| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | eng-zho | generaltest2022 | show (logfile) | 6.8 |
| 1 | eng-zho | newstest2017 | show (logfile) | 1.3 |
| 2 | eng-zho | newstest2018 | show (logfile) | 1.1 |
| 3 | eng-zho | newstest2019 | show (logfile) | 5.4 |
| 4 | eng-zho | newstest2020 | show (logfile) | 1.8 |
| 5 | eng-zho | newstest2021 | show (logfile) | 2.3 |
| 6 | eng-zho | newstestALL2020 | show (logfile) | 2 |
| 7 | eng-zho | newstestB2020 | show (logfile) | 0.3 |
| 8 | eng-zho | tatoeba-test-v2020-07-28 | show (logfile) | 0.2 |
| 9 | eng-zho | tatoeba-test-v2021-03-30 | show (logfile) | 0.2 |
| 10 | eng-zho | tatoeba-test-v2021-08-07 | show (logfile) | 0.2 |
| 11 | eng-zho | tico19-test | show (logfile) | 8.7 |
| average | 2.525 |