Abstract:

The rise of large language models (LLMs) has brought the knowledge-transforming function of translation technology into sharp relief, yet the relationship between translation technology and knowledge transformation remains undertheorized. From the perspective of Transknowletology, this study first redefines translation technology as a technical system centered on knowledge transformation, based on a review of previous definitions, and uncovers its triple attributes of locality, historicity, and mobility. Drawing on the DIKW framework, it then analyzes the hierarchical structure of translation technology and explicates its functions in knowledge transformation. Second, focusing on LLM-based translation technology, the study examines its implementation mechanism across four stages: input processing, contextual understanding, target-text generation, and quality assessment. Third, it reflects on the ethical challenges of the current knowledge transformation mechanism in terms of the fairness of knowledge resources, equality in mutual interpretation among knowledge systems, and the sustainability of knowledge dissemination. Finally, it envisions future directions for translation technology in terms of theoretical innovation, system renovation, application expansion, and educational upgrading. This study offers a Transknowletological perspective for translation technology studies and provides a theoretical reference for the development of translation technology in the era of generative AI.

Key words: Transknowletology, translation technology, DIKW, Large Language Models, knowledge transformation