章节

422 个页面

Posts

OPD (On-Policy Distillation)

LLM 算法岗面试手册 2026

Training extremely large neural networks across thousands of GPUs

MOPD：Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training

从 GPT-2 到 Kimi K3：大模型架构如何演进

Review：自进化Agent的RL Memory与Skill方法批判

verl V1 主线重构 · 给后训练工作者的三问

DeepSeek V4 Infra Deep Dive 4.4: PyTorch 训练显存管理

RL 面试题 2026

The Ultra-Scale Playbook 学习笔记