Questions about results in Table 2

[qinliuliuqin]

Thanks for the great work!
In Table 2, the training set should be PASCAL-10K instead of PASCAL because the model achieving 93.2% IOU in Table 2 is actually the model 'IOG-Resnet101' in Table 1 and 3. Am I correct?

Thank you!
Qin