We describe an unsupervised method to create pseudo-parallel corpora for machine translation (MT) from unaligned text. We use multilingual BERT to create source and target sentence embeddings for nearest-neighbor search and adapt the model via self-training. We validate our technique by extracting parallel sentence pairs on the BUCC 2017 bitext mining task and observe up to a 24.5 point increase (absolute) in F1 scores over previous unsupervised methods. We then improve an XLM-based unsupervised neural MT system pre-trained on Wikipedia by supplementing it with pseudo-parallel text mined from the same corpus, boosting unsupervised translation performance by up to 3.5 BLEU on the WMT'14 French-English and WMT'16 German-English tasks and outperforming the previous state-of-the-art. Finally, we enrich the IWSLT'15 English-Vietnamese corpus with pseudo-parallel Wikipedia sentence pairs, yielding a 1.2 BLEU improvement on the low-resource MT task. We demonstrate that unsupervised bitext mining is an effective way of augmenting MT datasets and complements existing techniques like initializing with pre-trained contextual embeddings.
University of Maryland（马里兰大学）；Facebook AI Research（脸书人工智能研究院）
Mined bitexts can contain imperfect translations that yield unreliable training signals for Neural Machine Translation (NMT). While filtering such pairs out is known to improve final model quality, we argue that it is suboptimal in low-resource conditions where even mined data can be limited. In our work, we propose instead, to refine the mined bitexts via automatic editing: given a sentence in a language xf, and a possibly imperfect translation of it xe, our model generates a revised version xf' or xe' that yields a more equivalent translation pair (i.e., <xf, xe'> or <xf', xe>). We use a sim…
Obtaining high-quality parallel corpora is of paramount importance for training NMT systems. However, as many language pairs lack adequate gold-standard training data, a popular approach has been to mine so-called "pseudo-parallel" sentences from paired documents in two languages. In this paper, we outline some problems with current methods, propose computationally economical solutions to those problems, and demonstrate success with novel methods on the Tatoeba similarity search benchmark and on a downstream task, namely NMT. We uncover the effect of resource-related factors (i.e. how much mon…
We show that margin-based bitext mining in a multilingual sentence space can be applied to monolingual corpora of billions of sentences. We are using ten snapshots of a curated common crawl corpus (Wenzek et al., 2019) totalling 32.7 billion unique sentences. Using one unified approach for 38 languages, we were able to mine 4.5 billions parallel sentences, out of which 661 million are aligned with English. 20 language pairs have more then 30 million parallel sentences, 112 more then 10 million, and most more than one million, including direct alignments between many European or Asian languages…
University of Chicago（芝加哥大学）；University of Washington（华盛顿大学）
This work presents methods for learning cross-lingual sentence representations using paired or unpaired bilingual texts. We hypothesize that the cross-lingual alignment strategy is transferable, and therefore a model trained to align only two languages can encode multilingually more aligned representations. We thus introduce dual-pivot transfer: training on one language pair and evaluating on other pairs. To study this theory, we design unsupervised models trained on unpaired sentences and single-pair supervised models trained on bitexts, both based on the unsupervised language model XLM-R wit…
Center for Language and Speech Processing（语言与语音处理中心）；Johns Hopkins University（约翰斯·霍普金斯大学）
Mining high-quality bitexts for low-resource languages is challenging. This paper shows that sentence representation of language models fine-tuned with multiple negatives ranking loss, a contrastive objective, helps retrieve clean bitexts. Experiments show that parallel data mined from our approach substantially outperform the previous state-of-the-art method on low resource languages Khmer and Pashto.
We learn a joint multilingual sentence embedding and use the distance between sentences in different languages to filter noisy parallel data and to mine for parallel data in large news collections. We are able to improve a competitive baseline on the WMT'14 English to German task by 0.3 BLEU by filtering out 25% of the training data. The same approach is used to mine additional bitexts for the WMT'14 system and to obtain competitive results on the BUCC shared task to identify parallel sentences in comparable corpora. The approach is generic, it can be applied to many language pairs and it is i…
University of Ljubljana（卢布尔雅那大学）；Jožef Stefan Institute（约瑟夫·斯蒂芬研究所）
Idiomatic expressions can be problematic for natural language processing applications as their meaning cannot be inferred from their constituting words. A lack of successful methodological approaches and sufficiently large datasets prevents the development of machine learning approaches for detecting idioms, especially for expressions that do not occur in the training set. We present an approach, called MICE, that uses contextual embeddings for that purpose. We present a new dataset of multi-word expressions with literal and idiomatic meanings and use it to train a classifier based on two stat…
Scaling multilingual representation learning beyond the hundred most frequent languages is challenging, in particular to cover the long tail of low-resource languages. A promising approach has been to train one-for-all multilingual models capable of cross-lingual transfer, but these models often suffer from insufficient capacity and interference between unrelated languages. Instead, we move away from this approach and focus on training multiple language (family) specific representations, but most prominently enable all languages to still be encoded in the same representational space. To achiev…
In this paper, we propose to align sentence representations from different languages into a unified embedding space, where semantic similarities (both cross-lingual and monolingual) can be computed with a simple dot product. Pre-trained language models are fine-tuned with the translation ranking task. Existing work (Feng et al., 2020) uses sentences within the same batch as negatives, which can suffer from the issue of easy negatives. We adapt MoCo (He et al., 2020) to further improve the quality of alignment. As the experimental results show, the sentence representations produced by our model…
University of Cape Town（开普敦大学）；South African Astronomical Observatory（南非天文台）
Language-agnostic sentence embeddings generated by pre-trained models such as LASER and LaBSE are attractive options for mining large datasets to produce parallel corpora for low-resource machine translation. We test LASER and LaBSE in extracting bitext for two related low-resource African languages: Luhya and Swahili. For this work, we created a new parallel set of nearly 8000 Luhya-English sentences which allows a new zero-shot test of LASER and LaBSE. We find that LaBSE significantly outperforms LASER on both languages. Both LASER and LaBSE however perform poorly at zero-shot alignment on L…
Creativity, i.e., the process of generating and developing fresh and original ideas or products that are useful or effective, is a valuable skill in a variety of domains. Creativity is called an essential 21st-century skill that should be taught in schools. The use of educational technology to promote creativity is an active study field, as evidenced by several studies linking creativity in the classroom to beneficial learning outcomes. Despite the burgeoning body of research on adaptive technology for education, mining creative thinking patterns from educational data remains a challenging tas…
Attention Mechanism and Context Modeling System for Text Mining Machine Translation
Boston University（波士顿大学）；Duke University（杜克大学）；Carnegie Mellon University（卡内基梅隆大学）；University of Pennsylvania（宾夕法尼亚大学）；North Carolina State University（北卡罗来纳州立大学）
We present an approach based on multilingual sentence embeddings to automatically extract parallel sentences from the content of Wikipedia articles in 85 languages, including several dialects or low-resource languages. We do not limit the the extraction process to alignments with English, but systematically consider all possible language pairs. In total, we are able to extract 135M parallel sentences for 1620 different language pairs, out of which only 34M are aligned with English. This corpus of parallel sentences is freely available at https://github.com/facebookresearch/LASER/tree/master/ta…
TTI-Chicago（丰田工业大学芝加哥分校）；University of Washington（华盛顿大学）；Facebook AI Research（脸书人工智能研究院）
Bilingual lexicons map words in one language to their translations in another, and are typically induced by learning linear projections to align monolingual word embedding spaces. In this paper, we show it is possible to produce much higher quality lexicons with methods that combine (1) unsupervised bitext mining and (2) unsupervised word alignment. Directly applying a pipeline that uses recent algorithms for both subproblems significantly improves induced lexicon quality and further gains are possible by learning to filter the resulting lexical entries, with both unsupervised and semi-supervi…
Modular Sentence Encoders: Separating Language Specialization from Cross-Lingual Alignment
Ubiquitous Knowledge Processing Lab (UKP Lab) Department of Computer Science Technical University of Darmstadt and National Research Center for Applied Cybersecurity ATHENE, Germany（技术大学达姆施塔特计算机科学系通用知识处理实验室（UKP实验室）和应用网络安全国家研究中心ATHENE德国）；Center for AI and Data Science, University of Würzburg（人工智能与数据科学中心，乌尔姆大学）
Alberta Machine Intelligence Institute（阿尔伯塔机器智能研究所）；University of Alberta（阿尔伯塔大学）
Acquisition of multilingual training data continues to be a challenge in word sense disambiguation (WSD). To address this problem, unsupervised approaches have been proposed to automatically generate sense annotations for training supervised WSD systems. We present three new methods for creating sense-annotated corpora which leverage translations, parallel bitexts, lexical resources, as well as contextual and synset embeddings. Our semi-supervised method applies machine translation to transfer existing sense annotations to other languages. Our two unsupervised methods refine sense annotations …
University College Dublin（都柏林大学学院）；Birkbeck, University of London（伦敦大学伯贝克学院）
The contextual word embedding model, BERT, has proved its ability on downstream tasks with limited quantities of annotated data. BERT and its variants help to reduce the burden of complex annotation work in many interdisciplinary research areas, for example, legal argument mining in digital humanities. Argument mining aims to develop text analysis tools that can automatically retrieve arguments and identify relationships between argumentation clauses. Since argumentation is one of the key aspects of case law, argument mining tools for legal texts are applicable to both academic and non-academi…
Diachronic text mining has frequently been applied to long-term linguistic surveys of word meaning and usage shifts over time. In this paper we apply short-term diachronic text mining to a rapidly growing corpus of scientific publications on COVID-19 captured in the CORD-19 dataset in order to identify co-occurrences and analyze the behavior of potential candidate treatments. We used a data set associated with a COVID-19 drug re-purposing study from Oak Ridge National Laboratory. This study identified existing candidate coronavirus treatments, including drugs and approved compounds, which had …
University of Bristol（布里斯托大学）；University of Sussex（萨塞克斯大学）
This research on data extraction methods applies recent advances in natural language processing to evidence synthesis based on medical texts. Texts of interest include abstracts of clinical trials in English and in multilingual contexts. The main focus is on information characterized via the Population, Intervention, Comparator, and Outcome (PICO) framework, but data extraction is not limited to these fields. Recent neural network architectures based on transformers show capacities for transfer learning and increased performance on downstream natural language processing tasks such as universal…
This paper describes our submission for the SemEval-2019 Suggestion Mining task. A simple Convolutional Neural Network (CNN) classifier with contextual word representations from a pre-trained language model was used for sentence classification. The model is trained using tri-training, a semi-supervised bootstrapping mechanism for labelling unseen data. Tri-training proved to be an effective technique to accommodate domain shift for cross-domain suggestion mining (Subtask B) where there is no hand labelled training data. For in-domain evaluation (Subtask A), we use the same technique to augment…
INESC-ID Lisboa（INESC-ID里斯本）；Instituto Superior Técnico（高等技术学院）；Universidade de Lisboa（里斯本大学）；Google DeepMind（谷歌DeepMind）
Recent approaches for sentiment lexicon induction have capitalized on pre-trained word embeddings that capture latent semantic properties. However, embeddings obtained by optimizing performance of a given task (e.g. predicting contextual words) are sub-optimal for other applications. In this paper, we address this problem by exploiting task-specific representations, induced via embedding sub-space projection. This allows us to expand lexicons describing multiple semantic properties. For each property, our model jointly learns suitable representations and the concomitant predictor. Experiments …
R2GenCSR: Mining Contextual and Residual Information for LLMs-based Radiology Report Generation
School of Computer Science and Technology, Anhui University（安徽大学计算机科学与技术学院）；First Affiliated Hospital of Anhui University of Chinese Medicine（安徽中医药大学第一附属医院）
University of Applied Sciences and Arts of Western Switzerland（瑞士西部应用科学与艺术大学）；Swiss Institute of Bioinformatics（瑞士生物信息学研究所）；University of Geneva（日内瓦大学）
Chemical patent documents describe a broad range of applications holding key reaction and compound information, such as chemical structure, reaction formulas, and molecular properties. These informational entities should be first identified in text passages to be utilized in downstream tasks. Text mining provides means to extract relevant information from chemical patents through information extraction techniques. As part of the Information Extraction task of the Cheminformatics Elsevier Melbourne University challenge, in this work we study the effectiveness of contextualized language models t…
Sarcasm is the use of words usually used to either mock or annoy someone, or for humorous purposes. Sarcasm is largely used in social networks and microblogging websites, where people mock or censure in a way that makes it difficult even for humans to tell if what is said is what is meant. Failure to identify sarcastic utterances in Natural Language Processing applications such as sentiment analysis and opinion mining will confuse classification algorithms and generate false results. Several studies on sarcasm detection have utilized different learning algorithms. However, most of these learni…
On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Survey
Institute for Clarity in Documentation（清晰文档研究所）；Inria Paris-Rocquencourt（巴黎- Rocquencourt 国家信息与自动化研究所）；Rajiv Gandhi University（拉贾·甘地大学）；Tsinghua University（清华大学）；Palmer Research Laboratories（帕勒尔研究实验室）；Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳））
Microblogging sites, like Twitter, have emerged as ubiquitous sources of information. Two important tasks related to the automatic extraction and analysis of information in Microblogs are Entity Mention Detection (EMD) and Entity Detection (ED). The state-of-the-art EMD systems aim to model the non-literary nature of microblog text by training upon offline static datasets. They extract a combination of surface-level features -- orthographic, lexical, and semantic -- from individual messages for noisy text modeling and entity extraction. But given the constantly evolving nature of microblog str…
Language Model as an Annotator: Unsupervised Context-aware Quality Phrase Generation
College of Economics and Management, Beijing University of Technology（北京工业大学经济与管理学院）；School of Economics and Management, Beihang University（北京航空航天大学经济与管理学院）；Department of Computer Science, University of Manchester（曼彻斯特大学计算机科学系）
Information Sciences Institute（信息科学研究所）；USC Viterbi School of Engineering（南加州大学维特比工程学院）
Word embeddings have made enormous inroads in recent years in a wide variety of text mining applications. In this paper, we explore a word embedding-based architecture for predicting the relevance of a role between two financial entities within the context of natural language sentences. In this extended abstract, we propose a pooled approach that uses a collection of sentences to train word embeddings using the skip-gram word2vec architecture. We use the word embeddings to obtain context vectors that are assigned one or more labels based on manual annotations. We train a machine learning class…
Institute of Mathematics and Computer Science, University of São Paulo（圣保罗大学数学与计算机科学研究所）；Universidade Católica de Brasília（巴西利亚天主教大学）
Detecting keywords in texts is important for many text mining applications. Graph-based methods have been commonly used to automatically find the key concepts in texts, however, relevant information provided by embeddings has not been widely used to enrich the graph structure. Here we modeled texts co-occurrence networks, where nodes are words and edges are established either by contextual or semantical similarity. We compared two embedding approaches -- Word2vec and BERT -- to check whether edges created via word embeddings can improve the quality of the keyword extraction method. We found th…
