Sumanth Dathathri

Oops. The current draft seems to be missing a link to where we got the wordlists from: https://www.enchantedlearning.com/wordlist/. Will add this back into the paper! Thanks for catching this. Aside:...

Can you elaborate what "gpt2 pplm runs on the bag of words" means? I don't know if I fully follow what is causing the problem. I haven't seen this error...

Thanks for the question! We have not run a detailed analysis on the inference speed, but it is slower than normal inference because of the gradient based updates to the...

Found the fix! Sorry

Oops, actually it isn't fixed. The outputs are still inconsistent when I load a saved torch model first and then do the transfer. Otherwise it is consistent... strange. But, also,...

Oops, sorry. It wasn't commented out and I do add the softmax outside, just copy pasted a snipped here -- I later commented out the batchnorms before copy pasting because...

This paper is not under review at CVPR. Could you please paste your error messages? Happy to help debug.

I am happy to help debug the errors privately, and follow up with the specific version of python/pytorch/cleverhans needed for this to run. But besides that, we do not have...