Haoqin Tu
Results
12
comments of
Haoqin Tu
From the view of model structure, I think yes. But there are differences between the training paradigm between Bert and Bart decoder (Bert has next sentence prediction and masked token...
Thanks for your interests in our work! Seems that there are issues in dependency using the LLaVA model. In our paper, we employed the initial version of LLaVA, you can...