Diffusion language models write a whole block of text at once instead of word by word.
Badges
Badges come from GitHub once the author claims this feed.
Recent posts
every link opens the author’s post-
Can a Trillion-Parameter Diffusion LLM Actually Work? Inside Nvidia's Bet on Parallel Text Generation
-
The King Is Dead: Jensen Huang's Nvidia Ousting and the Holy Grail of Founder Control
Why Nvidia's board could remove its founder CEO, and how startup founders can use dual-class shares, board seats, and voting control to protect their companies
-
SpaceX Wants to Be Your Phone Company. It Just Hired Qualcomm to Build the Radios.
The company that put thousands of routers in orbit is now coming down to earth: steel towers, rooftop leases, zoning boards and all.
-
Guido van Rossum: Why Python's Creator Quit as Dictator for Life
He wrote Python over a 1989 Christmas break.
-
Starship's Heat Shield: Can 18,000 Tiles Escape the Space Shuttle's Fatal Flaw?
A broken heat shield doomed Columbia.
-
Mercury 3 vs. the Fast Autoregressive Models: A Technical Report
How fast is a diffusion LLM in practice, what does it get right, and what does it cost per right answer?
-
Moxie Marlinspike: The Anarchist Sailor Who Built Signal and Encrypted Billions of Chats
Moxie Marlinspike, a dreadlocked squatter who sailed a $1,000 wreck to the Caribbean with bags of bricks for anchors, built Signal and the encryption guarding billions of WhatsApp and Messenger chats, then got Facebook, Google, and Microsoft to run his code.