Result in the CWMT2018 QE dataset

[chencong-jxnu]

I trained the TransQuest model with WMT2020 QE dataset, and the test result in en-zh task2 is 0.5999. I want to know the performance of the TransQuest model on CWMT2018(China Workshop on Machine Translation, 2018) QE dataset. So,I trained the TransQuest model with CWMT2018 QE dataset. The test result of en-zh and zh-en is 0.516 and 0.568 respectively. I am not sure if there is a problem with my training method. So, I was wondering if you could provide a test result in the CWMT2018 QE dataset.
Thank you for your help.