This study analyzes the near-block-diagonal structure of neural network Hessians, identifying static and dynamic forces influencing it, and providing insights into large language models' Hessian characteristics.

https://arxiv.org/abs//2505.02809

YouTube: https://www.youtube.com/@ArxivPapers

TikTok: https://www.tiktok.com/@arxiv_papers

Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016

Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers