What Should SFT Learn? Conditional Information Gain for Generalizable Post-Training

arXiv:2610.12995 · cs.CL, cs.AI, cs.LG · Submitted 2026-10-09 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2026-10-09

Updated: 2026-10-09

Terminology

Related papers