Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Confusion about the time cost of GCG and AutoDAN #14

Open
allblueJT opened this issue Apr 10, 2024 · 0 comments
Open

Confusion about the time cost of GCG and AutoDAN #14

allblueJT opened this issue Apr 10, 2024 · 0 comments

Comments

@allblueJT
Copy link

Hi!

According to the paper of AutoDAN, the time cost of each iter of AutoDAN is less than GCG. But as shown in your paper of EasyJailbreak, AutoDAN cost much higher time than GCG. I'm wandering what made the difference.

Does AutoDAN require more epochs to converge than GCG? Or in your exp settings, AutoDAN is executed on 520 samples(for example) to optimize the suffix while GCG just uses 25 samples(as mentioned in the origin paper)?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant