Subject: Request for vocab.3000 file and Complete Dataset for Attack Experiments in multinli.py

[jinyilun718]

Hello,

I hope this message finds you well. I am currently working with your codebase and I have a question regarding the vocab.3000 file referenced in the multinli.py script within the attack directory. I was unable to find this vocab.3000 file in the Hugging Face datasets library, and it seems not to be included in the resources of the repository. Could you please guide me on how to obtain this file? Additionally, I would appreciate if you could provide information on the complete set of data required for the attack experiments. The guidance on how to gather or generate all necessary files would be greatly beneficial for my work. Thank you very much for your time and help. Looking forward to your response.

[HanGuo97]

Thanks for reaching out! Please check out a related question at https://github.com/HanGuo97/soft-Q-learning-for-text-generation/issues/2. Somewhat related, we also have a follow-up work with better documentation.