新文科视域下口译教学自动测评系统建构的可解释研究

Interpretability Research on the Construction of an Automatic Evaluation System for Interpreting Teaching from the Perspective of New Liberal Arts

随着全球化进程的不断推进,口译教学在新文科视域下扮演着越来越重要的角色。为了提高口译教学的效率和质量,自动测评系统被引入口译教学之中。然而,目前大多数口译教学自动测评系统在其工作原理和结果的解释方面出现了诸如模型的黑盒化、评估结果的不透明及评分标准的不明确等问题。该研究以新文科翻译学视角,通过对口译任务的要素进行分析,设计了一种可解释的口译教学自动测评系统,并基于人工智能的可解释理论分析,提出了相应的可持续性优化路径,以提高系统的透明度、可信度和准确度。

With the continuous advancement of globalization,interpretation teaching is playing an increasingly important role in the field of new humanities.In order to improve the efficiency and quality of interpreting teaching,an automatic evaluation system has been introduced into interpreting teaching.However,currently most automated evaluation systems have encountered issues such as black box modeling,opaque evaluation results,and unclear scoring criteria in their working principles and interpretation of results.This study,from the perspective of new liberal arts translation studies,analyzes the elements of interpreting tasks and designs an interpretable automatic evaluation system for interpreting teaching.Based on the interpretable theory of artificial intelligence,corresponding optimization strategies are proposed to improve the transparency,credibility,and accuracy of the system.