"Machine Learning Engineering Open Book" by Stas Bekman This open book has a lot of information on the Engineering aspects of building DL / ML models specifically LLM and Multi-modal models. This open book is continuously being updated. A pdf version can also be downloaded.
MULTIMODAL AI
-
AI Dominates Super Bowl Commercials with Generative Video
By
–
AI just took over the Super Bowl commercials.
— Rowan Cheung (@rowancheung) 12 février 2024
Here are my 6 favorite AI commercials from the big game:
1. Despicable Me 4, featuring minions prompting comical AI video generations. pic.twitter.com/S3QH3A3yFGAI just took over the Super Bowl commercials. Here are my 6 favorite AI commercials from the big game: 1. Despicable Me 4, featuring minions prompting comical AI video generations.
-
Decision Transformer Learns Game Instructions from Multimodal Input
By
–
Read to Play (R2-Play): Decision Transformer with Multimodal Game Instruction Jin et al.: https://
arxiv.org/abs/2402.04154 #ArtificialIntelligence #DeepLearning #MachineLearning -
Images from Noise: Quantum Emergence in AI Generation
By
–
images arising from noise like life emerging from quantum foam
-
Resolution Limitations for Professional Work Applications
By
–
I don’t find the resolution sufficient for doing work. Good enough for movies though
-
VR Girlfriend Technology: LLM and AR Integration for Real-World Second Life
By
–
VR girlfriend is there. Insane, LLm+AR= realworld second life
-
User critique of AI model image generation performance
By
–
Feels a bit like it is still way behind on many tasks Rejections is a massive problem (so it forgot how to draw images)
-
MusicLM Reconstructs Music from Brain Activity Signals
By
–
How can we reconstruct music from brain activity? Learn how our music generation model, MusicLM, was used to reconstruct music in this Research@ NYC lightning talk from Timo Denk, Software Engineer at Google → https://t.co/U8ph36azjv pic.twitter.com/IOgM0iH8BN
— Google AI (@GoogleAI) 9 février 2024How can we reconstruct music from brain activity? Learn how our music generation model, MusicLM, was used to reconstruct music in this Research@ NYC lightning talk from Timo Denk, Software Engineer at Google → https://
goo.gle/3SNJll2 -
Stable Audio Paper Details High-Quality Audio Generation Model
By
–
Curious how @stableaudio works? The team’s just released a paper detailing how the model operates with high sampling rates, flexible audio lengths, and comprehensive composition: https://t.co/mIhbcXEVwT https://t.co/aHzaA28ILI
— Stability AI (@StabilityAI) 9 février 2024Curious how @stableaudio works? The team’s just released a paper detailing how the model operates with high sampling rates, flexible audio lengths, and comprehensive composition: https://
arxiv.org/abs/2402.04825 -
Programmable AI Agents in Hardware Enable Autonomous Actions
By
–
This is like programmable AI agents within your hardware. Just screen grab something and it does whatever actions you want. It’s insane.