arXivSenmiao Wang, Tiantian Fang, Haoran Zhang, Yushun Zhang, Kunxiang Zhao, Alex Schwing, Ruoyu SunThu, Jun 4, 2026, 10:55 AM PDT
score 17.2
Polynomial layer stabilizes weight training in large language models
Original: PC Layer: Polynomial Weight Preconditioning for Improving LLM Pre-Training
Source: arxiv.org ↗
Writing ELI5 summary…