CLJul 29, 2019
Machine Translation Evaluation with BERT Regressor
arXiv:1907.12679v126 citations
Originality Synthesis-oriented
AI Analysis
This addresses the need for accurate evaluation metrics in machine translation, but it is incremental as it applies an existing method (BERT) to a new task.
The paper tackled the problem of automatic machine translation evaluation by introducing a metric using BERT, achieving state-of-the-art performance in segment-level metrics for all to-English language pairs on the WMT-2017 dataset.
We introduce the metric using BERT (Bidirectional Encoder Representations from Transformers) (Devlin et al., 2019) for automatic machine translation evaluation. The experimental results of the WMT-2017 Metrics Shared Task dataset show that our metric achieves state-of-the-art performance in segment-level metrics task for all to-English language pairs.