RRHF
RRHF
copied to clipboard
Published
20 hours ago
•
GanjinZero
Reame
Issues
Wombat与RRHF
Open
Guochry
opened this issue 11 months ago
• 4 comments
想请问一下,有比较不同reward来源的实验吗?即比较“开源reward model”和“ChatGPT”分别作为reward score来源的效果孰优孰劣吗?
Aug 02 '23 13:08
Guochry