AI RESEARCH

Stabilizing LLM Supervised Fine-Tuning via Explicit Distributional Control

arXiv CS.AI

ArXi:2605.04468v1 Announce Type: cross Post-