AI RESEARCH
InfoSFT: Learn More and Forget Less with Information-Aware Token Weighting
arXiv CS.LG
•
ArXi:2605.14967v1 Announce Type: new Supervised fine-tuning (SFT) provides the standard approach for teaching LLMs new behaviors from offline expert nstrations. However, standard SFT uniformly fits all samples -- including those with low likelihood under the base model -- which can disproportionately drive