|
Modern Linguistics 2025
逻辑语义功能视域下Kimi ai写作优化效果研究
|
Abstract:
本研究基于系统功能语言学的逻辑语义功能理论,采用语篇对比分析与参与式跟踪调查相结合的方法,对陕西省某高校英语专业大二学生的写作样本进行研究,旨在评估Kimi ai在不同水平写作样本优化中的实际效能。结果显示,Kimi ai在低至中等水平文本优化方面表现显著,通过扩展意义和句式变换提升了文本质量;然而,在高水平写作样本优化中,尽管能够通过复合连词植入和句式重组实现表层句法改良,但在扩展关系句式比例和整体文本评分上未达到显著差异。此外,其显性连接策略与写作者的隐性连贯机制存在冲突,且受限于命题框架的突破能力,未能有效提升深层逻辑连贯性和概念层级重组,从而揭示了当前AI工具在语篇认知增强方面的功能性局限。这一研究为二语写作教学及GenAI写作优化功能的开发提供了实证支持。
This study, grounded in the logical semantic function theory of systemic functional linguistics, employs a combined approach of discourse contrastive analysis and participatory tracking surveys to investigate the writing samples of second-year English majors at a university in Shaanxi Province. The aim is to evaluate the practical effectiveness of Kimi AI in optimizing writing samples across different proficiency levels. The results show that Kimi AI significantly enhances the quality of low- to mid-level texts through meaning expansion and syntactic transformation. However, in high-level writing optimization, while Kimi AI can improve surface syntax through the insertion of complex conjunctions and sentence restructuring, it fails to achieve significant differences in the proportion of extended relational sentences and overall text scores. Furthermore, its explicit connection strategies conflict with the implicit coherence mechanisms of writers, and its limited ability to break through propositional frameworks prevents effective enhancement of deep logical coherence and conceptual reorganization. These findings reveal the functional limitations of current AI tools in enhancing discourse cognition and provide empirical support for second language writing instruction and the development of GenAI writing optimization functions.
[1] | 张明飞, 孔新梅, 王显闯, 方海光. 基于生成式人工智能的探究式教学设计与应用研究[J]. 中国现代教育装备, 2024(18): 1-4. |
[2] | 王一岩, 朱陶, 杨淑豪, 郑永和. 人机协同教学: 动因、本质与挑战[J]. 电化教育研究, 2024, 45(8): 51-57. |
[3] | 欧志刚, 刘玉屏, 覃可, 王艺, 李欣燚. 人工智能多模态教学资源的生成与评价——基于AIGC在国际中文教育的应用[J]. 现代教育技术, 2024, 34(9): 37-47. |
[4] | 曹晓明, 张永和, 潘萌, 朱姗, 闫海亮. 人工智能视域下的学习参与度识别方法研究——基于一项多模态数据融合的深度学习实验分析[J]. 远程教育杂志, 2019, 37(1): 32-44. |
[5] | 方建锋, 王克宇, 房欲飞. 生成式人工智能对教育的颠覆性影响和应对[J]. 全球教育展望, 2024, 53(8): 17-32. |
[6] | 刘骥, 丘霖. 生成式人工智能嵌入教育应用的风险生成及其规制[J]. 现代远距离教育, 2024(4): 12-19. |
[7] | 陈安天, 卢军, 张新庆. 生成式人工智能对医患共享决策的影响机制探究[J]. 中国医学伦理学, 2024, 37(9): 1087-1092. |
[8] | 蒋小英, 宁启兰, 闫小飞, 武丽涛, 李海韵. 生成式人工智能在基础医学教学中的应用前景[J]. 基础医学教育, 2024, 26(6): 508-511. |
[9] | 陈炳澍, 王玥. 人工智能医疗监管与治理创新: 突破现行规范框架的思考[J]. 中国医学伦理学, 2024, 37(9): 1030-1036. |
[10] | 严驰. 论人工智能标准与法律协同治理体系之构建[J]. 北京理工大学学报(社会科学版), 2024, 26(6): 109-121. |
[11] | 高志宏. 回应与超越: 生成式人工智能法律规制——以《生成式人工智能服务管理暂行办法》为视角[J]. 社会科学辑刊, 2024(5): 121-130. |
[12] | 卢家银. 美德与规范的内在主义协同: 生成式人工智能的传播伦理进路[J]. 新闻界, 2024(8): 15-23. |
[13] | 南婷. 主流意识形态传播中的生成式人工智能与媒体融合[J]. 融媒, 2024(9): 4-9. |
[14] | Yeo, M.A. (2023) Academic Integrity in the Age of Artificial Intelligence (AI) Authoring Apps. TESOL Journal, 14, e716. https://doi.org/10.1002/tesj.716 |
[15] | Ingley, S.J. and Pack, A. (2023) Leveraging AI Tools to Develop the Writer Rather than the Writing. Trends in Ecology & Evolution, 38, 785-787. https://doi.org/10.1016/j.tree.2023.05.007 |
[16] | Elliot, N. and Klobucar, A. (2013) Automated Essay Evaluation and the Teaching of Writing. In: Shermis, M.D. and Burstein, J., Eds., The Handbook of Automated Essay Evaluation: Current Applications and New Directions, Routledge, 16-35. https://www.scirp.org/reference/referencespapers?referenceid=2885861 |
[17] | 毛延生, 王一航, 邢艳茹. ChatGPT辅助高中英语写作反馈的实证研究[J]. 教育测量与评价, 2024(1): 3-13. |
[18] | 朱永生. 功能语言学导论[M]. 上海: 上海外语教育出版社, 2004. |
[19] | 彭程, 鲍珍. 二语水平及任务复杂度对概要写作任务的影响[J]. 当代外语研究, 2024(3): 110-120+139. |
[20] | 杨晓荣. 从答卷看教学——TEM4-93作文阅卷手记[J]. 语教学与研究, 1993, 25(4): 70-72. |
[21] | 杨玉晨, 闻兆荣. 中国学生英文写作的句子类型及分析[J]. 现代外语, 1994, 17(1): 39-42+67. |
[22] | 张在新, 吴红云, 王晓露, 张俊香. 我国英语写作教学中的主要问题[J]. 外语教学与研究, 1995, 27(4): 43-50. |
[23] | 马广惠, 文秋芳. 大学生英语写作能力的影响因素研究[J]. 外语教学与研究, 1999(4): 34-39+78. |
[24] | 胡萍萍. 不同写作水平学生在英语作文质量上的差异[J]. 浙江师范大学学报, 2005, 30(2): 75-79. |
[25] | VanLehn, K. (2011) The Relative Effectiveness of Human Tutoring, Intelligent Tutoring Systems, and Other Tutoring Systems. Educational Psychologist, 46, 197-221. https://doi.org/10.1080/00461520.2011.611369 |
[26] | Gulwani, S., Radiček, I. and Zuleger, F. (2018) Automated Clustering and Program Repair for Introductory Programming Assignments. Proceedings of the 39th ACM SIGPLAN Conference on Programming Language Design and Implementation, Philadelphia, 18-22 June 2018, 465-480. https://doi.org/10.1145/3192366.3192387 |
[27] | Mayer, R.E. and Fiorella, L. (2021) Introduction to Multimedia Learning. In: The Cambridge Handbook of Multimedia Learning, Cambridge University Press, 3-16. https://doi.org/10.1017/9781108894333.003 |
[28] | Radford, A. (2021) Learning Transferable Visual Models from Natural Language Supervision. |
[29] | Koedinger, K.R., Corbett, A.T. and Perfetti, C. (2012) The Knowledge-Learning-Instruction Framework. Cognitive Science, 36, 757-798. |
[30] | Baker, R.S. and Inventado, P.S. (2014) Educational Data Mining and Learning Analytics. In: Learning Analytics, Springer, 61-75. https://doi.org/10.1007/978-1-4614-3305-7_4 |
[31] | D’Mello, S., Graesser, A. and Hu, X. (2014) AutoTutor and Family: A Review of 17 Years of Natural Language Tutoring. International Journal of Artificial Intelligence in Education, 24, 427-469. |