Standardization of Translation of Green Leather Processing Terminology: Taking Chrome-Free Tanning and Bio-Based Tanning Agents as Examples

Accurate cross-linguistic representation of technical terminology is essential for knowledge sharing and international collaboration in emerging sustainable manufacturing technologies. This study addresses semantic ambiguity and conceptual inconsistency in the translation of green leather processing terminology by focusing on chrome-free tanning and bio-based tanning agents. A Domain-Adaptive Semantic Mapping (DASM) framework is proposed to integrate terminology theory with deep semantic learning. The method constructs a structured bilingual corpus and employs an improved XLM-RoBERTa-based semantic component parser to decompose terms into core concept, process attribute, and ecological attribute subspaces for fine-grained cross-language alignment. A semantic density–driven decision strategy further optimizes standardized translation selection by balancing conceptual equivalence and system consistency. Experimental evaluation demonstrates translation accuracy of 99.1% for Level 1 bio-based tanning terminology, terminology network modularity of 0.72, and a relation misalignment rate of only 3.2%, while maintaining robust performance for structurally complex concepts. Beyond supporting standardized terminology management in leather science, the proposed framework provides a transferable methodology for multilingual technical communication, semantic knowledge organization, and international standardization in engineering disciplines where precise information representation and interoperable terminology are critical, including digital communication and intelligent engineering systems.