Lior⚡
Covering the latest in AI R&D • ML Engineer • Ex-Mila researcher • MIT Lecturer • Building @TheAlphaSignal, a technical newsletter read by 180,000+ ML experts.
View on 𝕏Threads
Model details: https://t.co/QFRibMAUnn
Alternative 1 https://t.co/vYHL4XY7Pb
Current text-to-image models struggle with visual text since they lack character-level input features. This paper shows that character-aware variants outperform character-blind mod...
TLDR: Researchers extended text-to-image diffusion to generate multiple images. To learn continuous motion, they propose Tune-A-Video with a tailored Sparse-Causal Attention, which...
The most popular repo was "4K-NeRF", an official implementation of Arxiv paper "4K-NeRF: High Fidelity Neural Radiance Fields at Ultra High Resolutions" https://t.co/EMTiinNKM2 h...
The most liked tweet was the first ever Comcast bill negotiated 100% with A.I and LLMs. https://t.co/MBYfiBZD50
This paper suggests that measuring data can help you gain better control of what modern ML systems will learn. It proposes different measurements to quantitatively characterize t...
If you want to stay up to date with the latest breakthroughs in AI check out our weekly summary. We use ML to identify the top papers, news, and repos. It's read by 25,000+ AI en...
That's it! If you want to stay up to date with the latest breakthroughs in AI check out our weekly summary. We use ML to identify the top papers, news, and repos. It's read by 25...
Notable Papers: "Coder Reviewer Reranking for Code Generation" https://t.co/6jgKiNC0f2 "On Distillation of Guided Diffusion Models" https://t.co/IL1MCDdoZO
With a model capable of generating answers similar to humans, they needed a way to tell the AI what was a good/bad answer. To solve that, they used humans (again) to rank randomly...
DeepMind publishes DeepNash - the first AI to master Stratego, a game more complex than chess, Go and poker. @karl_tuyls https://t.co/jZerEfnWUi 3/6