Better aligned models obtained by weak-to-strong model extrapolation (ExPO)
-
Weak-to-Strong Extrapolation Expedites Alignment
Paper • 2404.16792 • Published • 2 -
chujiezheng/Smaug-Llama-3-70B-Instruct-ExPO
Text Generation • Updated • 196 • 2 -
chujiezheng/Llama-3-Instruct-8B-SimPO-ExPO
Text Generation • Updated • 97 • 5 -
chujiezheng/LLaMA3-iterative-DPO-final-ExPO
Text Generation • Updated • 204