Alex Volkov - targum.video Profile picture
May 23 8 tweets 5 min read Twitter logo Read on Twitter
Captain... it's only tuesday 😅

As always we are coming to your ears on Thursday to cover the major things that happened in AI for the past week.

twitter.com/i/spaces/1lPKq…

First hour: recap of AI insanity
Second hour: Talking #genAI inside @Photoshop with @jnack !

More in🧵 Image
GANs are back! DragGAN enables dragging control points to change the image in ways not possible before!
Deeper dive into MMS from @MetaAI , is it better than #whisper?

(watch my blind reaction here)
Multi Modal future is coming at us Faster and Faster

Plugins are coming to... everywhere. Bing chat, windows, chatGPT!

Also, browsing is coming to everyone, including free accounts!

And of course we'll cover in depth, what ever this insanity means to the world of creators!

#firefly right inside @Adobe @Photoshop 🤯

We're going to do a deep dive with @jnack the mind behind Adobe Firefly, Ex-Google AI and Photoshop’s longtime PM

So what are you waiting for? Hit that reminder button, and join us on our weekly AI recap 👏 Can't wait for this one!

twitter.com/i/spaces/1lPKq…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Alex Volkov - targum.video

Alex Volkov - targum.video Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @altryne

May 24
Watching @karpathy presentation from today and taking twitter notes, come along for the ride:

If you're like only the practical tips, skip to #32

@karpathy starts with stages:
1 - Pre-training - months x thousands of GPUs
2, 3, 4 - Finetuning stages that take hours or days

1/ Image
Before pre-training happens, there are 2 preparation steps.

Data collection - Get tons of data from different sources (here Andrej LLaMa mixture)

Tokenization - a lossless translations between pieces of words and integers.

2/ ImageImage
"You shouldn't judge the power of the model just by the number of parameters it contains"

LLaMa has trained on 1-1.4 Trillion tokens vs 300B tokens in GPT-3.

3/ Image
Read 44 tweets
May 22
🇮🇱🇯🇵🇩🇪🇪🇸🇮🇳 x 1000
This is MASSIVE folks! (blind reaction)

TTS and STT in one model, that understands 1100 languages, better than whisper! and is able to generate audio in those languages?

Incredible thanks to @ylecun @boztank and tons of other folks who made this happen and… twitter.com/i/web/status/1…
@ylecun @boztank And we get force alignment tooling open source as well to help with the quality of longer transcriptions?? Image
This is one HELL of a flex!

MMS was trained on 45K hours of labeled data (15x LESS than whisper) and has twice better WER (word error rate, lower = better) while also supporting 11x more languages?

This is.. how exponential feels folks. Whisper is not eve 1 year old yet! 🤯 Image
Read 5 tweets
May 22
This quote is from ImageBind by @MetaAI seemed like science fiction just a short few weeks ago!

And now we have... MindVideo. Image
Here's the Mind-Video page:
mind-video.com

And Github:
github.com/jqin4749/MindV…
Read 5 tweets
Mar 26
My 🧵 of the @lexfridman and @sama pod (which is incredible by the way, I'm having a hard time selecting the clips!) got...twittered.

Starting a new thread with Sam talking about @elonmusk being a sometimes funny Jerk on Twitter

Oh and reacting to the "GPT is woke" takes… twitter.com/i/web/status/1…
@lexfridman @sama @elonmusk This is one powerful quote:

"Despite @elonmusk being a jerk on Twitter,
or whatever, I'm happy he exists in the world.
But I wish he would do more to look at the
hard work we're doing to get this stuff right."

- @sama
I loved this one about @lexfridma and @sama geeking about code editors, where Lex admits he switched to VsCode largely because of coPilot and Sama being all giddy about vscode 😂

Read 6 tweets
Mar 25
.@sama thinks that the fact that you can chat with GPT-4 about your code, is a "really big deal" only after 6 days (before plugin system was released)
@sama .@sama on the pressure from outrage journalism as it comes to AI
@sama "The reason Steve Jobs insisted on the handles on the old macs is to give humans the feeling of control, that they would be able to throw the computer out the window if it misbehaves"

There seem to be no handles like these in these models 😆
Read 10 tweets
Mar 22
I just finished watching @ilyasut with @nvidia CEO Jensen, and I was really expecting the discussion about GPT-4 and multi-modality 🤯

Posting a few highlights, of what I think is going to be a huge, HUGE change in everyone's life soon:
(transcribed w/ Targum of course)
1/
.@ilyasut explains everything in simple terms:
Multimodality is important for two reasons: 1) It's useful, as vision enhances the practicality and value of neural networks. 2) It allows us to learn more about the world through images, in addition to text.
targum.video/v/2023/3/22/a6…
A third of the human cortex is dedicated to vision. We as human beings learn visually way way before we learn verbally!

There's a whole understanding of the world that lies in vision
Read 10 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(