Recently, the emergence of pre-trained models (PTMs) has brought natural language processing (NLP) to a new era. In this survey, we provide a comprehensive review of PTMs for NLP. We first briefly introduce language representation learning and its research progress. Then we systematically categorize existing PTMs based on a taxonomy with four perspectives. Next, we describe how to adapt the knowledge of PTMs to the downstream tasks. Finally, we outline some potential directions of PTMs for future research. This survey is purposed to be a hands-on guide for understanding, using, and developing PTMs for various NLP tasks.
Speech representations learned from Self-supervised learning (SSL) models can benefit various speech processing tasks. However, utilizing SSL representations usually requires fine-tuning the pre-trained models or designing task-specific downstream models and loss functions, causing much memory usage and human labor. Recently, prompting in Natural Language Processing (NLP) has been found to be an efficient technique to leverage pre-trained language models (LMs). Specifically, prompt tuning optimizes a limited number of task-specific parameters with a fixed pre-trained model; as a result, only a…
The field of clinical natural language processing has been advanced significantly since the introduction of deep learning models. The self-supervised representation learning and the transfer learning paradigm became the methods of choice in many natural language processing application, in particular in the settings with the dearth of high quality manually annotated data. Electronic health record systems are ubiquitous and the majority of patients' data are now being collected electronically and in particular in the form of free text. Identification of medical concepts and information extractio…
Recent progress in the Natural Language Processing domain has given us several State-of-the-Art (SOTA) pretrained models which can be finetuned for specific tasks. These large models with billions of parameters trained on numerous GPUs/TPUs over weeks are leading in the benchmark leaderboards. In this paper, we discuss the need for a benchmark for cost and time effective smaller models trained on a single GPU. This will enable researchers with resource constraints experiment with novel and innovative ideas on tokenization, pretraining tasks, architecture, fine tuning methods etc. We set up Sma…
Raytheon BBN Technologies（雷神BBN技术公司）；Harvard University（哈佛大学）；University of Pennsylvania（宾夕法尼亚大学）；University of Oregon（俄勒冈大学）；University of the Basque Country (UPV/EHU)（巴斯克大学 (UPV/EHU)）
Large, pre-trained transformer-based language models such as BERT have drastically changed the Natural Language Processing (NLP) field. We present a survey of recent work that uses these large language models to solve NLP tasks via pre-training then fine-tuning, prompting, or text generation approaches. We also present approaches that use pre-trained language models to generate data for training augmentation or other purposes. We conclude with discussions on limitations and suggested directions for future research.
Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
With the maturity and popularity of dialogue systems, detecting user's unknown intent in dialogue systems has become an important task. It is also one of the most challenging tasks since we can hardly get examples, prior knowledge or the exact numbers of unknown intents. In this paper, we propose SofterMax and deep novelty detection (SMDN), a simple yet effective post-processing method for detecting unknown intent in dialogue systems based on pre-trained deep neural network classifiers. Our method can be flexibly applied on top of any classifiers trained in deep neural networks without changin…
We observe a recent behaviour on social media, in which users intentionally remove consonantal dots from Arabic letters, in order to bypass content-classification algorithms. Content classification is typically done by fine-tuning pre-trained language models, which have been recently employed by many natural-language-processing applications. In this work we study the effect of applying pre-trained Arabic language models on "undotted" Arabic texts. We suggest several ways of supporting undotted texts with pre-trained models, without additional training, and measure their performance on two Arab…
Recent advancements in natural language processing \cite{gpt2} \cite{BERT} have led to near-human performance in multiple natural language tasks. In this paper, we seek to understand whether similar techniques can be applied to a highly structured environment with strict syntax rules. Specifically, we propose an end-to-end machine learning model for code generation in the Python language built on-top of pre-trained language models. We demonstrate that a fine-tuned model can perform well in code generation tasks, achieving a BLEU score of 0.22, an improvement of 46\% over a reasonable sequence-…
University of Copenhagen（哥本哈根大学）；the Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Baidu（百度）；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）
The emerging classical-quantum transfer learning paradigm has brought a decent performance to quantum computational models in many tasks, such as computer vision, by enabling a combination of quantum models and classical pre-trained neural networks. However, using quantum computing with pre-trained models has yet to be explored in natural language processing (NLP). Due to the high linearity constraints of the underlying quantum computing infrastructures, existing Quantum NLP models are limited in performance on real tasks. We fill this gap by pre-training a sentence state with complex-valued B…
Pre-trained speech Transformers have facilitated great success across various speech processing tasks. However, fine-tuning these encoders for downstream tasks require sufficiently large training data to converge or to achieve state-of-the-art. In text domain this has been partly attributed to sub-optimality of the representation space in pre-trained Transformers. In this work, we take a sober look into pre-trained speech encoders and rewire their representation space without requiring any task-specific labels. Our method utilises neutrally synthesised version of audio inputs along with frame …
Pre-trained word embeddings are the primary method for transfer learning in several Natural Language Processing (NLP) tasks. Recent works have focused on using unsupervised techniques such as language modeling to obtain these embeddings. In contrast, this work focuses on extracting representations from multiple pre-trained supervised models, which enriches word embeddings with task and domain specific knowledge. Experiments performed in cross-task, cross-domain and cross-lingual settings indicate that such supervised embeddings are helpful, especially in the low-resource setting, but the exten…
Harbin Institute of Technology（哈尔滨工业大学）；iFLYTEK Research（科大讯飞研究院）；iFLYTEK AI Research (Hebei)（科大讯飞人工智能研究院（河北））
Bidirectional Encoder Representations from Transformers (BERT) has shown marvelous improvements across various NLP tasks, and consecutive variants have been proposed to further improve the performance of the pre-trained language models. In this paper, we target on revisiting Chinese pre-trained language models to examine their effectiveness in a non-English language and release the Chinese pre-trained language model series to the community. We also propose a simple but effective model called MacBERT, which improves upon RoBERTa in several ways, especially the masking strategy that adopts MLM a…
University of Pennsylvania（宾夕法尼亚大学）；Adobe Research（Adobe研究院）
Keeping the performance of language technologies optimal as time passes is of great practical interest. We study temporal effects on model performance on downstream language tasks, establishing a nuanced terminology for such discussion and identifying factors essential to conduct a robust study. We present experiments for several tasks in English where the label correctness is not dependent on time and demonstrate the importance of distinguishing between temporal model deterioration and temporal domain adaptation for systems using pre-trained representations. We find that depending on the task…
University of Chicago（芝加哥大学）；Department of Linguistics（语言学系）
Pre-trained LMs have shown impressive performance on downstream NLP tasks, but we have yet to establish a clear understanding of their sophistication when it comes to processing, retaining, and applying information presented in their input. In this paper we tackle a component of this question by examining robustness of models' ability to deploy relevant context information in the face of distracting content. We present models with cloze tasks requiring use of critical context information, and introduce distracting content to test how robustly the models retain and use that critical information…
BNLP is an open source language processing toolkit for Bengali language consisting with tokenization, word embedding, POS tagging, NER tagging facilities. BNLP provides pre-trained model with high accuracy to do model based tokenization, embedding, POS tagging, NER tagging task for Bengali language. BNLP pre-trained model achieves significant results in Bengali text tokenization, word embedding, POS tagging and NER tagging task. BNLP is using widely in the Bengali research communities with 16K downloads, 119 stars and 31 forks. BNLP is available at https://github.com/sagorbrur/bnlp.
Transfer learning is a vital technique that generalizes models trained for one setting or task to other settings or tasks. For example in speech recognition, an acoustic model trained for one language can be used to recognize speech in another language, with little or no re-training data. Transfer learning is closely related to multi-task learning (cross-lingual vs. multilingual), and is traditionally studied in the name of `model adaptation'. Recent advance in deep learning shows that transfer learning becomes much easier and more effective with high-level abstract features learned by deep mo…
SikuGPT: A Generative Pre-trained Model for Intelligent Information Processing of Ancient Texts from the Perspective of Digital Humanities
College of Information Management, Nanjing Agricultural University（南京农业大学信息管理学院）；School of Economics & Management, Nanjing University of Science and Technology（南京理工大学经济管理学院）；College of Liberal Art, Nanjing Normal University（南京师范大学文学院）；School of Foreign Languages, China Pharmaceutical University（中国药科大学外国语学院）
A Survey of Pre-trained Language Models for Processing Scientific Text
National Institute of Informatics（情报学研究所）；The University of Tokyo（东京大学）；NTT Communications Corporation（NTT通信公司）；University of Information Technology（信息技术大学）；CNRS（法国国家科学研究中心）；LS2N, Université de Nantes（南特大学LS2N实验室）
Athens University of Economics and Business（雅典经济与商业大学）；University of Copenhagen（哥本哈根大学）；Cognitiv+
Pre-trained Transformers currently dominate most NLP tasks. They impose, however, limits on the maximum input length (512 sub-words in BERT), which are too restrictive in the legal domain. Even sparse-attention models, such as Longformer and BigBird, which increase the maximum input length to 4,096 sub-words, severely truncate texts in three of the six datasets of LexGLUE. Simpler linear classifiers with TF-IDF features can handle texts of any length, require far less resources to train and deploy, but are usually outperformed by pre-trained Transformers. We explore two directions to cope with…
This technical note describes a set of baseline tools for automatic processing of Danish text. The tools are machine-learning based, using natural language processing models trained over previously annotated documents. They are maintained at ITU Copenhagen and will always be freely available.
To investigate the role of linguistic knowledge in data augmentation (DA) for Natural Language Processing (NLP), we designed two adapted DA programs and applied them to LCQMC (a Large-scale Chinese Question Matching Corpus) for a binary Chinese question matching classification task. The two DA programs produce augmented texts by five simple text editing operations (or DA techniques), largely irrespective of language generation rules, but one is enhanced with a pre-trained n-gram language model to fuse it with prior linguistic knowledge. We then trained four neural network models (BOW, CNN, LST…
We examine whether neural natural language processing (NLP) systems reflect historical biases in training data. We define a general benchmark to quantify gender bias in a variety of neural NLP tasks. Our empirical evaluation with state-of-the-art neural coreference resolution and textbook RNN-based language models trained on benchmark datasets finds significant gender bias in how models view occupations. We then mitigate bias with CDA: a generic methodology for corpus augmentation via causal interventions that breaks associations between gendered and gender-neutral words. We empirically show t…
This open access book provides a comprehensive overview of the state of the art in research and applications of Foundation Models and is intended for readers familiar with basic Natural Language Processing (NLP) concepts. Over the recent years, a revolutionary new paradigm has been developed for training models for NLP. These models are first pre-trained on large collections of text documents to acquire general syntactic knowledge and semantic information. Then, they are fine-tuned for specific tasks, which they can often solve with superhuman accuracy. When the models are large enough, they c…
Most tasks in natural language processing can be cast into question answering (QA) problems over language input. We introduce the dynamic memory network (DMN), a neural network architecture which processes input sequences and questions, forms episodic memories, and generates relevant answers. Questions trigger an iterative attention process which allows the model to condition its attention on the inputs and the result of previous iterations. These results are then reasoned over in a hierarchical recurrent sequence model to generate answers. The DMN can be trained end-to-end and obtains state-o…
We present SentiMATE, a novel end-to-end Deep Learning model for Chess, employing Natural Language Processing that aims to learn an effective evaluation function assessing move quality. This function is pre-trained on the sentiment of commentary associated with the training moves and is used to guide and optimize the agent's game-playing decision making. The contributions of this research are three-fold: we build and put forward both a classifier which extracts commentary describing the quality of Chess moves in vast commentary datasets, and a Sentiment Analysis model trained on Chess commenta…
This paper seeks to address the classification of misinformation in news articles using a Long Short Term Memory Recurrent Neural Network. Articles were taken from 2018; a year that was filled with reporters writing about President Donald Trump, Special Counsel Robert Mueller, the Fifa World Cup, and Russia. The model presented successfully classifies these articles with an accuracy score of 0.779944. We consider this to be successful because the model was trained on articles that included languages other than English as well as incomplete, or fragmented, articles.
Stellenbosch University（斯泰伦博斯大学）；University of Edinburgh（爱丁堡大学）
Acoustic word embeddings are fixed-dimensional representations of variable-length speech segments. In settings where unlabelled speech is the only available resource, such embeddings can be used in "zero-resource" speech search, indexing and discovery systems. Here we propose to train a single supervised embedding model on labelled data from multiple well-resourced languages and then apply it to unseen zero-resource languages. For this transfer learning approach, we consider two multilingual recurrent neural network models: a discriminative classifier trained on the joint vocabularies of all t…
We introduce SciWING, an open-source software toolkit which provides access to pre-trained models for scientific document processing tasks, inclusive of citation string parsing and logical structure recovery. SciWING enables researchers to rapidly experiment with different models by swapping and stacking different modules. It also enables them declare and run models from a configuration file. It enables researchers to perform production-ready transfer learning from general, pre-trained transformers (i.e., BERT, SciBERT etc), and aids development of end-user applications. It includes ready-to-u…
Humans often communicate by using imprecise language, suggesting that fuzzy concepts with unclear boundaries are prevalent in language use. In this paper, we test the extent to which models trained to capture the distributional statistics of language show correspondence to fuzzy-membership patterns. Using the task of natural language inference, we test a recent state of the art model on the classical case of temperature, by examining its mapping of temperature data to fuzzy-perceptions such as "cool", "hot", etc. We find the model to show patterns that are similar to classical fuzzy-set theore…
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…
Natural Language Processing for Financial Regulation
Imperial College London（帝国理工学院）；CNRS, Complex Systems Institute of Paris Île-de-France（法国国家科学研究中心巴黎法兰西岛复杂系统研究所）；Oxford Man Institute of Quantitative Finance（牛津曼定量金融研究所）
National Center for Biotechnology Information（国家生物技术信息中心）；National Library of Medicine, National Institutes of Health（美国国立卫生研究院国家医学图书馆）
Inspired by the success of the General Language Understanding Evaluation benchmark, we introduce the Biomedical Language Understanding Evaluation (BLUE) benchmark to facilitate research in the development of pre-training language representations in the biomedicine domain. The benchmark consists of five tasks with ten datasets that cover both biomedical and clinical texts with different dataset sizes and difficulties. We also evaluate several baselines based on BERT and ELMo and find that the BERT model pre-trained on PubMed abstracts and MIMIC-III clinical notes achieves the best results. We m…
Convolutional Neural Networks for Sentiment Analysis on Weibo Data: A Natural Language Processing Approach
College Of Computing And Information Technologies（计算与信息技术学院）；National University Manila（马尼拉国立大学）；School of Information Engineering（信息工程学院）；Jiangxi College of Applied Technology（江西应用技术职业学院）
National Taiwan University（台湾大学）；Nagoya University（名古屋大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；Meta AI；Massachusetts Institute of Technology（麻省理工学院）；Carnegie Mellon University（卡内基梅隆大学）；LXT
Transfer learning has proven to be crucial in advancing the state of speech and natural language processing research in recent years. In speech, a model pre-trained by self-supervised learning transfers remarkably well on multiple tasks. However, the lack of a consistent evaluation methodology is limiting towards a holistic understanding of the efficacy of such models. SUPERB was a step towards introducing a common benchmark to evaluate pre-trained models across various speech tasks. In this paper, we introduce SUPERB-SG, a new benchmark focused on evaluating the semantic and generative capabi…
Spelling irregularities, known now as spelling mistakes, have been found for several centuries. As humans, we are able to understand most of the misspelled words based on their location in the sentence, perceived pronunciation, and context. Unlike humans, computer systems do not possess the convenient auto complete functionality of which human brains are capable. While many programs provide spelling correction functionality, many systems do not take context into account. Moreover, Artificial Intelligence systems function in the way they are trained on. With many current Natural Language Proces…
Federated Learning aims to learn machine learning models from multiple decentralized edge devices (e.g. mobiles) or servers without sacrificing local data privacy. Recent Natural Language Processing techniques rely on deep learning and large pre-trained language models. However, both big deep neural and language models are trained with huge amounts of data which often lies on the server side. Since text data is widely originated from end users, in this work, we look into recent NLP models and techniques which use federated learning as the learning framework. Our survey discusses major challeng…
Seoul National University（首尔国立大学）；University of Chicago（芝加哥大学）
With the recent success and popularity of pre-trained language models (LMs) in natural language processing, there has been a rise in efforts to understand their inner workings. In line with such interest, we propose a novel method that assists us in investigating the extent to which pre-trained LMs capture the syntactic notion of constituency. Our method provides an effective way of extracting constituency trees from the pre-trained LMs without training. In addition, we report intriguing findings in the induced trees, including the fact that pre-trained LMs outperform other approaches in corre…
University of Copenhagen（哥本哈根大学）；National University of Defense Technology（国防科技大学）
We present a benchmark suite of four datasets for evaluating the fairness of pre-trained language models and the techniques used to fine-tune them for downstream tasks. Our benchmarks cover four jurisdictions (European Council, USA, Switzerland, and China), five languages (English, German, French, Italian and Chinese) and fairness across five attributes (gender, age, region, language, and legal area). In our experiments, we evaluate pre-trained language models using several group-robust fine-tuning techniques and show that performance group disparities are vibrant in many cases, while none of …
Zhejiang University（浙江大学）；Alibaba-Zhejiang University Joint Research Institute of Frontier Technologies（阿里巴巴-浙江大学前沿技术联合研究院）；Alibaba Group（阿里巴巴集团）
Large-scale pre-trained language models have contributed significantly to natural language processing by demonstrating remarkable abilities as few-shot learners. However, their effectiveness depends mainly on scaling the model parameters and prompt design, hindering their implementation in most real-world applications. This study proposes a novel pluggable, extensible, and efficient approach named DifferentiAble pRompT (DART), which can convert small language models into better few-shot learners without any prompt engineering. The main principle behind this approach involves reformulating pote…
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）；National Research University Higher School of Economics（国立研究型大学高等经济学院）
Neural Architecture Search (NAS) is a promising and rapidly evolving research area. Training a large number of neural networks requires an exceptional amount of computational power, which makes NAS unreachable for those researchers who have limited or no access to high-performance clusters and supercomputers. A few benchmarks with precomputed neural architectures performances have been recently introduced to overcome this problem and ensure more reproducible experiments. However, these benchmarks are only for the computer vision domain and, thus, are built from the image datasets and convoluti…
While commonsense knowledge acquisition and reasoning has traditionally been a core research topic in the knowledge representation and reasoning community, recent years have seen a surge of interest in the natural language processing community in developing pre-trained models and testing their ability to address a variety of newly designed commonsense knowledge reasoning and generation tasks. This paper presents a survey of these tasks, discusses the strengths and weaknesses of state-of-the-art pre-trained models for commonsense reasoning and generation as revealed by these tasks, and reflects…
Stanford University（斯坦福大学）；Facebook AI Research（脸书人工智能研究院）；UPF Linguistics Department（庞培法布拉大学语言学系）
Recurrent neural networks (RNNs) have reached striking performance in many natural language processing tasks. This has renewed interest in whether these generic sequence processing devices are inducing genuine linguistic knowledge. Nearly all current analytical studies, however, initialize the RNNs with a vocabulary of known words, and feed them tokenized input during training. We present a multi-lingual study of the linguistic knowledge encoded in RNNs trained as character-level language models, on input data with word boundaries removed. These networks face a tougher and more cognitively rea…
With increasing globalization and immigration, various studies have estimated that about half of the world population is bilingual. Consequently, individuals concurrently use two or more languages or dialects in casual conversational settings. However, most research is natural language processing is focused on monolingual text. To further the work in code-switched sentiment analysis, we propose a multi-step natural language processing algorithm utilizing points of code-switching in mixed text and conduct sentiment analysis around those identified points. The proposed sentiment analysis algorit…
Faculty of Informatics Kaunas University of Technology（考纳斯理工大学信息学院）；Kaunas University of Technology（考纳斯理工大学）
A recent introduction of Transformer deep learning architecture made breakthroughs in various natural language processing tasks. However, non-English languages could not leverage such new opportunities with the English text pre-trained models. This changed with research focusing on multilingual models, where less-spoken languages are the main beneficiaries. We compare pre-trained multilingual BERT, XLM-R, and older learned text representation methods as encodings for the task of Lithuanian news clustering. Our results indicate that publicly available pre-trained multilingual Transformer models…
In this work, we release COVID-Twitter-BERT (CT-BERT), a transformer-based model, pretrained on a large corpus of Twitter messages on the topic of COVID-19. Our model shows a 10-30% marginal improvement compared to its base model, BERT-Large, on five different classification datasets. The largest improvements are on the target domain. Pretrained transformer models, such as CT-BERT, are trained on a specific target domain and can be used for a wide variety of natural language processing tasks, including classification, question-answering and chatbots. CT-BERT is optimised to be used on COVID-19…
German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）
Current state-of-the-art relation extraction methods typically rely on a set of lexical, syntactic, and semantic features, explicitly computed in a pre-processing step. Training feature extraction models requires additional annotated language resources, which severely restricts the applicability and portability of relation extraction to novel languages. Similarly, pre-processing introduces an additional source of error. To address these limitations, we introduce TRE, a Transformer for Relation Extraction, extending the OpenAI Generative Pre-trained Transformer [Radford et al., 2018]. Unlike pr…
