Tense inconsistency frequently occurs in machine translation. However, there are few criteria to assess the model's mastery of tense prediction from a linguistic perspective. In this paper, we present a parallel tense test set, containing French-English 552 utterances. We also introduce a corresponding benchmark, tense prediction accuracy. With the tense test set and the benchmark, researchers are able to measure the tense consistency performance of machine translation systems for the first time.
@article{arxiv.2305.13740,
title = {TeCS: A Dataset and Benchmark for Tense Consistency of Machine Translation},
author = {Yiming Ai and Zhiwei He and Kai Yu and Rui Wang},
journal= {arXiv preprint arXiv:2305.13740},
year = {2023}
}