Yuntao Du

Hi, thanks for your great work. I want to ask could this method apply to open VQA task where an open answer is needed instead of choosing from a given...

Prompt for generation

1

Hi, thanks for your great work. Could you share the text prompt for image generation? Or share some examples especially for Texture-based and Material-based?

Dataset

11

Hi, would you please a downloaded dataset for training? Downloading from Arvix is very unstable and often forbidden.

Traing code

1

Hi, good job! When would you like to release the training code?

HI, how is the experiment in Table 4 conducted? What if we have a large dataset e.g. 1000 VQA samples, how could we conduct the experiment?

Traing code

1

Hi, is there any plan to release the training code?

Losses

1

Hi, Great job. I have a question, what does this use for in losses.py of line 37-38 pos_loss = (torch.pow(1 - torch.exp(-pos_loss), 2) * pos_loss).sum(dim=1).mean() neg_loss = (torch.pow(1 - torch.exp(-neg_loss),...