Three things everyone should know about Vision Transformers

71•reqo•9mo ago

Comments

Centigonal•9mo ago

There's something that tickles me about this paper's title. The thought that everyone should know these three things. The idea of going to my neighbor who's a retired K-12 teacher and telling her about how adding MLP-based patch pre-processing layers improves Bert-like self-supervised training based on patch masking.

pixl97•9mo ago

Hey, when the AI powered T-rex is chasing you down you'll wish you paid attention that the vision transformers perception is based on movement!

Had to throw some Jurassic Park humor in here.

woopwoop•9mo ago

Clickbait titles are something of a tradition in this field by now. Some important paper titles include "One weird trick for parallelizing convolutional neural networks", "Attention is all you need", and "A picture is worth 16x16 words". Personally I still find it kind of irritating, but to each their own I guess.

minimaxir•9mo ago

Only the first one is clickbait in the style of blogs that incentivize you to click on the headline (i.e. the information gap), the last two are just fun puns.

janalsncm•9mo ago

Honestly I took the first one as making fun of that trope. Usually the “one weird trick to” ends in some tabloid-style thing like lose 15 pounds or find out if your husband is loyal. So “parallizing CNNs” is a joke, as if that’s something you’d see in a checkout isle.

woopwoop•9mo ago

In what sense is "Attention is all you need" a pun?

minimaxir•9mo ago

It's a reference to the lyric "love is all you need" from the song "All You Need Is Love" by the Beatles, and it uses a faux-synonym with a different meaning.

adultSwim•9mo ago

"Attention is all you need" is an outlier. They backed up their bold claim with breakthrough results.

For modest incremental improvements, I greatly prefer boring technical titles. Not everything needs to a stochastic parrot. We see this dynamic with building luxury condos. On any individual project, making that pick will help juice profit. When the whole city follows that , it leads to a less desirable outcome.

throwaway_x031•9mo ago

"Time and Space Are Not What You Think — Introducing the Special Theory of Relativity"

guerrilla•9mo ago

Yeah, I guess today was the day that I learned I am not part of "everyone". I feel so left out now.

i5heu•9mo ago

I put this paper into 4o so i can check if it is relevant, so that you do not have to do this too here are the bullet points:

- Vision Transformers can be parallelized to reduce latency and improve optimization without sacrificing accuracy.

- Fine-tuning only the attention layers is often sufficient for adapting ViTs to new tasks or resolutions, saving compute and memory.

- Using MLP-based patch preprocessing improves performance in masked self-supervised learning by preserving patch independence.

Jamesoncrate•9mo ago

just read the abstract

jmugan•9mo ago

You would think. I don't know about this paper in particular, but I'm continually surprised about how much more I get out of LLM summaries of papers than the abstracts of papers written by the authors.

tough•9mo ago

This would be an interesting metric to track, how different an abstract generated from LLM giving it the paper as source, vs the actual abstract is, and if it has any correlation whatsoever with the overall quality of the paper or not

mananaysiempre•9mo ago

Paper abstracts are not optimized by drive-by readers like you and me. They are optimized for active researchers in the field reading their daily arXiv digest that lists all the new papers across the categories they work in, and needing to take the read/don't-read decision for each entry there as efficiently as possible.

If you’ve already decided you’re interested in the paper, then the Introduction and/or Conclusion sections are what you’re looking for.

andai•9mo ago

Wouldn't a more comprehensive, digestible bullet point summary be even more helpful to actual researchers choosing which papers to read?

kridsdale3•9mo ago

Same. I don't think GP deserves the downvotes.

155M US land parcel boundaries

Private Inference

Font Rendering from First Principles

Show HN: Seedance 2.0 AI video generator for creators and ecommerce

Wally: A fun, reliable voice assistant in the shape of a penguin

Rewriting Pycparser with the Help of an LLM

Lobsters Vibecoding Challenge

E-Commerce vs. Social Commerce

Avoiding Modern C++ – Anton Mikhailov [video]

Show HN: AegisMind–AI system with 12 brain regions modeled on human neuroscience

Zig – Package Management Workflow Enhancements

AI-powered text correction for macOS

AppSecMaster – Learn Application Security with hands on challenges

Fibonacci Number Certificates

AI Overviews are killing the web search, and there's nothing we can do about it

City skylines need an upgrade in the face of climate stress

1979: The Model World of Robert Symes [video]

Satellites Have a Lot of Room

1980s Farm Crisis

Show HN: FSID - Identifier for files and directories (like ISBN for Books)

Show HN: Holy Grail: Open-Source Autonomous Development Agent

Show HN: Minecraft Creeper meets 90s Tamagotchi

Show HN: Termiteam – Control center for multiple AI agent terminals

The only U.S. particle collider shuts down

Ask HN: Why do purchased B2B email lists still have such poor deliverability?

Show HN: Remotion directory (videos and prompts)

Portable C Compiler

Show HN: Kokki – A "Dual-Core" System Prompt to Reduce LLM Hallucinations

Software Engineering Transformation 2026

Microsoft purges Win11 printer drivers, devices on borrowed time