MQM Human Evaluation
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MQM Human Evaluation Japanese→English
11.6MQM Score
3
MQM Human Evaluation English→Serbian
15.8MQM Score
3
MQM Human Evaluation English→Chinese
6.3MQM Score
3
MQM Human Evaluation Czech→Ukrainian
5.3MQM Score
3
MQM Human Evaluation English→Swahili
4.2MQM Score
3
MQM Human Evaluation English→Korean
3.1MQM Score
3
MQM Human Evaluation English→German
2.2MQM Score
3
MQM Human Evaluation English→Italian
1.8MQM Score
3