post-training 5
- Reinforcement Learning from Human Feedback: Mechanics of PPO, DPO, and Reward Modeling
- Llama 3.1 Architecture: Dense Scaling to 405B and 128K Context
- Frontier LLM Research Trends: Test-Time Compute, State-Space Models, and MoE
- Self-Evolving Language Models: The Mechanics of Autonomous Post-Training
- LLaMA-Factory Architecture: Unified Post-Training Across 100+ LLMs