google-research / l2p

Learning to Prompt (L2P) for Continual Learning @ CVPR22 and DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning @ ECCV22
https://arxiv.org/pdf/2112.08654.pdf
Apache License 2.0
416 stars 42 forks source link

question on the G(eneral)-Prompt learning #34

Open liudakai2 opened 1 year ago

liudakai2 commented 1 year ago

Thanks for your impressive work. There is still one thing that I can't figure out. According to the setting of continual learning, when learning on the current task, the data from previous tasks is unavailable. My question is, how do you make sure that the G(eneral)-Prompt learned on the current task also works for previous tasks? How do you deal with the forgetting problem of the G(eneral)-Prompt itself?

lyjgo commented 1 year ago

I have the same question when I read the paper