Chu et al. (2025) provided empirical evidence that Supervised Fine-Tuning (SFT) tends to memorize training data, leading to poor performance on out-of-distribution (OOD) tasks, whereas Reinforcement Learning (RL) demonstrates superior generalization capabilities.

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 (5)

• reinforcement learning concept
• training data concept
• generalization concept
• supervised fine-tuning concept
• Out-of-Distribution concept