single mention · post-training
Three-stage post-training with SFT, RL, and multi-teacher on-policy distillation
Initializes baseline agent capabilities with supervised fine-tuning, develops domain experts with reinforcement learning, and consolidates them through multi-teacher on-policy distillation.
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 three-stage post-training paradigm.
Filed under post-training