Stanford CME295 Transformers and LLMs | Autumn 2025 | Lecture 6 - LLM Reasoning

Stanford Online: University Courses & Lectures · Oct 17, 2026 · 1h 53m

For more information about Stanford’s graduate programs, visit: https://online.stanford.edu/graduate-educationNovember 7, 2025This lecture covers: • Reasoning models• RL for reasoning• GRPO• ScalingTo follow along with the course schedule and syllabus, visit: https://cme295.stanford.edu/syllabus/Chapters:00:00:00 Introduction00:12:43 Reasoning models00:27:49 Benchmarks00:32:04 Pass@k metric00:48:07 Scaling with RL00:57:44 GRPO01:06:03 Comparison between GRPO and PPO01:16:14 Length bias01:25:00 DAPO, Dr. GRPO01:29:38 DeepSeek R1 recipeUploaded 14-11-25Afshine Amidi is an Adjunct Lecturer at…