认知翻译框架下DeepSeek提示语优化研究与评测

——以高校校园新闻翻译为考察中心

打开文本图片集

中图分类号：H315.9；H33 文献标志码：A 文章编号：1672-1101（2025）04-0057-08

Research and Evaluation on DeepSeek Prompt Optimization under the Cognitive Translation Framework -Take the Translation of Campus News in Colleges and Universities as the Research Focus

LIU Xiaoyu，FANG Fang （SchoolofForeignLanguages，HefeiUniversity，Hefei23o6O1，China）

Abstract： With the continuous deepening of international educational exchanges，campus news translation is playing an increasingly important role in shaping the international image of colleges and universities. Integrating the theory of cognitive language knowledge interpretation into the DeepSeek translation process， a prompt framework suitable for campus news translation in coleges and universities is constructed around three phases： the translating proposition phase， the integrating proposition phase， and the polishing proposition phase. It is thus proposed that，（1） in the translating proposition phase，four key dimensions，namely specificity， scope/background， perspective and prominence，should be comprehensively considered to enable the translation to accurately convey the news content，conform to the cognitive mode of the target readers and enhance the acceptability of the information.（2）In the integrating proposition phase，the textual logic of the translation should be optimized to meet the requirements of news style.（3） In the polishing proposition phase， prompt phrases should be designed based on the principles of external publicity and the characteristics of news language to enhance the international dissmination power and cross-cultural communication effect of campus news.Based on the construction of the above-mentioned DeepSeek prompt framework， a case analysis of campus news translation in colleges and universities was conducted，and the optimization effect was evaluated using the machine translation evaluation systems BLEU and TER. The analysis and evaluation results show that the optimized prompts can effectively improve the translation quality of DeepSeek， enabling it to beter meet the reading habits and cognitive expectations of the target readers， which can provide useful references for the later development of artificial intelligence translation tools.

Key words： cognitive translation; DeepSeek; prompt; campus news; Chinese-German translation

全球化背景下，国内高校校园新闻翻译在塑造自身形象和促进跨文化交流方面发挥着重要作用。（剩余12704字）