Stanford has released the second part of its CS329A "Self-Improving AI Agents" lecture, a deep dive into test-time compute scaling led by instructor Azalia Mirhoseini that walks through the techniques increasingly used to boost model performance without touching the underlying weights. The video is available through Stanford Online's channel as part of the course playlist, and covers repeated sampling, inference-time scaling laws, reward models, and the tradeoffs between parallel and sequential approaches to reasoning.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.