Recent advancements in the NLP field showed that transfer learning helps with achieving state-of-the-art results for new tasks by tuning pre-trained models instead of starting from scratch. Transformers have made a significant improvement in creating new state-of-the-art results for many NLP tasks including but not limited to text classification, text generation, and sequence labeling. Most of these success stories were based on large datasets. In this paper we focus on a real-life scenario that scientists in academia and industry face frequently: given a small dataset, can we use a large pre-trained model like BERT and get better results than simple models? To answer this question, we use a small dataset for intent classification collected for building chatbots and compare the performance of a simple bidirectional LSTM model with a pre-trained BERT model. Our experimental results show that bidirectional LSTM models can achieve significantly higher results than a BERT model for a small dataset and these simple models get trained in much less time than tuning the pre-trained counterparts. We conclude that the performance of a model is dependent on the task and the data, and therefore before making a model choice, these factors should be taken into consideration instead of directly choosing the most popular model.
MahaParaphrase: A Marathi Paraphrase Detection Corpus and BERT-based Models
Pune Institute of Computer Technology（普那计算机技术研究所）；MKSSS’ Cummins College of Engineering for Women（MKSSS女子工程学院）；Indian Institute of Technology Madras（印度理工学院马德拉斯分校）；L3Cube Labs（L3Cube实验室）
University of Massachusetts, Amherst（马萨诸塞大学阿默斯特分校）
Phrase representations derived from BERT often do not exhibit complex phrasal compositionality, as the model relies instead on lexical similarity to determine semantic relatedness. In this paper, we propose a contrastive fine-tuning objective that enables BERT to produce more powerful phrase embeddings. Our approach (Phrase-BERT) relies on a dataset of diverse phrasal paraphrases, which is automatically generated using a paraphrase generation model, as well as a large-scale dataset of phrases in context mined from the Books3 corpus. Phrase-BERT outperforms baselines across a variety of phrase-…
Microsoft Corporation（微软公司）；Coleridge Initiative（科尔里奇倡议组织）
Many prior language modeling efforts have shown that pre-training on an in-domain corpus can significantly improve performance on downstream domain-specific NLP tasks. However, the difficulties associated with collecting enough in-domain data might discourage researchers from approaching this pre-training task. In this paper, we conducted a series of experiments by pre-training Bidirectional Encoder Representations from Transformers (BERT) with different sizes of biomedical corpora. The results demonstrate that pre-training on a relatively small amount of in-domain data (4GB) with limited trai…
This paper presents results of our experiments for the next utterance ranking on the Ubuntu Dialog Corpus -- the largest publicly available multi-turn dialog corpus. First, we use an in-house implementation of previously reported models to do an independent evaluation using the same data. Second, we evaluate the performances of various LSTMs, Bi-LSTMs and CNNs on the dataset. Third, we create an ensemble by averaging predictions of multiple models. The ensemble further improves the performance and it achieves a state-of-the-art result for the next utterance ranking on this dataset. Finally, we…
Rootroo Ltd（Rootroo有限公司）；University of Helsinki（赫尔辛基大学）；Mahidol University（玛希隆大学）
We present the first openly available corpus for detecting depression in Thai. Our corpus is compiled by expert verified cases of depression in several online blogs. We experiment with two different LSTM based models and two different BERT based models. We achieve a 77.53\% accuracy with a Thai BERT model in detecting depression. This establishes a good baseline for future researcher on the same corpus. Furthermore, we identify a need for Thai embeddings that have been trained on a more varied corpus than Wikipedia. Our corpus, code and trained models have been released openly on Zenodo.
Jordan University of Science and Technology（约旦科技大学）
This research presents our team KEIS@JUST participation at SemEval-2020 Task 12 which represents shared task on multilingual offensive language. We participated in all the provided languages for all subtasks except sub-task-A for the English language. Two main approaches have been developed the first is performed to tackle both languages Arabic and English, a weighted ensemble consists of Bi-GRU and CNN followed by Gaussian noise and global pooling layer multiplied by weights to improve the overall performance. The second is performed for other languages, a transfer learning from BERT beside t…
Toxic comments in online platforms are an unavoidable social issue under the cloak of anonymity. Hate speech detection has been actively done for languages such as English, German, or Italian, where manually labeled corpus has been released. In this work, we first present 9.4K manually labeled entertainment news comments for identifying Korean toxic speech, collected from a widely used online news platform in Korea. The comments are annotated regarding social bias and hate speech since both aspects are correlated. The inter-annotator agreement Krippendorff's alpha score is 0.492 and 0.496, res…
In this paper, we introduce the Chinese corpus from CLUE organization, CLUECorpus2020, a large-scale corpus that can be used directly for self-supervised learning such as pre-training of a language model, or language generation. It has 100G raw corpus with 35 billion Chinese characters, which is retrieved from Common Crawl. To better understand this corpus, we conduct language understanding experiments on both small and large scale, and results show that the models trained on this corpus can achieve excellent performance on Chinese. We release a new Chinese vocabulary with a size of 8K, which …
Faculty of Engineering, Alexandria University（亚历山大大学工程学院）
Arabic is one of the most important and growing languages in the world. With the rise of social media platforms such as Twitter, Arabic spoken dialects have become more in use. In this paper, we describe our approach on the NADI Shared Task 1 that requires us to build a system to differentiate between different 21 Arabic dialects, we introduce a deep learning semi-supervised fashion approach along with pre-processing that was reported on NADI shared Task 1 Corpus. Our system ranks 4th in NADI's shared task competition achieving a 23.09% F1 macro average score with a simple yet efficient approa…
Dialogue act (DA) classification has been studied for the past two decades and has several key applications such as workflow automation and conversation analytics. Researchers have used, to address this problem, various traditional machine learning models, and more recently deep neural network models such as hierarchical convolutional neural networks (CNNs) and long short-term memory (LSTM) networks. In this paper, we introduce a new model architecture, directed-acyclic-graph LSTM (DAG-LSTM) for DA classification. A DAG-LSTM exploits the turn-taking structure naturally present in a multi-party…
University of São Paulo（圣保罗大学）；Instituto de Matemática e Estatística（数学与统计研究所）
At present, different deep learning models are presenting high accuracy on popular inference datasets such as SNLI, MNLI, and SciTail. However, there are different indicators that those datasets can be exploited by using some simple linguistic patterns. This fact poses difficulties to our understanding of the actual capacity of machine learning models to solve the complex task of textual inference. We propose a new set of syntactic tasks focused on contradiction detection that require specific capacities over linguistic logical forms such as: Boolean coordination, quantifiers, definite descrip…
RAAMove: A Corpus for Analyzing Moves in Research Article Abstracts
Faculty of Information Technology, Beijing University of Technology（北京工业大学信息技术学院）；School of Computer Science, Beijing Institute of Technology（北京理工大学计算机学院）
This study presents a new dataset on rumor detection in Finnish language news headlines. We have evaluated two different LSTM based models and two different BERT models, and have found very significant differences in the results. A fine-tuned FinBERT reaches the best overall accuracy of 94.3% and rumor label accuracy of 96.0% of the time. However, a model fine-tuned on Multilingual BERT reaches the best factual label accuracy of 97.2%. Our results suggest that the performance difference is due to a difference in the original training data. Furthermore, we find that a regular LSTM model works b…
The Arabic language suffers from a great shortage of datasets suitable for training deep learning models, and the existing ones include general non-specialized classifications. In this work, we introduce a new Arab medical dataset, which includes two thousand medical documents collected from several Arabic medical websites, in addition to the Arab Medical Encyclopedia. The dataset was built for the task of classifying texts and includes 10 classes (Blood, Bone, Cardiovascular, Ear, Endocrine, Eye, Gastrointestinal, Immune, Liver and Nephrological) diseases. Experiments on the dataset were perf…
Ubuntu dialogue corpus is the largest public available dialogue corpus to make it feasible to build end-to-end deep neural network models directly from the conversation data. One challenge of Ubuntu dialogue corpus is the large number of out-of-vocabulary words. In this paper we proposed a method which combines the general pre-trained word embedding vectors with those generated on the task-specific training set to address this issue. We integrated character embedding into Chen et al's Enhanced LSTM method (ESIM) and used it to evaluate the effectiveness of our proposed method. For the task of …
Recognizing causal elements and causal relations in text is one of the challenging issues in natural language processing; specifically, in low resource languages such as Persian. In this research we prepare a causality human annotated corpus for the Persian language which consists of 4446 sentences and 5128 causal relations and three labels of cause, effect and causal mark -- if possibl -- are specified for each relation. We have used this corpus to train a system for detecting causal elements boundaries. Also, we present a causality detection benchmark for three machine learning methods and t…
Russian-Armenian University（俄罗斯-亚美尼亚大学）；Ivannikov Laboratory for System Programming（伊万尼科夫系统编程实验室）
In this work, we employ a semi-automatic method based on back translation to generate a sentential paraphrase corpus for the Armenian language. The initial collection of sentences is translated from Armenian to English and back twice, resulting in pairs of lexically distant but semantically similar sentences. The generated paraphrases are then manually reviewed and annotated. Using the method train and test datasets are created, containing 2360 paraphrases in total. In addition, the datasets are used to train and evaluate BERTbased models for detecting paraphrase in Armenian, achieving results…
This work presents a new resource for borrowing identification and analyzes the performance and errors of several models on this task. We introduce a new annotated corpus of Spanish newswire rich in unassimilated lexical borrowings -- words from one language that are introduced into another without orthographic adaptation -- and use it to evaluate how several sequence labeling models (CRF, BiLSTM-CRF, and Transformer-based models) perform. The corpus contains 370,000 tokens and is larger, more borrowing-dense, OOV-rich, and topic-varied than previous corpora available for this task. Our result…
University of Minnesota（明尼苏达大学）；Arizona State University（亚利桑那州立大学）；Mayo Clinic（梅奥诊所）
Since no effective therapies exist for Alzheimer's disease (AD), prevention has become more critical through lifestyle factor changes and interventions. Analyzing electronic health records (EHR) of patients with AD can help us better understand lifestyle's effect on AD. However, lifestyle information is typically stored in clinical narratives. Thus, the objective of the study was to demonstrate the feasibility of natural language processing (NLP) models to classify lifestyle factors (e.g., physical activity and excessive diet) from clinical texts. We automatically generated labels for the trai…
This paper describes the system submitted by ANA Team for the SemEval-2019 Task 3: EmoContext. We propose a novel Hierarchical LSTMs for Contextual Emotion Detection (HRLCE) model. It classifies the emotion of an utterance given its conversational context. The results show that, in this task, our HRCLE outperforms the most recent state-of-the-art text classification framework: BERT. We combine the results generated by BERT and HRCLE to achieve an overall score of 0.7709 which ranked 5th on the final leader board of the competition among 165 Teams.
We focus on multi-turn response selection in a retrieval-based dialog system. In this paper, we utilize the powerful pre-trained language model Bi-directional Encoder Representations from Transformer (BERT) for a multi-turn dialog system and propose a highly effective post-training method on domain-specific corpus. Although BERT is easily adopted to various NLP tasks and outperforms previous baselines of each task, it still has limitations if a task corpus is too focused on a certain domain. Post-training on domain-specific corpus (e.g., Ubuntu Corpus) helps the model to train contextualized r…
Indian Institute of Technology Madras（印度理工学院马德拉斯分校）；L3Cube
We present L3Cube-MahaCorpus a Marathi monolingual data set scraped from different internet sources. We expand the existing Marathi monolingual corpus with 24.8M sentences and 289M tokens. We further present, MahaBERT, MahaAlBERT, and MahaRoBerta all BERT-based masked language models, and MahaFT, the fast text word embeddings both trained on full Marathi corpus with 752M tokens. We show the effectiveness of these resources on downstream Marathi sentiment analysis, text classification, and named entity recognition (NER) tasks. We also release MahaGPT, a generative Marathi GPT model trained on M…
Fake news detection is a challenging task aiming to reduce human time and effort to check the truthfulness of news. Automated approaches to combat fake news, however, are limited by the lack of labeled benchmark datasets, especially in languages other than English. Moreover, many publicly available corpora have specific limitations that make them difficult to use. To address this problem, our contribution is threefold. First, we propose a new, publicly available German topic related corpus for fake news detection. To the best of our knowledge, this is the first corpus of its kind. In this rega…
Explainable question answering for science questions is a challenging task that requires multi-hop inference over a large set of fact sentences. To counter the limitations of methods that view each query-document pair in isolation, we propose the LSTM-Interleaved Transformer which incorporates cross-document interactions for improved multi-hop ranking. The LIT architecture can leverage prior ranking positions in the re-ranking setting. Our model is competitive on the current leaderboard for the TextGraphs 2020 shared task, achieving a test-set MAP of 0.5607, and would have gained third place h…
Web search engines focus on serving highly relevant results within hundreds of milliseconds. Pre-trained language transformer models such as BERT are therefore hard to use in this scenario due to their high computational demands. We present our real-time approach to the document ranking problem leveraging a BERT-based siamese architecture. The model is already deployed in a commercial search engine and it improves production performance by more than 3%. For further research and evaluation, we release DaReCzech, a unique data set of 1.6 million Czech user query-document pairs with manually assi…
Guangdong University of Technology（广东工业大学）；Guangdong Ocean University（广东海洋大学）；University of Surrey（萨里大学）；Tianjin University of Technology（天津理工大学）
Visual storytelling is a creative and challenging task, aiming to automatically generate a story-like description for a sequence of images. The descriptions generated by previous visual storytelling approaches lack coherence because they use word-level sequence generation methods and do not adequately consider sentence-level dependencies. To tackle this problem, we propose a novel hierarchical visual storytelling framework which separately models sentence-level and word-level semantics. We use the transformer-based BERT to obtain embeddings for sentences and words. We then employ a hierarchica…
In this work, we study leveraging extra text data to improve low-resource end-to-end ASR under cross-lingual transfer learning setting. To this end, we extend our prior work [1], and propose a hybrid Transformer-LSTM based architecture. This architecture not only takes advantage of the highly effective encoding capacity of the Transformer network but also benefits from extra text data due to the LSTM-based independent language model network. We conduct experiments on our in-house Malay corpus which contains limited labeled data and a large amount of extra text. Results show that the proposed a…
In the present paper, two models are presented namely LSTM-CRF and BERT-LSTM-CRF for semantic tagging of universal semantic tag dataset. The experiments show that the first model is much easier to converge while the second model that leverages BERT embedding, takes a long time to converge and needs a big dataset for semtagging to be effective.
ESRC Business and Local Government Data Research Centre（ESRC商业与地方政府数据研究中心）
We build a sentence-level political discourse classifier using existing human expert annotated corpora of political manifestos from the Manifestos Project (Volkens et al., 2020a) and applying them to a corpus ofCOVID-19Press Briefings (Chatsiou, 2020). We use manually annotated political manifestos as training data to train a local topic ConvolutionalNeural Network (CNN) classifier; then apply it to the COVID-19PressBriefings Corpus to automatically classify sentences in the test corpus.We report on a series of experiments with CNN trained on top of pre-trained embeddings for sentence-level cl…
Aston University（阿斯顿大学）；Folding Space（Folding Space公司）；University of Wolverhampton（伍尔弗汉普顿大学）
In this paper, an approach for hate speech detection against women in Arabic community on social media (e.g. Youtube) is proposed. In the literature, similar works have been presented for other languages such as English. However, to the best of our knowledge, not much work has been conducted in the Arabic language. A new hate speech corpus (Arabic\_fr\_en) is developed using three different annotators. For corpus validation, three different machine learning algorithms are used, including deep Convolutional Neural Network (CNN), long short-term memory (LSTM) network and Bi-directional LSTM (Bi-…
Pre-trained language model word representation, such as BERT, have been extremely successful in several Natural Language Processing tasks significantly improving on the state-of-the-art. This can largely be attributed to their ability to better capture semantic information contained within a sentence. Several tasks, however, can benefit from information available at a corpus level, such as Term Frequency-Inverse Document Frequency (TF-IDF). In this work we test the effectiveness of integrating this information with BERT on the task of identifying abuse on social media and show that integrating…
This paper presents Wojood, a corpus for Arabic nested Named Entity Recognition (NER). Nested entities occur when one entity mention is embedded inside another entity mention. Wojood consists of about 550K Modern Standard Arabic (MSA) and dialect tokens that are manually annotated with 21 entity types including person, organization, location, event and date. More importantly, the corpus is annotated with nested entities instead of the more common flat annotations. The data contains about 75K entities and 22.5% of which are nested. The inter-annotator evaluation of the corpus demonstrated a str…
University of Malta（马耳他大学）；Utrecht University（乌得勒支大学）；Idiap Research Institute（Idiap研究所）
Multilingual language models such as mBERT have seen impressive cross-lingual transfer to a variety of languages, but many languages remain excluded from these models. In this paper, we analyse the effect of pre-training with monolingual data for a low-resource language that is not included in mBERT -- Maltese -- with a range of pre-training set ups. We conduct evaluations with the newly pre-trained models on three morphosyntactic tasks -- dependency parsing, part-of-speech tagging, and named-entity recognition -- and one semantic classification task -- sentiment analysis. We also present a ne…
BERT is a widely used pre-trained model in natural language processing. However, since BERT is quadratic to the text length, the BERT model is difficult to be used directly on the long-text corpus. In some fields, the collected text data may be quite long, such as in the health care field. Therefore, to apply the pre-trained language knowledge of BERT to long text, in this paper, imitating the skimming-intensive reading method used by humans when reading a long paragraph, the Skimming-Intensive Model (SkIn) is proposed. It can dynamically select the critical information in the text so that the…
Deep language models have achieved remarkable success in the NLP domain. The standard way to train a deep language model is to employ unsupervised learning from scratch on a large unlabeled corpus. However, such large corpora are only available for widely-adopted and high-resource languages and domains. This study presents the first deep language model, DPRK-BERT, for the DPRK language. We achieve this by compiling the first unlabeled corpus for the DPRK language and fine-tuning a preexisting the ROK language model. We compare the proposed model with existing approaches and show significant im…
Research on Information Extraction of LCSTS Dataset Based on an Improved BERTSum-LSTM Model
College of Information Science and Technology, Donghua University（东华大学信息科学与技术学院）；College of Physics, Donghua University（东华大学物理学院）；Glorious Sun School of Business & Management, Donghua University（东华大学旭日工商管理学院）；College of Medical Technology, Shanghai University of Medicine & Health Sciences（上海健康医学院医学技术学院）；College of Information Sciences and Technology, Donghua University（东华大学信息科学与技术学院）
Contextualized word embeddings have demonstrated state-of-the-art performance in various natural language processing tasks including those that concern historical semantic change. However, language models such as BERT was trained primarily on contemporary corpus data. To investigate whether training on historical corpus data improves diachronic semantic analysis, we present a pre-trained BERT-based language model, HistBERT, trained on the balanced Corpus of Historical American English. We examine the effectiveness of our approach by comparing the performance of the original BERT and that of Hi…
Integrating LSTM and BERT for Long-Sequence Data Analysis in Intelligent Tutoring Systems
School of Electronics and Computer Science（电子与计算机科学学院）；University of Southampton（南安普顿大学）；Durham University（杜伦大学）；School of Computing（计算机学院）；Newcastle University（纽卡斯尔大学）
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…
We convert the Chinese medical text attributes extraction task into a sequence tagging or machine reading comprehension task. Based on BERT pre-trained models, we have not only tried the widely used LSTM-CRF sequence tagging model, but also other sequence models, such as CNN, UCNN, WaveNet, SelfAttention, etc, which reaches similar performance as LSTM+CRF. This sheds a light on the traditional sequence tagging models. Since the aspect of emphasis for different sequence tagging models varies substantially, ensembling these models adds diversity to the final system. By doing so, our system achie…
National Science Library, Chinese Academy of Science（中国科学院国家科学图书馆）；University of Chinese Academy of Science（中国科学院大学）；City University of New York（纽约城市大学）；Tsinghua University（清华大学）
Computational chemistry develops fast in recent years due to the rapid growth and breakthroughs in AI. Thanks for the progress in natural language processing, researchers can extract more fine-grained knowledge in publications to stimulate the development in computational chemistry. While the works and corpora in chemical entity extraction have been restricted in the biomedicine or life science field instead of the chemistry field, we build a new corpus in chemical bond field annotated for 7 types of entities: compound, solvent, method, bond, reaction, pKa and pKa value. This paper presents a …
University of Helsinki（赫尔辛基大学）；École Normale Supérieure（巴黎高等师范学院）
We present a method for extracting a multilingual sentiment annotated dialog data set from Fallout New Vegas. The game developers have preannotated every line of dialog in the game in one of the 8 different sentiments: \textit{anger, disgust, fear, happy, neutral, pained, sad } and \textit{surprised}. The game has been translated into English, Spanish, German, French and Italian. We conduct experiments on multilingual, multilabel sentiment analysis on the extracted data set using multilingual BERT, XLMRoBERTa and language specific BERT models. In our experiments, multilingual BERT outperformed…
Politehnica Univ. of Bucharest（布加勒斯特理工大学）；University of Turku（图尔库大学）
Large-scale pretrained language models have become ubiquitous in Natural Language Processing. However, most of these models are available either in high-resource languages, in particular English, or as multilingual models that compromise performance on individual languages for coverage. This paper introduces Romanian BERT, the first purely Romanian transformer-based language model, pretrained on a large text corpus. We discuss corpus composition and cleaning, the model training process, as well as an extensive evaluation of the model on various Romanian datasets. We open source not only the mo…
Bidirectional Encoder Representations from Transformers (BERT) has recently achieved state-of-the-art performance on a broad range of NLP tasks including sentence classification, machine translation, and question answering. The BERT model architecture is derived primarily from the transformer. Prior to the transformer era, bidirectional Long Short-Term Memory (BLSTM) has been the dominant modeling architecture for neural machine translation and question answering. In this paper, we investigate how these two modeling techniques can be combined to create a more powerful model architecture. We pr…
Department of Electrical and Electronics Engineering（电气与电子工程系）
Fine-tuning a pretrained BERT model is the state of the art method for extractive/abstractive text summarization, in this paper we showcase how this fine-tuning method can be applied to the Arabic language to both construct the first documented model for abstractive Arabic text summarization and show its performance in Arabic extractive summarization. Our model works with multilingual BERT (as Arabic language does not have a pretrained BERT of its own). We show its performance in English corpus first before applying it to Arabic corpora in both extractive and abstractive tasks.
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…
National Sun Yat-sen University（国立中山大学）；The University of London（伦敦大学）；Academia Sinica（中央研究院）
Varieties of Democracy (V-Dem) is a new approach to conceptualizing and measuring democracy and politics. It has information for 200 countries and is one of the biggest databases for political science. According to the V-Dem annual democracy report 2019, Taiwan is one of the two countries that got disseminated false information from foreign governments the most. It also shows that the "made-up news" has caused a great deal of confusion in Taiwanese society and has serious impacts on global stability. Although there are several applications helping distinguish the false information, we found ou…
This paper summarizes our studies on propaganda detection techniques for news articles in the SemEval-2020 task 11. This task is divided into the SI and TC subtasks. We implemented the GloVe word representation, the BERT pretraining model, and the LSTM model architecture to accomplish this task. Our approach achieved good results for both the SI and TC subtasks. The macro-F1-score for the SI subtask is 0.406, and the micro-F1-score for the TC subtask is 0.505. Our method significantly outperforms the officially released baseline method, and the SI and TC subtasks rank 17th and 22nd, respective…
