{"id":"2602.12222","title":"Towards On-Policy SFT: Distribution Discriminant Theory and its Applications in LLM Training","url":"https://codeofpaper.com/paper/2602.12222","arxiv_url":"https://arxiv.org/abs/2602.12222","published":"2026-02-12","code":"official","best":{"full_name":"zhangmiaosen2000/Towards-On-Policy-SFT","url":"https://github.com/zhangmiaosen2000/Towards-On-Policy-SFT","tier":"official","stars":27,"license":null,"framework":null,"language":"Python","last_commit":"2026-03-26T05:35:50"},"most_starred_community":null,"implementations":[{"full_name":"zhangmiaosen2000/Towards-On-Policy-SFT","url":"https://github.com/zhangmiaosen2000/Towards-On-Policy-SFT","tier":"official","stars":27,"license":null,"framework":null,"language":"Python","last_commit":"2026-03-26T05:35:50"}],"possible_matches":0}