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 as correction (Mac). We carried out extensive experiments on eight Chinese NLP tasks to revisit the existing pre-trained language models as well as the proposed MacBERT. Experimental results show that MacBERT could achieve state-of-the-art performances on many NLP tasks, and we also ablate details with several findings that may help future research. Resources available: https://github.com/ymcui/MacBERT
East China Normal University（华东师范大学）；Alibaba Group（阿里巴巴集团）；Chongqing University（重庆大学）
Recently, knowledge-enhanced pre-trained language models (KEPLMs) improve context-aware representations via learning from structured relations in knowledge graphs, and/or linguistic knowledge from syntactic or dependency analysis. Unlike English, there is a lack of high-performing open-source Chinese KEPLMs in the natural language processing (NLP) community to support various language understanding applications. In this paper, we revisit and advance the development of Chinese natural language understanding with a series of novel Chinese KEPLMs released in various parameter sizes, namely CKBERT…
State Key Lab of Software Development Environment, Beihang University（北京航空航天大学软件开发环境国家重点实验室）；School of Computer Science and Engineering, Northeastern University（东北大学计算机科学与工程学院）；Faculty of Computing, Harbin Institute of Technology（哈尔滨工业大学计算学部）；Lark Platform Engineering-AI（飞书平台工程-人工智能）
Pretrained language models (PLMs) have shown marvelous improvements across various NLP tasks. Most Chinese PLMs simply treat an input text as a sequence of characters, and completely ignore word information. Although Whole Word Masking can alleviate this, the semantics in words is still not well represented. In this paper, we revisit the segmentation granularity of Chinese PLMs. We propose a mixed-granularity Chinese BERT (MigBERT) by considering both characters and words. To achieve this, we design objective functions for learning both character and word-level representations. We conduct exte…
Shanghai Jiao Tong University（上海交通大学）；City University of Hong Kong（香港城市大学）；University of Pennsylvania（宾夕法尼亚大学）；Brigham Young University（杨百翰大学）；University of Washington（华盛顿大学）；Indiana University Bloomington（印第安纳大学伯明顿分校）
Ancient Chinese is the essence of Chinese culture. There are several natural language processing tasks of ancient Chinese domain, such as ancient-modern Chinese translation, poem generation, and couplet generation. Previous studies usually use the supervised models which deeply rely on parallel data. However, it is difficult to obtain large-scale parallel data of ancient Chinese. In order to make full use of the more easily available monolingual ancient Chinese corpora, we release AnchiBERT, a pre-trained language model based on the architecture of BERT, which is trained on large-scale ancient…
With the success of down streaming task using English pre-trained language model, the pre-trained Chinese language model is also necessary to get a better performance of Chinese NLP task. Unlike the English language, Chinese has its special characters such as glyph information. So in this article, we propose the Chinese pre-trained language model StyleBERT which incorporate the following embedding information to enhance the savvy of language model, such as word, pinyin, five stroke and chaizi. The experiments show that the model achieves well performances on a wide range of Chinese NLP tasks.
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…
iFLYTEK Research（科大讯飞研究院）；Harbin Institute of Technology（哈尔滨工业大学）；Jilin Kexun Information Technology Co., Ltd.（吉林科讯信息技术有限公司）
Multilingual pre-trained language models have shown impressive performance on cross-lingual tasks. It greatly facilitates the applications of natural language processing on low-resource languages. However, there are still some languages that the current multilingual models do not perform well on. In this paper, we propose CINO (Chinese Minority Pre-trained Language Model), a multilingual pre-trained language model for Chinese minority languages. It covers Standard Chinese, Yue Chinese, and six other ethnic minority languages. To evaluate the cross-lingual ability of the multilingual model on e…
Adversarial attacks in texts are mostly substitution-based methods that replace words or characters in the original texts to achieve success attacks. Recent methods use pre-trained language models as the substitutes generator. While in Chinese, such methods are not applicable since words in Chinese require segmentations first. In this paper, we propose a pre-train language model as the substitutes generator using sentence-pieces to craft adversarial examples in Chinese. The substitutions in the generated adversarial examples are not characters or words but \textit{'pieces'}, which are more nat…
In recent years, after the neural-network-based method was proposed, the accuracy of the Chinese word segmentation task has made great progress. However, when dealing with out-of-vocabulary words, there is still a large error rate. We used a simple bidirectional LSTM architecture and a large-scale pretrained language model to generate high-quality contextualize character representations, which successfully reduced the weakness of the ambiguous meanings of each Chinese character that widely appears in Chinese characters, and hence effectively reduced OOV error rate. State-of-the-art performance…
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. In this technical report, we present our practice of pre-training language models named NEZHA (NEural contextualiZed representation for CHinese lAnguage understanding) on Chinese corpora and finetuning for the Chinese NLU tasks. The current version of NEZHA is based on BERT with a collection of proven improvements, which include Functional Relative Positional Encodin…
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）
Most Chinese pre-trained models take character as the basic unit and learn representation according to character's external contexts, ignoring the semantics expressed in the word, which is the smallest meaningful utterance in Chinese. Hence, we propose a novel word-aligned attention to exploit explicit word information, which is complementary to various character-based Chinese pre-trained language models. Specifically, we devise a pooling mechanism to align the character-level attention to the word level and propose to alleviate the potential issue of segmentation error propagation by multi-so…
Pre-trained Language Models (PLMs) have proven to be beneficial for various downstream NLP tasks. Recently, GPT-3, with 175 billion parameters and 570GB training data, drew a lot of attention due to the capacity of few-shot (even zero-shot) learning. However, applying GPT-3 to address Chinese NLP tasks is still challenging, as the training corpus of GPT-3 is primarily English, and the parameters are not publicly available. In this technical report, we release the Chinese Pre-trained Language Model (CPM) with generative pre-training on large-scale Chinese training data. To the best of our knowl…
Recent NLP tasks have benefited a lot from pre-trained language models (LM) since they are able to encode knowledge of various aspects. However, current LM evaluations focus on downstream performance, hence lack to comprehensively inspect in which aspect and to what extent have they encoded knowledge. This paper addresses both queries by proposing four tasks on syntactic, semantic, commonsense, and factual knowledge, aggregating to a total of $39,308$ questions covering both linguistic and world knowledge in Chinese. Throughout experiments, our probes and knowledge data prove to be a reliable …
Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model
Multimodal Art Projection Research Community（多模态艺术投影研究社区）；HKUST（香港科技大学）；Fudan University（复旦大学）；Peking University（北京大学）；Shanghai Jiaotong University（上海交通大学）；Nanjing University（南京大学）；University of Waterloo（滑铁卢大学）；Vector Institute（矢量研究所）
Bangladesh University of Engineering and Technology（孟加拉国工程技术大学）；United International University（联合国际大学）
Though there has been a large body of recent works in language modeling (LM) for high resource languages such as English and Chinese, the area is still unexplored for low resource languages like Bengali and Hindi. We propose an end to end trainable memory efficient CNN architecture named CoCNN to handle specific characteristics such as high inflection, morphological richness, flexible word order and phonetical spelling errors of Bengali and Hindi. In particular, we introduce two learnable convolutional sub-models at word and at sentence level that are end to end trainable. We show that state-o…
We present a simple yet effective method for generating high quality classical Chinese poetry with Generative Pre-trained Language Model (GPT). The method adopts a simple GPT model, without using any human crafted rules or features, or designing any additional neural components. While the proposed model learns to generate various forms of classical Chinese poems, including Jueju, Lüshi, various Cipai and Couples, the generated poems are of very high quality. We also propose and implement a method to fine-tune the model to generate acrostic poetry. To the best of our knowledge, this is the firs…
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 its consecutive variants have been proposed to further improve the performance of the pre-trained language models. In this paper, we aim to first introduce the whole word masking (wwm) strategy for Chinese BERT, along with a series of Chinese pre-trained language models. Then we also propose a simple but effective model called MacBERT, which improves upon RoBERTa in several ways. Especially, we propose a new masking strategy called MLM as correction (Mac). To demonstrat…
State Key Laboratory of Media Convergence Production Technology and Systems（媒体融合生产技术与系统国家重点实验室）；MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University（上海交通大学人工智能研究院教育部人工智能重点实验室）
Chinese short text matching is a fundamental task in natural language processing. Existing approaches usually take Chinese characters or words as input tokens. They have two limitations: 1) Some Chinese words are polysemous, and semantic information is not fully utilized. 2) Some models suffer potential issues caused by word segmentation. Here we introduce HowNet as an external knowledge base and propose a Linguistic knowledge Enhanced graph Transformer (LET) to deal with word ambiguity. Additionally, we adopt the word lattice graph as input to maintain multi-granularity information. Our model…
Biomedical text mining is becoming increasingly important as the number of biomedical documents and web data rapidly grows. Recently, word representation models such as BERT has gained popularity among researchers. However, it is difficult to estimate their performance on datasets containing biomedical texts as the word distributions of general and biomedical corpora are quite different. Moreover, the medical domain has long-tail concepts and terminologies that are difficult to be learned via language models. For the Chinese biomedical text, it is more difficult due to its complex structure an…
EEG2TEXT-CN: An Exploratory Study of Open-Vocabulary Chinese Text-EEG Alignment via Large Language Model and Contrastive Learning on ChineseEEG
Department of Artificial Intelligence in Healthcare, International Academia of Biomedical Innovation Technology（人工智能医疗系，国际生物医学创新技术学院）；National Taiwan University（台湾大学）；Neuro Industry Research, Neuro Industry, Inc.（神经产业研究，神经产业公司）；Department of Biomedical Engineering, University of Southern California（生物医学工程系，南加州大学）；Taiwan Artificial Intelligence Association（台湾人工智能协会）
The current paper concerns language transfer at the phonetic level and concentrates on the transfer phenomenon in an advanced English language learner's acquisition of the English vowels /i/ and its lax counterpart. By determining whether the Chinese English-language learner (ELL), named Vanya, can accurately distinguish between /i/ and its lax counterpart, and pronounce them precisely in General American English (GAE), this paper serves as a reference for further studying language transfer among Chinese ELLs. There were two objectives: first, the learner's perceptual ability to distinguish be…
University of Toronto（多伦多大学）；Vector Institute for Artificial Intelligence（Vector人工智能研究所）
Classical Chinese is a language notable for its word class flexibility: the same word may often be used as a noun or a verb. Bisang (2008) claimed that Classical Chinese is a precategorical language, where the syntactic position of a word determines its part-of-speech category. In this paper, we apply entropy-based metrics to evaluate these claims on historical corpora. We further explore differences between nouns and verbs in Classical Chinese: using psycholinguistic norms, we find a positive correlation between concreteness and noun usage. Finally, we align character embeddings from Classica…
Shuo Wen Jie Zi: Rethinking Dictionaries and Glyphs for Chinese Language Pre-training
Peking University（北京大学）；Beijing Institute for General Artificial Intelligence (BIGAI)（北京通用人工智能研究院）；National Key Laboratory of General Artificial Intelligence（通用人工智能全国重点实验室）
The First ChineseBabyLM Challenge: training data-efficient and cognitively plausible language models for Chinese
Princeton University（普林斯顿大学）；Shanghai Jiao Tong University（上海交通大学）；Chinese Academy of Sciences（中国科学院）；Columbia University（哥伦比亚大学）；Beijing Normal University（北京师范大学）；Tsinghua University（清华大学）；University of California San Diego（加利福尼亚大学圣地亚哥分校）；The Hong Kong Polytechnic University（香港理工大学）
In this paper, we propose a joint algorithm for the word segmentation on Chinese social media. Previous work mainly focus on word segmentation for plain Chinese text, in order to develop a Chinese social media processing tool, we need to take the main features of social media into account, whose grammatical structure is not rigorous, and the tendency of using colloquial and Internet terms makes the existing Chinese-processing tools inefficient to obtain good performance on social media. In our approach, we combine CRF and MMSEG algorithm and extend features of traditional CRF algorithm to trai…
Tianjin University（天津大学）；University of Edinburgh（爱丁堡大学）
Text corpora annotated with language-related properties are an important resource for the development of Language Technology. The current work contributes a new resource for Chinese Language Technology and for Chinese-English translation, in the form of a set of TED talks (some originally given in English, some in Chinese) that have been annotated with discourse relations in the style of the Penn Discourse TreeBank, adapted to properties of Chinese text that are not present in English. The resource is currently unique in annotating discourse-level properties of planned spoken monologues rather…
Tibetan is a low-resource language. In order to alleviate the shortage of parallel corpus between Tibetan and Chinese, this paper uses two monolingual corpora and a small number of seed dictionaries to learn the semi-supervised method with seed dictionaries and self-supervised adversarial training method through the similarity calculation of word clusters in different embedded spaces and puts forward an improved self-supervised adversarial learning method of Tibetan and Chinese monolingual data alignment only. The experimental results are as follows. First, the experimental results of Tibetan …
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 …
International Digital Economy Academy (IDEA)（国际数字经济研究院（IDEA））；Beijing University of Posts and Telecommunications（北京邮电大学）；Nanjing University（南京大学）
Bidirectional Encoder Representations from Transformers or BERT~\cite{devlin-etal-2019-bert} has been one of the base models for various NLP tasks due to its remarkable performance. Variants customized for different languages and tasks are proposed to further improve the performance. In this work, we investigate supervised continued pre-training~\cite{gururangan-etal-2020-dont} on BERT for Chinese topic classification task. Specifically, we incorporate prompt-based learning and contrastive learning into the pre-training. To adapt to the task of Chinese topic classification, we collect around 2…
In this paper, we improve Chinese spoken language understanding (SLU) by injecting word information. Previous studies on Chinese SLU do not consider the word information, failing to detect word boundaries that are beneficial for intent detection and slot filling. To address this issue, we propose a multi-level word adapter to inject word information for Chinese SLU, which consists of (1) sentence-level word adapter, which directly fuses the sentence representations of the word information and character information to perform intent detection and (2) character-level word adapter, which is appli…
China University of Geosciences (Beijing)（中国地质大学（北京））；Shenzhen University（深圳大学）；Tencent AI Lab（腾讯人工智能实验室）
Scientific literature serves as a high-quality corpus, supporting a lot of Natural Language Processing (NLP) research. However, existing datasets are centered around the English language, which restricts the development of Chinese scientific NLP. In this work, we present CSL, a large-scale Chinese Scientific Literature dataset, which contains the titles, abstracts, keywords and academic fields of 396k papers. To our knowledge, CSL is the first scientific document dataset in Chinese. The CSL can serve as a Chinese corpus. Also, this semi-structured data is a natural annotation that can constitu…
School of Systems Science, Beijing Normal University（北京师范大学系统科学学院）；New England Complex Systems Institute（新英格兰复杂系统研究所）；Brandeis University（布兰迪斯大学）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；Business College, Shanghai Normal University（上海师范大学商学院）
Corpus-based statistical analysis plays a significant role in linguistic research, and ample evidence has shown that different languages exhibit some common laws. Studies have found that letters in some alphabetic writing languages have strikingly similar statistical usage frequency distributions. Does this hold for Chinese, which employs ideogram writing? We obtained letter frequency data of some alphabetic writing languages and found the common law of the letter distributions. In addition, we collected Chinese literature corpora for different historical periods from the Tang Dynasty to the p…
Chinese couplet is a special form of poetry composed of complex syntax with ancient Chinese language. Due to the complexity of semantic and grammatical rules, creation of a suitable couplet is a formidable challenge. This paper presents a transformer-based sequence-to-sequence couplet generation model. With the utilization of AnchiBERT, the model is able to capture ancient Chinese language understanding. Moreover, we evaluate the Glyph, PinYin and Part-of-Speech tagging on the couplet grammatical rules to further improve the model.
Although pre-trained language models have remarkably enhanced the generation ability of dialogue systems, open-domain Chinese dialogue systems are still limited by the dialogue data and the model size compared with English ones. In this paper, we propose EVA, a Chinese dialogue system that contains the largest Chinese pre-trained dialogue model with 2.8B parameters. To build this model, we collect the largest Chinese dialogue dataset named WDC-Dialogue from various public social media. This dataset contains 1.4B context-response pairs and is used as the pre-training corpus of EVA. Extensive ex…
