Quick Search
  Office Online
  Journal Online
Current Issue
Archive
Advanced Search
Read Articles
Download Articles
RSS
Email Alert
2024 Vol. 9, No. 5
Published: 2024-09-10

 
1 Language Planning in the Digital Intelligence Era Must Go with the Flow
Liu Haitao
2024 Vol. 9 (5): 1-1 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
5 Introduction: Large Language Models and Linguistic Research: Mutual Enhancement and Synergistic Development
Yang Erhong and Hu Renfen
2024 Vol. 9 (5): 5-6 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
7 SpaCE: A Linguistic Knowledge-Driven Benchmark for Spatial Cognition Evaluation
Zhan Weidong, Sun Chunhui and Xiao Liming
DOI: 10.19689/j.cnki.cn10-1361/h.20240501

Over the past two decades, deep learning technology has propelled machine natural language processing capabilities to rival or even surpass human levels in many tasks. Machine learning does not directly utilize the outcomes of human linguistic research (knowledge), but rather from human language materials (data). This situation should garner significant attention from linguists. As large language models, driven purely by data and computational power, have nearly constructed a modern Tower of Babel, the question of how to realize the value of linguistic knowledge through in-depth exploration of specific and subtle language phenomena looms large over every linguistic researcher. This paper proposes a research approach that generates text data from linguistic knowledge for evaluating machine understanding of spatial semantics. Over the past four years, we have organized four consecutive competitions on Chinese Spatial Cognition Evaluation (SpaCE): from SpaCE2021 to SpaCE2024, including 6 sub-tasks: Determination of Spatial Information Validity, Detection of Spatial Anomalies, Recovery of Spatial References, Identification of Spatial Semantic Roles, Recognition of Spatial Equivalences, and Spatial Position Reasoning. This paper introduces the design philosophy, dataset creation process, dataset overview, and the performance characteristics of machines in SpaCE tasks. Overall, large language models participating in the SpaCE competitions perform relatively well on tasks that rely on surface distribution features, that is, tasks with formal cues, but poorly on tasks that depend on deep semantic understanding, that is, tasks requiring cognitive abilities. In the current era of rapid AI development, where linguistic knowledge is passively marginalized in the field of natural language processing, the value of linguistic knowledge needs to be redefined. It should be used to guide the production of small, high-quality language data to enhance the effectiveness and efficiency of machine learning. For computational applications, grammatical research should pursue more challenging goals—adequate generation—beyond the objectives of adequate observation, description, and explanation.

2024 Vol. 9 (5): 7-21 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
22 Construction and Application of Ancient Chinese Large Language Model
Li Shen, Hu Renfen and Wang Lijun
DOI: 10.19689/j.cnki.cn10-1361/h.20240502
General-purpose large language models demonstrate notable capabilities in language comprehension and generation, achieving results comparable to or surpassing human performance in many language information processing tasks. Nevertheless, when general models are applied to ancient Chinese texts, their effectiveness is often unsatisfactory. Similarly, fine-tuning open-source foundational models encounter difficulties in adequately incorporating domain-specific knowledge. To address this challenge, this study introduces a large language model, AI Taiyan, specifically designed for understanding and generating ancient Chinese. Experimental results show that with an appropriately designed model, data processing, foundational training, and fine-tuning, satisfactory results can be achieved with merely 1.8 billion parameters. In key tasks related to ancient Chinese information processing such as punctuation, identification of allusions, explanation of word meanings, and translation between ancient and modern Chinese, this model exhibits a clear advantage over both general-purpose large models and domain-specific models, achieving or surpassing human baseline performance. This research provides a reference framework for the efficient construction of specialized domain-specific large language models. Furthermore, the paper discusses the application of this model in fields such as the collation of ancient texts, dictionary editing, and language research, supplemented by case studies.
2024 Vol. 9 (5): 22-33 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
34 A Study on the Evaluation of Large Language Models’ Capabilities in Chinese Text Simplification
Yang Erhong, Zhu Junhui, Zhu Haonan, Zong Xuquan and Yang Lin’er
DOI: 10.19689/j.cnki.cn10-1361/h.20240503
Large language models (LLMs) offer new approaches for automatic text simplification. To explore the capabilities of LLMs in simplifying Chinese texts, this study constructed a Chinese passage-level text simplification dataset and conducted a feature analysis of the parallel text pairs within it. Based on this, an experiment was designed to assess the automatic text simplification performance of LLMs using four prompting strategies: zero-shot, few-shot, few-shot with lexicon, and few-shot with rules. The study evaluated the performance of six commonly used domestic and international LLMs in Chinese text simplification under different prompting strategies, utilizing a combination of existing and study-specific linguistic feature evaluation metrics. The findings revealed that the few-shot prompting strategy performed best in terms of text features, significantly enhancing information retention. Incorporating external lexicons in the prompts helped the LLMs use relatively simpler words, while integrating simplification rules enabled the LLMs to employ more concise syntactic structures. Different LLMs exhibited distinct strengths and limitations in controlling complexity and preserving semantics, but all showed a noticeable gap compared to human experts in discourse cohesion, coherence, and paragraph segmentation, with varying degrees of hallucination also observed. Future research should focus on constructing larger-scale, high-quality Chinese simplification datasets and exploring multi-faceted approaches to enhance the text simplification capabilities of LLMs.
2024 Vol. 9 (5): 34-47 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
48 Diverse Views on Large Language Model
Han Xianpei, Li Juanzi, Liu Pengyuan, Liu Zhiyuan, Wang Bin and Zong Chengqing
2024 Vol. 9 (5): 48-51 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
52 Introduction: Some Reflections on the Era of Teaching Chinese as a Second Language
Li Quan
2024 Vol. 9 (5): 52-53 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
54 The Disciplinary Foundations of “Population Profile” in International Chinese Language Education
Wang Hanwei and Zhang Xinyue
DOI: 10.19689/j.cnki.cn10-1361/h.20240504
The notion of “international” in the field of international Chinese language education is essentially defined by the international population, which comprises two heterogeneous groups: overseas Chinese and non-Chinese. The development of international Chinese language education necessitates a detailed analytical understanding of these populations. Overseas Chinese form the core group of international Chinese language education, while non-Chinese represent the potential main group. Chinese language education tailored to these groups pertains to the inheritance of Chinese ancestral language and the dissemination of Chinese as a second language, respectively. Considering the current situation of Chinese ancestral language preservation within both established and emerging Chinese communities, the target audience for Chinese ancestral language inheritance can be categorized into three groups: those with a robust family language environment, those with a moderate family language environment, and those with minimal or no family language environment. This categorization informs the design of distinct core curricula, textbooks, and teaching methods. For the dissemination of Chinese as a second language, the target audience can be further classified based on their native languages into Chinese character cultural circle and non-Chinese character cultural circle. The latter can be further divided into “peripheral” and “non-peripheral” groups based on their geographical proximity to China. Theoretical exploration and teaching practices that consider the relative linguistic and script distances between Chinese and other languages can avoid the Chinese centrism and address the intrinsic differences between Chinese and other languages. The progressive refinement of the target population and the corresponding deepening of research represent the developmental trajectory of international Chinese language education. By enriching the composition of the target population, a more rational division of labor and allocation of academic and social resources can be achieved, thereby fulfilling the objective of providing precise and effective educational services.
2024 Vol. 9 (5): 54-63 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
64 A Study on the Establishment of Chinese Language Special Course for Eastern European Exchange Students at Tsinghua University
Shi Zhengyu and Zhao Mei
DOI: 10.19689/j.cnki.cn10-1361/h.20240505
The Chinese Language Special Course for Eastern European Exchange Students at Tsinghua University was presided over by the Premier of the Government Administration Council Zhou Enlai, and operated in collaboration with multiple departments such as the Ministry of Foreign Affairs, the Cultural and Education Commission, and the Ministry of Education, with direct participation of Chairman Mao Zedong, Vice Chairman Liu Shaoqi and Zhu De, Vice Premier of the Government Administration Council and Director of the Financial and Economic Commission Chen Yun, after the establishment of the People’s Republic of China. The program originated from an initiative to exchange students between China and five Eastern European countries. Both parties extensively negotiated curriculum content, program duration, student qualifications, and living arrangements, forming an early model for international student education and management in the early years of the founding of the People’s Republic of China. Officially inaugurated in September 1950, the program operated at Tsinghua until its transfer to Peking University in autumn 1952. During this period, it employed 17 faculty and staff members and enrolled 36 students. Its program’s administration operated on three levels: Governmental, i.e. an Eastern European exchange student working group comprised of relevant ministry personnel; Institutional, i.e. a guidance committee for foreign students, consisting of renowned scholars from the university; Departmental, i.e. the Chinese Language Specialization Program for Eastern European Exchange Students at Tsinghua University, administratively equivalent to a department, with Deng Yi overseeing teaching operations. The program adhered to two key principles: ‘division of labor and cooperation’ and ‘democratic centralism’. In retrospect, this program not only cultivated a cohort of foreign students knowledgeable about and friendly towards China but also laid a solid foundation for the development of Chinese as a foreign language education. It holds a significant place in the history of Chinese language education both in China and even in the world.
2024 Vol. 9 (5): 64-75 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
76 The Engineering of International Chinese Language Education: Definition, Features, and Principles
Shi Chunhong and Ma Ruiling
DOI: 10.19689/j.cnki.cn10-1361/h.20240506
With the rapid development of international Chinese language education, much academic attention has been paid to the application of engineering thinking to solve the relevant problems in international Chinese language education. This study focuses on the engineering of international Chinese language education, which can be roughly defined as the use of relevant knowledge and technology to configure educational resources and teaching methods, and the construction of products that can be used in international Chinese language education in an integrated way according to certain norms and standards. Its features can be summarised from the perspectives of its goal and operation. The former includes demand-driven, product-oriented, and function-assisted, while the latter includes element integration, structural modularity, protocol methodization, and operational platform. Starting from ontology, epistemology, methodology, and value theory, four principles can be proposed for the engineering of international Chinese language education: the combination of holistic and block thinking, the combination of instrumentality and humanity, the combination of data-driven and human-computer collaboration, and the combination of scaling and personalisation. The practice of engineering has become an important and fundamental path to promote the emergence and development of a new ecology of international Chinese language education.
2024 Vol. 9 (5): 76-86 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
87 Discussion on Chinese Teaching in the Intelligent Era from the Perspective of Differences Between Chinese and English as Foreign Languages
Lu Jianming
DOI: 10.19689/j.cnki.cn10-1361/h.20240507
This study examines the distinct characteristics of teaching Chinese and English as foreign languages which are highlighted by the following: For Chinese learners of English, oral and written language acquisition typically progress concurrently. However, for English-speaking students learning Chinese, there is a notable disparity between oral and written language acquisition as they cannot synchronize learning of both spoken and written language. The transition from oral to written Chinese necessitates additional effort and time investment in mastering Chinese characters, presenting a significant challenge. The advent of intelligent technologies, which enable character input through pinyin or voice recognition, offers new opportunities to dispel the myth that Chinese is inherently difficult to learn. International scholars have achieved positive outcomes by implementing a “primarily digital writing, supplemented by handwriting” teaching approach, a teaching model that prioritizes digital writing supplemented by handwriting. This paper proposes a new approach: “Pave the way with digital character input, followed by handwriting practice.” In the initial stages of Chinese language learning, students are first taught and guided to input characters digitally, alleviating anxiety associated with character learning. Subsequently, handwriting exercises are gradually integrated into the curriculum, facilitating a transition from digital input to handwriting. This approach recognizes that handwriting contributes to character memorization, appreciation, and comprehension, ultimately ensuring proficiency in written Chinese. The intelligent era presents an opportunity to reevaluate and innovate Chinese language teaching methodologies, potentially revolutionizing the field of Chinese as a foreign language education.
2024 Vol. 9 (5): 87-91 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
92 Diverse Views on International Chinese Language Education
Cui Xiliang, Cui Yonghua, Wang Haifeng, Wu Yongyi, Zhang Wangxi, Zheng Yanqun and Zhou Xiaobing
2024 Vol. 9 (5): 92-96 [Abstract] ( 0 ) HTML (1 KB)  PDF  (0 KB)  ( 0 )
Copyright © Editorial Board of
Supported by:Beijing Magtech