Grammar Accuracy Evaluation (GAE): Quantifiable Quantitative Evaluation of Machine Translation Models
This addresses the issue of score deviation in human evaluations for machine translation, offering a more objective method, though it appears incremental as it builds on existing evaluation frameworks.
The paper tackles the problem of subjective human evaluation in Natural Language Generation by proposing Grammar Accuracy Evaluation (GAE) to provide specific criteria, showing that GAE compensates for BLEU's shortcomings in evaluating machine translation models.
Natural Language Generation (NLG) refers to the operation of expressing the calculation results of a system in human language. Since the quality of generated sentences from an NLG model cannot be fully represented using only quantitative evaluation, they are evaluated using qualitative evaluation by humans in which the meaning or grammar of a sentence is scored according to a subjective criterion. Nevertheless, the existing evaluation methods have a problem as a large score deviation occurs depending on the criteria of evaluators. In this paper, we propose Grammar Accuracy Evaluation (GAE) that can provide the specific evaluating criteria. As a result of analyzing the quality of machine translation by BLEU and GAE, it was confirmed that the BLEU score does not represent the absolute performance of machine translation models and GAE compensates for the shortcomings of BLEU with flexible evaluation of alternative synonyms and changes in sentence structure.