity to capture the deep contextualized information in text by pre-training on large-scale corpora. One of the fundamental components in pre-trained language models is the vocabulary, especially for training multilingual models on many different languages. In the technical report, we present our practices on training multilingual pre-trained language models with BBPE: Byte-Level BPE (i.e., Byte Pair Encoding). In the experiment, we adopted the architecture of NEZHA…
The standard BERT adopts subword-based tokenization, which may break a word into two or more wordpieces (e.g., converting "lossless" to "loss" and "less"). This will bring inconvenience in following situations: (1) what is the best way to obtain the contextual vector of a word that is divided into multiple wordpieces? (2) how to predict a word via cloze test without knowing the number of wordpieces in advance? In this work, we explore the possibility of developing BERT-style pretrained model over a vocabulary of words instead of wordpieces. We call such word-level BERT model as WordBERT. We tr…
INESC TEC（葡萄牙国家工程与系统创新研究所）；University of Porto（波尔图大学）；Cardiff University（卡迪夫大学）
Progress on commonsense reasoning is usually measured from performance improvements on Question Answering tasks designed to require commonsense knowledge. However, fine-tuning large Language Models (LMs) on these specific tasks does not directly evaluate commonsense learned during pre-training. The most direct assessments of commonsense knowledge in pre-trained LMs are arguably cloze-style tasks targeting commonsense assertions (e.g., A pen is used for [MASK].). However, this approach is restricted by the LM's vocabulary available for masked predictions, and its precision is subject to the con…
Recent developments in Natural Language Processing have led to the introduction of state-of-the-art Neural Language Models, enabled with unsupervised transferable learning, using different pretraining objectives. While these models achieve excellent results on the downstream NLP tasks, various domain adaptation techniques can improve their performance on domain-specific tasks. We compare and analyze the pretrained Neural Language Models, XLNet (autoregressive), and BERT (autoencoder) on the Legal Tasks. Results show that XLNet Model performs better on our Sequence Classification task of Legal …
To understand what kinds of linguistic knowledge are encoded by pretrained Chinese language models (LMs), we introduce the benchmark of Sino LINGuistics (SLING), which consists of 38K minimal sentence pairs in Mandarin Chinese grouped into 9 high-level linguistic phenomena. Each pair demonstrates the acceptability contrast of a specific syntactic or semantic phenomenon (e.g., The keys are lost vs. The keys is lost), and an LM should assign lower perplexity to the acceptable sentence. In contrast to the CLiMP dataset (Xiang et al., 2021), which also contains Chinese minimal pairs and was create…
Since the appearance of BERT, recent works including XLNet and RoBERTa utilize sentence embedding models pre-trained by large corpora and a large number of parameters. Because such models have large hardware and a huge amount of data, they take a long time to pre-train. Therefore it is important to attempt to make smaller models that perform comparatively. In this paper, we trained a Korean-specific model KR-BERT, utilizing a smaller vocabulary and dataset. Since Korean is one of the morphologically rich languages with poor resources using non-Latin alphabets, it is also important to capture l…
The content on the web is in a constant state of flux. New entities, issues, and ideas continuously emerge, while the semantics of the existing conversation topics gradually shift. In recent years, pre-trained language models like BERT greatly improved the state-of-the-art for a large spectrum of content understanding tasks. Therefore, in this paper, we aim to study how these language models can be adapted to better handle continuously evolving web content. In our study, we first analyze the evolution of 2013 - 2019 Twitter data, and unequivocally confirm that a BERT model trained on past twee…
Named entity recognition (NER) is frequently addressed as a sequence classification task where each input consists of one sentence of text. It is nevertheless clear that useful information for the task can often be found outside of the scope of a single-sentence context. Recently proposed self-attention models such as BERT can both efficiently capture long-distance relationships in input as well as represent inputs consisting of several sentences, creating new opportunitites for approaches that incorporate cross-sentence information in natural language processing tasks. In this paper, we prese…
University of Copenhagen（哥本哈根大学）；ROCKWOOL Foundation Research Unit（洛克伍德基金会研究部）；Johns Hopkins University（约翰斯·霍普金斯大学）；KU Leuven（荷语鲁汶大学）；Pioneer Centre for AI（先锋人工智能中心）
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…
Learning to represent free text is a core task in many clinical machine learning (ML) applications, as clinical text contains observations and plans not otherwise available for inference. State-of-the-art methods use large language models developed with immense computational resources and training data; however, applying these models is challenging because of the highly varying syntax and vocabulary in clinical free text. Structured information such as International Classification of Disease (ICD) codes often succinctly abstracts the most important facts of a clinical encounter and yields good…
Fine-Tuning Large Language Models for Scientific Text Classification: A Comparative Study
Doctoral School of Applied Informatics（应用信息学博士学院）；Applied Mathematics Óbuda University Budapest, Hungary（应用数学 奥布达大学 布达佩斯 匈牙利）；John von Neumann Faculty of Informatics（约翰·冯·诺依曼信息学学院）
Paul G. Allen School of Computer Science & Engineering, University of Washington（华盛顿大学保罗·G·艾伦计算机科学与工程学院）；Department of Linguistics, University of Washington（华盛顿大学语言学系）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）
Pretrained multilingual contextual representations have shown great success, but due to the limits of their pretraining data, their benefits do not apply equally to all language varieties. This presents a challenge for language varieties unfamiliar to these models, whose labeled \emph{and unlabeled} data is too limited to train a monolingual model effectively. We propose the use of additional language-specific pretraining and vocabulary augmentation to adapt multilingual models to low-resource settings. Using dependency parsing of four diverse low-resource language varieties as a case study, w…
Evaluation Of P300 Speller Performance Using Large Language Models Along With Cross-Subject Training
University of Illinois at Urbana Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Washington（华盛顿大学）；UCLA（加州大学洛杉矶分校）；PE Investments（PE投资公司）；MIT（麻省理工学院）；University of Texas, Southwestern（德克萨斯大学西南医学中心）
Language models based on deep neural networks have facilitated great advances in natural language processing and understanding tasks in recent years. While models covering a large number of languages have been introduced, their multilinguality has come at a cost in terms of monolingual performance, and the best-performing models at most tasks not involving cross-lingual transfer remain monolingual. In this paper, we consider the question of whether it is possible to pre-train a bilingual model for two remotely related languages without compromising performance at either language. We collect pr…
German Research Centre for Artificial Intelligence (DFKI)（德国人工智能研究中心（DFKI））；phase-6 GmbH（phase-6有限公司）
Vocabulary learning is vital to foreign language learning. Correct and adequate feedback is essential to successful and satisfying vocabulary training. However, many vocabulary and language evaluation systems perform on simple rules and do not account for real-life user learning data. This work introduces Multi-Language Vocabulary Evaluation Data Set (MuLVE), a data set consisting of vocabulary cards and real-life user answers, labeled indicating whether the user answer is correct or incorrect. The data source is user learning data from the Phase6 vocabulary trainer. The data set contains voca…
Clinical interactions are initially recorded and documented in free text medical notes. ICD coding is the task of classifying and coding all diagnoses, symptoms and procedures associated with a patient's visit. The process is often manual and extremely time-consuming and expensive for hospitals. In this paper, we propose a machine learning model, BERT-XML, for large scale automated ICD coding from EHR notes, utilizing recently developed unsupervised pretraining that have achieved state of the art performance on a variety of NLP tasks. We train a BERT model from scratch on EHR notes, learning w…
Large Transformer-based language models such as BERT have led to broad performance improvements on many NLP tasks. Domain-specific variants of these models have demonstrated excellent performance on a variety of specialised tasks. In legal NLP, BERT-based models have led to new state-of-the-art results on multiple tasks. The exploration of these models has demonstrated the importance of capturing the specificity of the legal language and its vocabulary. However, such approaches suffer from high computational costs, leading to a higher ecological impact and lower accessibility. Our findings, fo…
This paper describes our submission on the COVID-19 literature annotation task at Biocreative VII. We proposed an approach that exploits the knowledge of the globally non-optimal weights, usually rejected, to build a rich representation of each label. Our proposed approach consists of two stages: (1) A bagging of various initializations of the training data that features weakly trained weights, (2) A stacking of heterogeneous vocabulary models based on BERT and RoBERTa Embeddings. The aggregation of these weak insights performs better than a classical globally efficient model. The purpose is t…
Dublin City University（都柏林城市大学）；ADAPT Centre（ADAPT中心）；Fiontar & Scoil na Gaeilge（芬塔尔与爱尔兰语学院）
The BERT family of neural language models have become highly popular due to their ability to provide sequences of text with rich context-sensitive token encodings which are able to generalise well to many NLP tasks. We introduce gaBERT, a monolingual BERT model for the Irish language. We compare our gaBERT model to multilingual BERT and the monolingual Irish WikiBERT, and we show that gaBERT provides better representations for a downstream parsing task. We also show how different filtering criteria, vocabulary size and the choice of subword tokenisation model affect downstream performance. We …
Sun Yat-sen University（中山大学）；National Institutes of Health（美国国立卫生研究院）；University of Pennsylvania（宾夕法尼亚大学）
Pretrained language models such as Bidirectional Encoder Representations from Transformers (BERT) have achieved state-of-the-art performance in natural language processing (NLP) tasks. Recently, BERT has been adapted to the biomedical domain. Despite the effectiveness, these models have hundreds of millions of parameters and are computationally expensive when applied to large-scale NLP applications. We hypothesized that the number of parameters of the original BERT can be dramatically reduced with minor impact on performance. In this study, we present Bioformer, a compact BERT model for biomed…
The Word and the Way: Strategies for Domain-Specific BERT Pre-Training in German Medical NLP
School of Computation, Information and Technology（计算信息科技学院）；Technical University of Munich（慕尼黑技术大学）；Chair of IT Infrastructure for Translational Medical Research（转化医学研究信息基础设施主任）；Faculty of Applied Computer Science（应用计算机科学学院）；University of Augsburg（艾希施泰特大学）；Institute of General Practice（基础医学研究所）；Faculty of Medicine and Medical Center（医学学院和医学中心）
Pre-trained Transformer-based models are achieving state-of-the-art results on a variety of Natural Language Processing data sets. However, the size of these models is often a drawback for their deployment in real production applications. In the case of multilingual models, most of the parameters are located in the embeddings layer. Therefore, reducing the vocabulary size should have an important impact on the total number of parameters. In this paper, we propose to generate smaller models that handle fewer number of languages according to the targeted corpora. We present an evaluation of smal…
While transfer learning has become a ubiquitous technique used across Natural Language Processing (NLP) tasks, it is often unable to replicate the performance of pre-trained models on text of niche domains like Automotive. In this paper we aim to understand the main characteristics of the distribution shift with automotive domain text (describing technical functionalities such as Cruise Control) and attempt to explain the potential reasons for the gap in performance. We focus on performing the Named Entity Recognition (NER) task as it requires strong lexical, syntactic and semantic understandi…
HiTZ Center, University of the Basque Country (UPV/EHU)（巴斯克大学 HiTZ 中心）；DeepMind
State-of-the-art unsupervised multilingual models (e.g., multilingual BERT) have been shown to generalize in a zero-shot cross-lingual setting. This generalization ability has been attributed to the use of a shared subword vocabulary and joint training across multiple languages giving rise to deep multilingual abstractions. We evaluate this hypothesis by designing an alternative approach that transfers a monolingual model to new languages at the lexical level. More concretely, we first train a transformer-based masked language model on one language, and transfer it to a new language by learnin…
University of New South Wales（新南威尔士大学）；Indian Institute of Technology Guwahati（印度理工学院古瓦哈蒂分校）
It is well known that translations of songs and poems not only break rhythm and rhyming patterns, but can also result in loss of semantic information. The Bhagavad Gita is an ancient Hindu philosophical text originally written in Sanskrit that features a conversation between Lord Krishna and Arjuna prior to the Mahabharata war. The Bhagavad Gita is also one of the key sacred texts in Hinduism and is known as the forefront of the Vedic corpus of Hinduism. In the last two centuries, there has been a lot of interest in Hindu philosophy from western scholars; hence, the Bhagavad Gita has been tran…
While self-supervised learning has made rapid advances in natural language processing, it remains unclear when researchers should engage in resource-intensive domain-specific pretraining (domain pretraining). The law, puzzlingly, has yielded few documented instances of substantial gains to domain pretraining in spite of the fact that legal language is widely seen to be unique. We hypothesize that these existing results stem from the fact that existing legal NLP tasks are too easy and fail to meet conditions for when domain pretraining can help. To address this, we first present CaseHOLD (Case …
