Followin LogoFollowin
  • icon of HOMEicon of HOME
    Today
  • icon of INTELicon of INTEL
    Intel
  • icon of EXPLOREicon of EXPLORE
    Market
  • icon of EARNicon of EARN
    Earn
  • icon of SETTINGicon of SETTING
    Settings
    • Account
    • Theme Selection
      • Light
      • Dark
    • Language
      • English
      • 简体中文
      • 繁體中文
      • Tiếng Việt
      • 한국어
Followin APP
Mine Web3 Possibilities
avatar
Log in
avatar
Yann LeCun
388,057 Twitter followers
Follow
Professor at NYU. Chief AI Scientist at Meta. Researcher in AI, Machine Learning, Robotics, etc. ACM Turing Award Laureate.
Posts
avatar
Yann LeCun
02-08
UCLA doesn't pay Terry Tao $100M. Even if he's one of the best paid professors in the UC system, he makes considerably less than $1M. Like most self-respecting scientists, he is not motivated by money. Those "IMO winners hired by Meta" are engineers. Leaving Meta was my own decision. You have no idea what you're talking about.
TAO
5.74%
avatar
Yann LeCun
01-05
I think you missed the main ideas. - The basic premise of JEPA is that training by reconstructio/prediction in input space is evil (or counterproductive). The details are almost always unpredictable. Hence prediction must take place in representation space, where unpredictable
avatar
Yann LeCun
10-23
One cannot show that turbojets are safe before actually building turbojets and carefully refining them for reliability. The same goes for AI.
avatar
Yann LeCun
04-03
New paper from FAIR+NYU: Q: Is language supervision required to learn effective visual representations for multimodal tasks? A: No. ⬇️⬇️⬇️
-- END --