生成式AI译文初稿的启动效应与译者主体性受限——基于旅游宣传文本翻译的实证研究
Priming Effect of AI-Generated Translation Drafts and Constraints on Translator Subjectivity—Evidence from Tourism Promotional Translation
摘要: 在生成式AI广泛应用于翻译实践的背景下,AI译文初稿对译后编辑及译者主体性的影响已成为当前研究关注的重要议题。由此,本文聚焦旅游宣传文本,在生成式AI翻译、专家独立翻译与基于AI初稿的专家译后编辑三种条件下,考察生成式AI初稿的启动效应及其对译者主体性的影响。结果显示,以生成式AI初稿为基础的译后编辑译文在表层词句与深层语义两个层面均更趋近于AI初稿,说明译后编辑过程中存在较为明显的启动效应。同时,译者对AI初稿的修订主要表现为局部表达优化,在整体重写与结构重组等深层调整方面相对有限。这表明生成式AI初稿不仅会持续影响译后编辑,还在一定程度上压缩了译者在文本重写与结构重组中的主体性。这一结果不仅有助于深化对于译文形成机制的认识,也为专业翻译实践的流程优化、AI初稿的合理应用以及译者角色的重新定位提供了实证支持。
Abstract: As generative AI has become widely used in translation practice, how AI-generated drafts shape post-editing and affect translator subjectivity has emerged as an important issue in current research. Within this context, the present study investigated the priming effect of generative AI drafts and the impact of such drafts on translator subjectivity in the translation of tourism promotional texts. Specifically, three translation conditions were compared: generative AI translation, expert human translation, and expert post-editing based on AI-generated drafts. Results showed that post-edited translations were closer to the AI-generated drafts at both the textual and semantic levels, indicating a pronounced priming effect in post-editing. Meanwhile, revisions to AI-generated drafts mainly took the form of local textual refinement, whereas deeper interventions involving substantial rewriting and structural reorganization were relatively limited. This suggests that AI-generated drafts not only exert a sustained influence on post-edited output, but also constrain the exercise of translator subjectivity in textual rewriting and structural reorganization. These findings help to deepen our understanding of translation production under generative AI conditions and provide empirical support for workflow optimization in professional translation practice, the judicious use of AI-generated drafts, and the repositioning of translators’ roles.
文章引用:毛尚勇, 黄津湘, 王婧睿. 生成式AI译文初稿的启动效应与译者主体性受限——基于旅游宣传文本翻译的实证研究[J]. 现代语言学, 2026, 14(4): 484-492. https://doi.org/10.12677/ml.2026.144318

参考文献

[1] 胡开宝, 李晓倩. 大语言模型背景下翻译研究的发展: 问题与前景[J]. 中国翻译, 2023, 44(6): 64-73, 192.
[2] 于蕾. ChatGPT翻译的词汇多样性和句法复杂度研究[J]. 外语教学与研究, 2024, 56(2): 297-307, 321.
[3] 左百瑶. 生成式人工智能与国内语言学研究: 基于Grok3的CSSCI发文分析[J]. 外语学刊, 2026(1): 31-39.
[4] 杨艳霞, 陈莹, 魏向清. 生成式智能时代的人机协同翻译素养研究[J]. 上海翻译, 2025(1): 39-45.
[5] 李德俊. 生成式人工智能时代对译者任务的再思考[J]. 外语学刊, 2026(1): 7-14.
[6] 王律, 王湘玲. ChatGPT时代机器翻译译后编辑能力培养模式研究[J]. 外语电化教学, 2023(4): 16-23, 115.
[7] 李奉栖, 张云, 丁丽杰. 大语言模型与神经网络机器翻译系统专业文本翻译质量对比——以法律汉英翻译为例[J]. 上海翻译, 2025(6): 62-67.
[8] 贾艳芳, 孙三军. 机器翻译译后编辑难度测量体系构建研究[J]. 中国外语, 2022, 19(3): 16-24.
[9] 王湘玲, 陈广姣, 周祥艳. 国际机器翻译译后编辑认知研究路线图(2011-2021) [J]. 外国语(上海外国语大学学报), 2023, 46(5): 90-100.
[10] 余静, 刘康龙. 重塑翻译研究: AI技术影响下的范式转换与未来方向探索[J]. 外国语(上海外国语大学学报), 2024, 47(4): 72-81.
[11] 王少爽, 陈媛媛. 技术社会学视域下生成式人工智能对翻译行业发展的影响探析[J]. 上海翻译, 2025(4): 59-65.
[12] 刘云虹. 人工智能时代翻译批评研究的核心问题[J]. 外国语(上海外国语大学学报), 2025, 48(6): 107-114.
[13] 赵苓岑. 智能时代翻译的伦理危机与反思[J]. 外国语(上海外国语大学学报), 2025, 48(6): 115-121.
[14] 王均松, 张政. 人机共生视域下的人智协同翻译模式构建[J]. 上海翻译(中英文), 2026(1): 13-19, 95.
[15] 陈刚. 跨文化意识——导游词译者之必备——兼评《走遍中国》英译本[J]. 中国翻译, 2002(2): 37-40.
[16] 洪明. 论接受美学与旅游外宣广告翻译中的读者关照[J]. 外语与外语教学, 2006(8): 56-59, 65.
[17] International Organization for Standardization (2015) ISO 17100:2015 Translation Services—Requirements for Translation Services. International Organization for Standardization.
[18] Popović, M. (2017) chrF++: Words Helping Character N-Grams. Proceedings of the Second Conference on Machine Translation, Copenhagen, 7-8 September 2017, 612-618. [Google Scholar] [CrossRef
[19] Zhang, T., Kishore, V., Wu, F., Weinberger, K.Q. and Artzi, Y. (2020) BERTScore: Evaluating Text Generation with BERT. 8th International Conference on Learning Representations, Addis Ababa, 26-30 April 2020.
[20] Snover, M., Dorr, B., Schwartz, R., Micciulla, L. and Makhoul, J. (2006) A Study of Translation Edit Rate with Targeted Human Annotation. Proceedings of the 7th Conference of the Association for Machine Translation in the Americas: Technical Papers, Association for Machine Translation in the Americas, Cambridge, 8-12 August 2006, 223-231.
[21] 王晓庄, 白学军. 判断与决策中的锚定效应[J]. 心理科学进展, 2009, 17(1): 37-43.
[22] 张文煜, 赵璧. 生成式人工智能开创机器翻译的新纪元了吗?——一项质量对比研究及对翻译教育的思考[J]. 北京第二外国语学院学报, 2024, 46(1): 83-98.
[23] 文旭, 田亚灵. ChatGPT应用于中国特色话语翻译的有效性研究[J]. 上海翻译, 2024(2): 27-34, 94-95.
[24] 孙彦, 李纾, 殷晓莉. 决策与推理的双系统——启发式系统和分析系统[J]. 心理科学进展, 2007(5): 721-726.