Zhangzhe's Blog

03-17

Transformers without Normalization

03-14

2025.02 DeepSeek 开源周第五弹 —— 3FS

03-12

2025.02 DeepSeek 开源周第四弹 —— DualPipe ＆ EPLB

03-10

2025.02 DeepSeek 开源周第三弹 —— DeepGEMM

03-06

2025.02 DeepSeek 开源周第二弹 —— DeepEP

03-06

2025.02 DeepSeek 开源周第一弹 —— FlashMLA

02-20

Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention

02-18

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

02-02

DeepSeek-V3 Technical Report

01-20

Denoising Diffusion Probabilistic Models