What Should SFT Learn? Conditional Information Gain for Generalizable Post-Training
Supervised fine-tuning (SFT) remains a widely used approach for post-training large language models, because fixed demonstrations can directly introduce desired reasoning strategies and behavioral patterns. Yet forcing a model to imitate externally provided trajectories may also interfere with behaviors already support...
Yuxuan Gu, Ke-Quan Niu, Xiao-Cheng Feng et al.
· 0 citations