Fascinating how human style institutions are precipitating out of mass agentic systems more and more. First moltbook, then agentic religions, now project/organisational legislatures. Great post!
I've got a pretty clear picture now of where we're headed next year, when Fable-class models become ubiquitous and cheap, and every enterprise is flooded with hundreds of new Fable-class AI employees.
Spoiler: They create a constitutional legal system.
https://t.co/9KqssPG6Od
@DanielleFong You might find this side project interesting, got models to keep a diary as they play and track what they look at: https://t.co/Yps1apzltf
Mark your calendars: @tiangolo and I will be live at https://t.co/mcrtjORJ4X by @pyconit on June 17 🎉 We'll talk about Python, FastAPI, spaCy & probably lots of other cool things. See you there!
📆 June 17, 9:10pm CET
🌎 Free & online
🐍 Details: https://t.co/MKgm1sVCU6
The most frequent Qs we receive at 🤗?
- I want to do this, which model should I choose
- How can I adapt the model on my data
We are working on a new project for this: ✨AutoNLP✨
For text classif, NER, summarize, translation…
Join the waiting list => https://t.co/yBcUW1LZjy
If you want to play with the new (or your own custom trained) @spacy_io v3 models, check out this app I built, hosted with the new @Streamlit sharing – also launched today, so perfect timing! 🥳
🖼️ App: https://t.co/SmhTUsQHHJ
👑 spacy-streamlit: https://t.co/BHXUkBkBgH
I saw a job post the other day. 👔
It required 4+ years of experience in FastAPI. 🤦
I couldn't apply as I only have 1.5+ years of experience since I created that thing. 😅
Maybe it's time to re-evaluate that "years of experience = skill level". ♻
Thrilled to share new work! “Retrieval-Augmented Generation for Knowledge-Intensive NLP tasks”.
Big gains on Open-Domain QA, with new State-of-the-Art results on NaturalQuestions, CuratedTrec and WebQuestions.
check out here: https://t.co/SVZ6K4tDn5.
1/N
Long-range sequence modeling meets 🤗 transformers! We are happy to officially release Reformer, a transformer that can process sequences as long as 500.000 tokens from @GoogleAI. Thanks a million, Nikita Kitaev and @lukaszkaiser! Try it out here: https://t.co/GwvMrt9lYk
@huggingface Quick question about your model repo. Is there an open API we can search against? I want to integrate a bunch of alternative models for my app with filterable attributes (language, task, etc.)
🎊 We just launched our new #website and logo 🔥!
✔️ Check out the new website: https://t.co/otpbzd81rJ
💌 To celebrate, we'd love to give everyone free stickers! Get your stickers via this form: https://t.co/pG3s7rPthK
🙌 We'll send these to you wherever you are:
What is the safest major source of energy today?
Nuclear!
No airborne pollution and no CO2 emissions (unlike coal, oil and gas).
Death rate from nuclear is 300 times lower than coal.
Yes, hydro, wind and solar... https://t.co/uQfSd1r9QC