Implement Tunix-based DPO/ORPO integration.#3668

Open

igorts-git wants to merge 3 commits into

mainfrom

igorts/dpo-feature-integration

Commits on May 11, 2026

Refactor: extract shared post-training hooks and update SFT implementation
igorts-git
committed
Feature: integrate post-training Direct Preference Optimization (DPO)
igorts-git
committed
Integrate standalone headless DPO and ORPO notebooks, and automated CI validation loop
igorts-git
committed