Skip to content

Latest commit

 

History

History
18 lines (15 loc) · 898 Bytes

README.md

File metadata and controls

18 lines (15 loc) · 898 Bytes

HJQE: Human judgement on the word-level quality estimation for the machine translation

Data

HJQE is a benchmark dataset for word-level quality estimation (QE) of machine translation, where all examples are annotated by expert translators. The goal of the dataset is to measure the translation errors from the human judgement.

HJQE contains the corpus for two translation directions: English-German and English-Chinese. For each corpus, the source and mt sentences are same to WMT20 (https://www.statmt.org/wmt20/quality-estimation-task.html).

Citation

Please cite the following paper if you found the resources in this repository useful.

@article{yang2021HJQE,
  title={Rethink about the Word-level Quality Estimation for Machine Translation from Human Judgement},
  author={Yang Zhen, Meng Fandong, Yuanmeng Yan, and Zhou, Jie},
  year={2022}
}