Some questions about the paper

[yuanrr]

Your paper has inspired me a lot! After reading the paper, I have some questions and hope you can reply to me!

1. Will the situation observed in Fig.2 appear in some models that use Qformer?
2. Can this approach be used in training as a means of reducing GPU memory?

Thanks a lot!