Source code for paper "Don't Forget Your Reward Values: Language Model Alignment via Value-based Calibration"
We will release the code soon!
Source code for paper "Don't Forget Your Reward Values: Language Model Alignment via Value-based Calibration"
License
MaoXinn/VCB
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
About
Source code for paper "Don't Forget Your Reward Values: Language Model Alignment via Value-based Calibration"
Resources
License
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published