PinnedPublished inGeek CultureHow to learn Machine Learning in 2022A step by step guide to getting into machine learningJan 20, 2022A response icon3Jan 20, 2022A response icon3
The End of the Scaling Era: How Recursive Reasoning Outperforms Billion-Parameter ModelsSamsung’s Tiny Recursive Model achieves 87% accuracy where 671-billion parameter LLMs score 0% — by learning computation instead of…Oct 31Oct 31
LET’S EXPLORE THE NEW FRONTIER OF EFFICIENT LANGUAGE MODELSThe Problem: Matrix Multiplication BottleneckOct 31Oct 31
Why Jensen Huang Loves the “AI Bubble” StoriesAnswering why AI is not a Bubble and the Deeper Story at playOct 17A response icon13Oct 17A response icon13
Why Step-by-Step Prompting Works in AIUnderstanding the Secrets to behind this prompt engineering superpowerOct 8Oct 8
Dissecting Meta AI’s “Revolutionary” Code World ModelsWhat the performance means, beyond the headlinesOct 3Oct 3
The Chocolate Milk Cult’s Guide to Inference Scaling for AI ModelsHow to Reduce the costs of Running LLMsSep 21A response icon3Sep 21A response icon3
How to break the memory wall for large-scale Graph Neural NetworksHow to Train Graph Neural Nets 100x FasterSep 11Sep 11
The real significance of Google’s Nano BananaGoogle using a Diffusion Model in Gemini is a really big dealSep 3A response icon3Sep 3A response icon3
How Google and Stanford made AI more Interpretable with a 20 year old TechniqueWhy AI Interpretability Has Hit Its First Fundamental WallAug 28A response icon4Aug 28A response icon4