mirror of
https://github.com/facebookresearch/ReAgent.git
synced 2026-06-16 12:44:41 +00:00
Summary: Compress Seq2Reward Model through supervised learning another model Reviewed By: czxttkl Differential Revision: D22826516 fbshipit-source-id: b20bd0582ef895aef228be827889511580e8c84e