AIBullisharXiv – CS AI · 10h ago7/10
🧠
Memorize Theorems, Not Instances: Probing SFT Generalization through Mathematical Reasoning
Researchers propose Theorem-SFT, a novel supervised fine-tuning approach that teaches language models to apply mathematical rules explicitly rather than memorize surface-level correlations between problems and solutions. The method demonstrates significant performance improvements across benchmarks while revealing that feed-forward layers, not memorization itself, are the primary locus of reasoning capability.