This project implements a self-improving language model evaluator using synthetic data, inspired by the paper "Self-Taught Evaluators". The model iteratively generates, evaluates, and fine-tunes itself using its own synthetic data, eliminating the need for costly human annotations.
-
Notifications
You must be signed in to change notification settings - Fork 1
sanowl/Self-Taught-Evaluator
About
this is based on the paper Self-Taught Evaluators
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published