Zhu et al. (2025c) proved that Reinforcement Learning updates occur in low-curvature subspaces orthogonal to the principal components updated by Supervised Fine-Tuning (SFT), suggesting that Reinforcement Learning operates in a distinct optimization regime that fine-tunes behavior without significantly altering primary feature representations.

Authors

• Person: Not available Organization: arXiv
  A Survey on the Theory and Mechanism of Large Language Models

Sources

• A Survey on the Theory and Mechanism of Large Language Models arxiv.org arXiv via serper

Referenced by nodes (2)

• reinforcement learning concept
• supervised fine-tuning concept