Data for the paper Online Versus Offline NMT quality: An In-depth Analysis on English–German and German-English published in COLING 2020.
- Segment-level annotations: data/deen.tsv
- Token-level annotations: data/deen_tokens.tsv
- Segment-level annotations: data/ende.tsv
- Token-level annotations: data/ende_tokens.tsv
- Guidelines for error annotation
- MQM decision tree
If you use this dataset, please cite:
@inproceedings{elbayad2020online,
title={Online Versus Offline NMT Quality: An In-depth Analysis on English-German and German-English},
author={Elbayad, Maha and Ustaszewski, Michael and Esperan{\c{c}}a-Rodier, Emmanuelle and Manquat, Francis Brunet and Besacier, Laurent},
journal={COLING},
year={2020}
}