Home
avatar

ychhh

2026 2篇文章

10-13 SmolLM2: When Smol Goes Big — Data-Centric Training of a Small Language Model #大模型 #数据 #多阶段训练 #论文阅读 10-13 RegMix: Data Mixture as Regression for Language Model Pre-training #大模型 #数据配比 #论文阅读