Log inSign up
Vaishaal Shankar
530 posts
Image
user avatar
Vaishaal Shankar
@Vaishaal
Birth date Add your date of birth Switch to professional
Berkeley, CA
vaishaal.com
Joined August 2008
362
Following
1,821
Followers
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Finally get to try the new bart trains! I've been literally waiting months for this. Will live tweet the experience.
    Image
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jul 18, 2024
    We have released our DCLM models on huggingface! To our knowledge these are by far the best performing truly open-source models (open data, open weight models, open training code) 1/5
    51K
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 18, 2024
    I am really excited to introduce DataComp for Language Models (DCLM), our new testbed for controlled dataset experiments aimed at improving language models. 1/x
    Image
    120K
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    Looks like the good-ol reboot trick didn't work. Now we are going to exit the train! In the tunnel! This is the most exciting commute I've had
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    They are trying to fix the train by turning the computer off and on again. I am sure this will work.
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    The bart technician says this is all because "all this technology is too new" [points to train].
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    Made it outside. Thanks @SFBART workers for handling the situation and getting everybody out safely!
    Image
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Oct 6, 2023
    I had an argument with @PreetumNakkiran about MLPs 4 years ago. He said with enough data + compute the MLP/ConvNet gap would go to 0. I was convolution-pilled and convinced this wasn't possible. He was right:
    arXiv logo
    arxiv.org
    Scaling MLPs: A Tale of Inductive Bias
    In this work we revisit the most fundamental building block in deep learning, the multi-layer perceptron (MLP), and study the limits of its performance on vision tasks. Empirical insights into...
    28K
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Dec 6, 2023
    Replying to @hankgreen
    It's a multiple choice exam that covers ~57 subjects. It's generally a good benchmark for capabilities of a model. 90% just means the model got 90% of these questions right. The paper is not a terrible read:
    arXiv logo
    arxiv.org
    Measuring Massive Multitask Language Understanding
    We propose a new test to measure a text model's multitask accuracy. The test covers 57 tasks including elementary mathematics, US history, computer science, law, and more. To attain high accuracy...
    15K
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    these trains are so much quieter I can hear the sound of my own thoughts
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Mar 6, 2020
    Neural Kernels Without Tangents arxiv.org/abs/2003.02237 Joint work with Alex Fang, @WSguo, Sara Fridovich-Keil, @lschmidt3, @jrk and @beenwrekt Taking inspiration from convolutional networks, we construct high performance kernel functions for image classification (1/6)
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    Well we evacuated the train successfully! Turns out we were only few hundred yards from 12 street bart
    Image
    00:00
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    Welp so much for that honeymoon phase. Looks like the train stopped abruptly in the tunnel before 12th street.
  • user avatar
    Vaishaal Shankar
    @Vaishaal
    Jun 24, 2019
    Replying to @Vaishaal
    They have succesfully dimmed the lights and turned off the AC.

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
Advertisement
Advertisement