Clarification on ImageNet-1K Experiments and Validation Protocol

[sunhyeok-lee]

Hi,

Thank you for your great work on MinimaxDiffusion! I have a question regarding the ImageNet-1K experiments mentioned in your repo.

You refer to using a "better validation protocol provided in RDED," but the exact details on how this was implemented are unclear. Could you clarify:

1. How was the validation protocol from RDED applied in your ImageNet-1K experiments?

2. Were there any specific settings (e.g., preprocessing, hyperparameters) or evaluation details unique to these experiments?

Any additional details or resources would be greatly appreciated.

Thank you!

[vimar-gu]

Thank you for your interest in our paper and sorry for the late reply.

We used the validation protocol of RDED exactly the way they used it. No specific settings are adopted other than their original setting.