Open Models Almanac
Methodspost-training

single mention · post-training

Joint supervised fine-tuning and reinforcement learning

Combines supervised fine-tuning and reinforcement learning, without evidence specifying their sequencing or a more detailed algorithm.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See post-training for the methods filed beside it.

Also called Joint SFT and RL.

Filed under post-training