主题报告人介绍

（1）梁茂成

梁茂成，北京航空航天大学外国语学院教授、博士生导师、院长。中国英汉语比较研究会语料库语言学专业委员会会长、语料库翻译学专业委员会副会长、中国高等教育学会外语教学研究分会副理事长，国务院外国语言文学学科评议组成员，担任国家社科基金、教育部人文社科基金、中国博士后基金、中国留学基金委基金等项目评审专家，主要研究兴趣涉及语料库语言学、数据科学、应用语言学等，主持国家社科基金、教育部人文社科基金、教育部人文社科重点研究基地重大项目等省部级以上项目十余项，著有《中国学生英语作文自动评分模型的构建》、《大规模考试英语作文自动评分系统的研制》、《语料库应用教程》（合著）、《什么是语料库语言学》等，发表各类学术论文70余篇，开发各类语言数据处理和语料库分析工具近30款。

（2）卫乃兴

Wei Naixing is a funding member of the Asia Pacific Corpus Linguistics Association and founding Chairman of the Corpus Linguistics Society of China. He is currently Professor of English and Chairman of Academic Committee of the Foreign Languages School at Beihang University. His major research interests cover corpus linguistics, phraseology and discourse analysis, among others. He has published profusely, along with undertaking numerous state-sponsored research projects. He has also served successively as Professor of English at Shanghai Jiao Tong University and Academic Adviser, Adjunct/ Visiting Professor at Hong Kong City University, Hong Kong Polytechnic University and other key universities acrossChina, supervising/training corpus linguists of the younger generation, helping push forward corpus linguistics in China and internationally.

（3）Susan Hunston

Susan Hunston is Professor of English Language at the University of Birmingham, UK. She specialises in corpus linguistics and discourse analysis and has published books and articles on topics such as: the interface between grammar and lexis of English, evaluative language, academic discourse, and the interface between corpus and discourse studies. She is co-author of Pattern Grammar (2000 Benjamins, with Gill Francis), and her most recent book is the 2nd edition of Corpora in Applied Linguistics (2022 Cambridge University Press). She is currently holder of a Leverhulme Emeritus Fellowship, working on a project to unify Pattern Grammar, Construction Grammar and Systemic Functional Grammar.

（4）Hilary Nesi

Hilary’s research activities mostly concern corpus development and analysis, the discourse of English for academic purposes, and the design and use of dictionaries and reference tools in academic contexts. She is Co-Editor of the Journal of English for Academic Purposes and the 3rd edition of the Elsevier Encyclopedia of Language and Linguistics (forthcoming), and was principal investigator for the projects to create the BASE corpus of British Academic Spoken English and the BAWE corpus of British Academic Written English. She has published more than 100 articles and book chapters.

（5）邓耀臣

邓耀臣，大连外国语大学教授、博士生导师，《外语与外语教学》主编，享受国务院政府特殊津贴专家。研究方向为语料库语言学理论及应用、计算术语学。主持完成国家社科基金一般项目2 项，国家社科基金重大项目子课题1 项，参与国家社科基金重大项目、国家社科基金、教育部社科基金一般项目多项。在Applied Linguistics、International Journal of Corpus Linguistics、Journal of Quantitative Linguistics、《外国语》、《中国外语》、《外语教学》、《外语与外语教学》等国内外外语类权威期刊发表论文40余篇，主编出版学术专著3部。

（6）甄凤超

甄凤超，外国语言学及应用语言学博士，上海交通大学外国语学院教授、博士生导师，《当代外语研究》副主编，International Journal of Corpus Linguistics、Applied Corpus Linguistics编委。研究方向为语料库语言学、语料库辅助的外语教学、语料库驱动的话语分析。在《外语教学与研究》《外国语》、TESOL Quarterly等期刊上发表学术论文40余篇，出版学术专著3部、教材3本。主持1项国家社科基金青年项目和1项国家社科基金重点项目。

（7）苏杭

苏杭，伯明翰大学博士，现为四川外国语大学外国语文研究中心主任、教授、博士生导师、校学术委员会委员。研究兴趣包括语料库语言学、系统功能语言学、（语料库）语用学、学术英语等；主持国家社科基金等科研项目三项，并在Applied Linguistics、《外语教学与研究》等刊物发表论文50余篇。2019年入选重庆英才•青年拔尖人才支持计划，2024年入选重庆市学术技术带头人后备人选；获得高等教育（研究生）国家级教学成果奖二等奖（2023）、重庆市高等教育教学成果奖三等奖（2022）以及四川外国语大学优秀科研成果一等奖（2022、2024）等奖项。

（8）姜峰

姜峰，吉林大学公共外语教育学院副院长，匡亚明特聘教授，博士生导师。国家“万人计划”青年拔尖人才，吉林省有突出贡献中青年专家，吉林省拔尖创新人才（第二层次），吉林省社会科学研究“十四五”规划学科专家。香港大学应用语言学博士，师从国际著名应用语言学家Ken Hyland教授，研究方向为学术语篇、语料库和学术写作。担任三部SSCI一区期刊Journal of English for Academic Purposes、English for Specific Purposes和Teaching in Higher Education编委。在Applied Linguistics, English for Specific Purposes等SSCI国际权威期刊发表论文70余篇，以及多篇中文论文发表在《外语教学与研究》、《现代外语》等外语类CSSCI核心期刊，在Routledge、外语教学与研究出版社出版学术专著4部，主持国家社科基金重点项目、吉林省教育厅重点项目等各类课题十余项。入选Elsevier中国高被引学者以及全球前2%顶尖科学家。

（9）邹艳丽

邹艳丽，伯明翰大学应用语言学博士，师从国际著名语料库语言学家Susan Hunston教授，主要研究兴趣包括语料库语言学、数据驱动学习、学术语篇、语料库文体学等。完美体育平台官网教授，硕士生导师，语料库与数据驱动学习研究所负责人。主持省级项目5项，出版专著和规划教材5部，在English for Specific Purposes，Journal of Pragmatics等期刊发表多篇论文。主持省级研究生精品示范课《学术英语阅读与写作》。

主题报告内容介绍

（1）局部语法研究：分析与实践

主讲：苏杭 四川外国语大学

助理：叶军 四川外国语大学

The concept of ‘local grammar’ initially originated from Gross (1993) and has been further developed by Sinclair and his colleagues (e.g., Hunston & Sinclair 2000; Barnbrook & Sinclair 2001; Hunston 2002; Barnbrook 2002; Sinclair 2010, 2016). While Gross (1993) was primarily concerned with applying local grammar into develop a finite automata, recent local grammar research mainly focused on describing particular meaning and/or function. Hunston (2002: 178), for example, points out that each local grammar “seeks to account for, not the whole of a language, but one meaning only’’, and Hunston and Su (2019: 571) explicitly state that a local grammar “is always a grammar of a discourse function”.

The rationale of promoting local grammar research has been adequately justified by previous research. In particular, Barnbrook and Sinclair (2001) note that the kind of specialised local grammar can “outperform” general grammar, and Butler (2004: 158) further argue that “rather than a single general grammar, we might end up with a set of local grammars for particular areas defined by their communicative functions in the discourse”.

Not surprisingly, then, recent years have witnessed some studies that employed the local grammar approach to develop specialised grammars of specific meanings or functions (e.g., Cheng & Ching 2016; Su 2017, 2018; Su & Wei 2018; Su & Fu 2023; Yu et al. 2024; Zhang et al. 2024). Despite its increasing popularity, however, newcomers of local grammar research face a number of challenges (e.g., data retrieval, terminology identification). To overcome these challenges, this workshop will address methodological issues concerning local grammar research and provide guidelines for practicing local grammar analyses, aiming to promote local grammar research and, more importantly, to develop new analytic techniques in corpus linguistics.

（2）AI-assisted Corpus-based Discourse Studies:

Significance and limitations of machine learning techniques

卫乃兴 北京航空航天大学

Corpus-based discourse study (CBDS) scholars are now increasingly faced with the issue of maintaining the total accountability of data on one hand, and also attending to hidden attitudes and meanings in text on the other hand, when dealing with sizable data-sets in the present mega-corpora era. To address this issue, this talk sets out to discuss the significance of implementing in CBDS work the machine learning techniques of Topic Modelling, Vector Modelling and Clustering analysis, for a possible improvement of data treatment. I will first review the methodological features of current CBDS, commenting on their strengths and weaknesses. Secondly, I explore the uses of Topic Modelling, Vector Modelling and Clustering techniques for data treatment in discourse analysis respectively, with a focus on their significance and limitations. I show that these AI-techniques, used appropriately, can provide a satisfactorily usable data base upon which various research issues can be explored more effectively, whilst they also suffer weakness and limitations on linguistic grounds. I argue that these machine learning techniques have to be used in combination with well-established analytic techniques in discourse studies, and importantly, a close reading of the text is irreplaceable. Finally, I sum up my viewpoints, suggestions and caveats for this strand of research in the conclusion.

（3）语料库驱动的配价型式研究

甄凤超 上海交通大学

语料库语言学研究主张词汇与语法统一，语法是词汇的语法，语法型式是用来识别词汇意义的重要路径。Hunston和Francis所提出的型式语法（pattern grammar）过分强调了型式本身与意义的共选，但相对弱化了词汇的作用。配价型式研究，坚持语法型式服务于词汇的理念，既可以是词汇的一般语法，也可以是具体词汇的局部语法。该研究将通过具体的案例，详细介绍配价型式的描写框架及其拓展研究。

（4）Communicating disciplinary knowledge to a wide audience:How students engage with popularisation of science in 3MT presentations

姜峰 吉林大学

The rapid development of communication technologies and changing ways of knowledge exchange are bringing up new academic contexts in which students, as future scientists, are exposed to a more unpredictable grouping of audience and diversified forms of interaction. Three Minute Thesis (3MT) presentations emerged as a new academic genre, which train research postgraduate students to communicate their research within three minutes to a general audience. In this talk, I seek to present the rhetorical language use students draw on to construct a persuasive speech while building an inclusive relationship with the audience. The analysis helps us to see how the particular rhetorical contexts of genre, register and disciplinary knowledge shape students’ use of the interactional resources. This also suggests how students understand the connection between their disciplinary knowledge and real-world concerns and the way they adapt their discourse accordingly.

（5）数据驱动学习在EAP教学中的应用：三角验证视角

邹艳丽 完美体育平台官网

本报告聚焦数据驱动学习（Data-driven Learning, DDL）在学术英语（English for Academic Purposes, EAP）课程中的应用效果与教学价值。在简要概述DDL核心概念及其在语言教学中实现路径的基础上，基于具体的EAP写作课程案例，通过三角验证方法（triangulation）探索DDL的教学效果。研究方法包括：（1）学生利用语料库识别学术文本中的修辞语步词束；（2）将识别出的语步词束应用于具体写作任务；（3）学习者学科素养与体裁意识的自测问卷；（4）学生对DDL教学的评价。研究结果表明，DDL在培养学习者体裁意识和提升写作素养方面具有积极作用，多数学习者认为DDL在EAP学习中具有正面效应。通过多角度综合分析，本文呈现了DDL在EAP教学中的潜在优势，为数据驱动学习的语言教学实践提供新的研究视角和实证支持。

（6）Pattern, Construction, System: Approaches to lexis and grammar

Susan Hunston 伯明翰大学

There is a large body of agreement that lexis and grammar form a continuum from the most general features of grammar to specific behaviours of words. This has been described in a number of ways. From the perspective of Cognitive Grammar, it is Constructions. From the perspective of Systemic Functional Linguistics, it is System Networks. From the perspective of Corpus Linguistics, it is Pattern Grammar. This paper reports on a project designed to bring together these perspectives. Starting with grammar patterns, the project has identified over 800 Verb Argument Constructions in English. The constructions belonging to specific semantic fields are then identified, and system networks devised to show the choices involved as the different constructions contribute to that meaning. The project is illustrated using the semantic field of Equivalence as an example.

（7）Double negation in academic lectures

Hilary Nesi 考文垂大学

Studies such as Jiang and Hyland (2022) and Sun, Jiang and Liu (2024) have investigated the way negation is used in academic genres to express attitude, and engage with readers/listeners. However, it has long been established that negated propositions require more complex cognitive processing than affirmative propositions. Double negation (DN) is even more difficult to make sense of (Sherman, 1976), especially as it can imply a variety of different stance positions (see e.g. Tessler & Franke, 2019). These features make DN likely to cause speakers of other languages comprehension difficulties, especially if DN is less used, or used differently, in their own first languages. DN in English has been examined in literary and rhetorical studies, but research into its occurrence and use in university instruction is lacking. Our study investigated the forms and functions of DN in academic lectures delivered in English, consulting the British Academic Spoken English (BASE) corpus, the Michigan Corpus of Academic Spoken English (MICASE), a corpus of lectures delivered by Turkish lecturers in Turkey using English as the Medium of Instruction (TEMI), and a corpus of open courseware lectures from the US (OCC). We used AntConc (Anthony 2024) to search the corpora for negative words such as “not” and “no” followed by a word with a negative meaning (e.g. "not uncommon” or “not entirely wrong”). The co-text where DN occurred was subsequently analysed to develop a framework of DN functions in academic discourse, and reveal that a precise interpretation of DN meaning is often dependent on the lecture’s informational content and the listener’s cultural knowledge. DN was found not infrequently in corpora composed of lectures delivered in anglophone countries, but was vanishingly rare in the EMI corpus – it is therefore possible that participants in EMI lectures lack exposure to this important means of (possibly ironic) understatement which serves to heighten or reduce the force of claims or criticism. The talk will introduce the framework and discuss examples showing the subtle employment of DN in lectures. It is hoped that it will raise educators’ awareness of this rather neglected linguistic feature, and lead to a deeper understanding of variation in academic discourse across cultural and instructional settings.

（8）基于大语言模型的术语自动抽取研究

邓耀臣 大连外国语大学

术语自动抽取技术旨在从不断更新的专业文本中高效提炼专业术语，并为知识图谱构建、文本分类、主题建模等任务提供基础支持。近年来，大语言模型(Large Language Model, LLM)的兴起为术语自动抽取带来了新的机遇。LLM基于海量文本进行预训练，从而学习到了丰富的语言表示，具有较强的上下文理解及语义捕获能力，因而在复杂语言领域的信息抽取任务中表现出色。为了深入发掘LLM在术语自动抽取任务中的潜力，本研究探讨了基于LLM的术语自动抽取方法，并分析不同提示设计对抽取性能的影响。此外，本研究将基于LLM的方法与传统方法进行对比实验，以揭示前者在术语抽取中的独特优势。实验结果显示，基于LLM的术语自动抽取方法在准确性和效率上均有显著提升，并在多个领域的术语抽取任务中展现了稳定的性能。相较于传统方法，基于LLM的方法不依赖专业术语词典等外部资源，能够在小规模的抽取任务中取得理想的成果。该研究结果将为术语自动抽取相关领域的研究与实践带来新的启示。

（9）局部语法视角下的多语种言语行为研究

苏杭 四川外国语大学

本报告介绍言语行为研究的局部语法路径。报告将首先概述局部语法概念、研究方法及其研究价值，然后以英语、汉语和意大利语言语行为研究作为案例，具体呈现局部语法在多语种言语行为研究中的应用，并探讨相关研究的教学价值。

（10）大语言模型驱动的语料库语言学专业知识图谱构建

梁茂成 北京航空航天大学

大语言模型具有强大的推理能力和文本理解能力，在语言研究和语言教学领域具有广阔的应用前景。本研究尝试运用大语言模型构建学科和专业知识图谱。研究中对用户提供的文本进行优化，进而从优化后的文本中自动提取实体和关系，构建专业知识图谱。研究对各类专业知识图谱构建具有重要启示。