Machine Learning Algorithms & Natural Language Processing

Apr 4, 2026 · Artificial Intelligence

Why the Best SFT Checkpoint May Hurt RL Performance: Adaptive Early‑Stop Loss (AESL) for LLM Cold‑Start

The paper reveals that over‑optimizing supervised fine‑tuning (SFT) for large language models can diminish their reinforcement‑learning (RL) potential, proposes an Adaptive Early‑Stop Loss (AESL) that balances accuracy and output diversity during cold‑start, and demonstrates across multiple LLMs that AESL consistently yields superior RL results.