Post-Training Techniques: On-Policy Distillation | Afshine & Shervine Amidi

Browse Stanford's online computational and mathematical engineering graduate courses: https://stanford.io/grad-cme

Enrollment for the winter quarter is open now!
https://stanford.io/Winter-2026

Is making a model bigger still the only way to improve it? Adjunct Professors Afshine Amidi and Shervine Amidi explain how post-training techniques such as finetuning, reinforcement learning, preference optimization, verifier-guided training, and distillation help a model reason better, follow instructions better, solve harder problems, and adapt to specialized domains.

They then focus on on-policy distillation. In standard distillation, a smaller student model learns from examples produced by a stronger teacher. In on-policy distillation, the student generates its own responses during training and learns from how the teacher evaluates those exact attempts. According to the Amidi brothers, this aligns the learning process with how the model behaves in practice, which matters most for reasoning tasks, where intermediate steps, mistakes, and corrections shape the result.

CME295 is a Stanford graduate course that distills the latest advances in LLMs, explains the core ideas from first principles, and connects them to practical systems. Learn more about this course: https://stanford.io/cme295

About the speakers:
Afshine Amidi and Shervine Amidi are twin brothers and Adjunct Professors at Stanford University. They teach CME295: Transformers & Large Language Models and CME296: Diffusion & Large Vision Models.

#PostTraining #LLMs #StanfordOnline Receive SMS online on sms24.me

TubeReader video aggregator is a website that collects and organizes online videos from the YouTube source. Video aggregation is done for different purposes, and TubeReader take different approaches to achieve their purpose.

Our try to collect videos of high quality or interest for visitors to view; the collection may be made by editors or may be based on community votes.

Another method is to base the collection on those videos most viewed, either at the aggregator site or at various popular video hosting sites.

TubeReader site exists to allow users to collect their own sets of videos, for personal use as well as for browsing and viewing by others; TubeReader can develop online communities around video sharing.

Our site allow users to create a personalized video playlist, for personal use as well as for browsing and viewing by others.

@YouTubeReaderBot allows you to subscribe to Youtube channels.

By using @YouTubeReaderBot Bot you agree with YouTube Terms of Service.

Use the @YouTubeReaderBot telegram bot to be the first to be notified when new videos are released on your favorite channels.

Look for new videos or channels and share them with your friends.

You can start using our bot from this video, subscribe now to Post-Training Techniques: On-Policy Distillation | Afshine & Shervine Amidi

What is YouTube?

YouTube is a free video sharing website that makes it easy to watch online videos. You can even create and upload your own videos to share with others. Originally created in 2005, YouTube is now one of the most popular sites on the Web, with visitors watching around 6 billion hours of video every month.