New dataset (Explainable Quality Estimation)

[plkumjorn]

Hi, thank you for creating this helpful website. I'd like to add a new dataset concerning explanations for quality estimation of machine translation, as detailed below.

• Explanation type: highlights
• Dataset name: X-QE
• Link to the associated publication/preprint: https://aclanthology.org/2021.eval4nlp-1.17/
• Associated task: Quality Estimation of Machine Translation
• Granurality restriction (for highlights): words
• Collection method: experts
• Number of instances: 4590
• Number of explanations per instance: 1
• Total number of annotators: 14
• Language: 4 Language pairs -- Estonian-English (Et-En), Romanian-English (Ro-En), German-Chinese (De-Zh), Russian-German (Ru-De)

[sarahwie]

Thanks for your contribution, and I was excited to see this shared task at EMNLP! Added to table.