AI & machine learning2024 · shared on Hacker News
Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking
Explore Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.
GO BEYOND THE ABSTRACT
The conversation around the idea.
Original Hacker News threads. Different perspectives, questions, and things worth thinking about.
Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.
Links found in the discussion
Up to 50 recent source links, in HN item order.
Yfofoz ·
Learning to Reason with LLMs
1,654 points1,261 comments
Ygfortaine ·
g1: Using Llama-3.1 70B on Groq to create o1-like reasoning chains
334 points148 comments
Ymilliondreams ·
Mixture-of-Depths: Dynamically allocating compute in transformers
281 points83 comments
Yhackerlight ·
Quiet-STaR: Language Models Can Teach Themselves to Think Before Speaking
280 points264 comments
Yamichail ·
How Chain-of-Thought Reasoning Helps Neural Networks Compute
270 points153 comments
Ybookofjoe ·
The Threat to OpenAI
150 points152 comments
Ylaurex ·
Can Demis Hassabis save Google?
133 points127 comments
Yspwa4 ·
OpenAI shows 'Strawberry' to feds, races to launch it
39 points73 comments
Yweinzierl ·
The "Chain of Thought" Delusions
22 points36 comments