Cite
Notes
Only stored in your browser.
Attribution
Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models
arXiv 2024
Efficient Adversarial Training in LLMs with Continuous Attacks
from 2 papers
Aaron Courville
Alessandro Sordoni
Arian Hosseini
Gauthier Gidel
Leo Schwinn
Michael Noukhovitch
Rishabh Agarwal
Shengyi "Costa" Huang
researcher
Stephan Günnemann