Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Evolving to Match Capabilities: Improving Harness Self-Evolution via Skill Adaptation
Zhihao Yang, Haojin Wang, Juncheng Wu +7
Harness self-evolution offers a practical route for LLM agents to improve from execution experience without updating model parameters. Its effectiveness depends on whether models c…
cs.LG2026
Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning
Dylan Zhang, Yufeng Xu, Haojin Wang +2
Post-training of reasoning LLMs is a holistic process that typically consists of an offline SFT stage followed by an online reinforcement learning (RL) stage. However, SFT is often…