Simone Romano โš—๏ธ๐Ÿ‘๏ธ๐Ÿ’ญ๐Ÿ•๏ธ (@ialuronico) 's Twitter Profile
Simone Romano โš—๏ธ๐Ÿ‘๏ธ๐Ÿ’ญ๐Ÿ•๏ธ

@ialuronico

๐Ÿ–ค machine learning products. X-(Bigtech+, researcher)

ID: 12414242

linkhttps://simoneromano.com calendar_today18-01-2008 20:53:12

5,5K Tweet

1,1K Followers

4,4K Following

Simone Romano โš—๏ธ๐Ÿ‘๏ธ๐Ÿ’ญ๐Ÿ•๏ธ (@ialuronico) 's Twitter Profile Photo

Struggling to reach users on generative search platforms? Learn how Generative Engine Optimization (GEO) helps you capture and convert high-intent traffic. moz.com/blog/generativโ€ฆ via @moz

Heurist (@heurist_ai) 's Twitter Profile Photo

Claude deep research is a living proof that specialized agents working together unlock exponentially better results. The future of AI is multi-agent. Heurist is here for it: An open network of agents = insights that no single agent could achieve alone

Lennart Purucker (@lennartpurucker) 's Twitter Profile Photo

๐ŸšจWhat is SOTA on tabular data, really? We are excited to announce ๐—ง๐—ฎ๐—ฏ๐—”๐—ฟ๐—ฒ๐—ป๐—ฎ, a living benchmark for machine learning on IID tabular data with: ๐Ÿ“Š an online leaderboard (submit!) ๐Ÿ“‘ carefully curated datasets ๐Ÿ“ˆ strong tree-based, deep learning, and foundation models ๐Ÿงต

๐ŸšจWhat is SOTA on tabular data, really? We are excited to announce ๐—ง๐—ฎ๐—ฏ๐—”๐—ฟ๐—ฒ๐—ป๐—ฎ, a living benchmark for machine learning on IID tabular data with:

๐Ÿ“Š an online leaderboard (submit!)
๐Ÿ“‘ carefully curated datasets
๐Ÿ“ˆ strong tree-based, deep learning, and foundation models

๐Ÿงต
Keyon Vafa (@keyonv) 's Twitter Profile Photo

Can an AI model predict perfectly and still have a terrible world model? What would that even mean? Our new ICML paper formalizes these questions One result tells the story: A transformer trained on 10M solar systems nails planetary orbits. But it botches gravitational laws ๐Ÿงต

Ethan Mollick (@emollick) 's Twitter Profile Photo

Right now, AI adoption in organizations is constrained by the need to figure out how to integrate AI with the complex & often poorly-understood processes inside companies But ChatGPT agent suggests that The Bitter Lesson of AI may come for real work, too. open.substack.com/pub/oneusefultโ€ฆ

Emad (@emostaque) 's Twitter Profile Photo

Grok Imagine is amazing, heading towards what I always wanted to build at Stability AI The generation speed & relative quality is great When the new video model is trained you can see how thoughts will be given form in future Great job โฆxAIโฉ team!

Andrew Ng (@andrewyng) 's Twitter Profile Photo

I'm thrilled to announce the definitive course on Claude Code, created with Anthropic and taught by Elie Schoppik Elie Schoppik. If you want to use highly agentic coding - where AI works autonomously for many minutes or longer, not just completing code snippets - this is it.

Maxime Rivest ๐Ÿง™โ€โ™‚๏ธ๐Ÿฆ™ (@maximerivest) 's Twitter Profile Photo

Today, I pruned 87.24% of Qwen 30B for a sentiment classification task while keeping 100% of its accuracy. This means we get to use big models on gpus with not that much RAM (potentially running models that would normally require an H100 on 3090 type gpus)! Imagine pruning

Today, I pruned 87.24% of Qwen 30B for a sentiment classification task while keeping 100% of its accuracy. This means we get to use big models on gpus with not that much RAM (potentially running models that would normally require an H100 on 3090 type gpus)!

Imagine pruning
ericosiu (@ericosiu) 's Twitter Profile Photo

Lenny Rachitsky a bunch - mainly leveraging subagents + MCPs: - SEO revenue optimizer to find high converting pages and coming up with new pillars to go after. Then creates blog post or programmatic SEO templates that sit in our CMS as drafts. It also finds page cannibalization issues as well

clem ๐Ÿค— (@clementdelangue) 's Twitter Profile Photo

If you think Apple is not doing much in AI, you're getting blindsided by the chatbot hype and not paying enough attention! They just released FastVLM and MobileCLIP2 on Hugging Face. The models are up to 85x faster and 3.4x smaller than previous work, enabling real-time vision