Automated Multi-dimensional English Translation Evaluation via LLMs: From Tool Prototyping to Classroom Integration

This study presents an Intelligent Translation Evaluation Tool (ITET) that leverages large language models (LLMs) to enhance translation quality assessment, optimize model architecture, and support personalized learning pathways. Classroom implementation shows that combining multidimensional assessment with the Spark Cognitive LLM improves translation proficiency by 77.0%, autonomous learning by 72.0%, and critical thinking by 74.0%. Innovations like dynamic competency radar charts and adaptive learning paths reduce feedback latency and enhance instructional adaptability compared to traditional tools. These findings offer interpreting educators evidence-based strategies for leveraging LLMs to optimize specialized language training, demonstrating measurable gains in both translational accuracy and learner agency. The system’s scalable design supports broader adoption in AI-enhanced translation pedagogy.