Lightnews — Scholar-powered news

Clem Delangue 🤗 @clem.hf.co · Dec 24

Happy holidays everyone! What would you want to see more from Hugging Face in 2025?

Reposted by Clem Delangue 🤗

Ben Burtenshaw @benburtenshaw.bsky.social · Dec 19

People are flexing their end of year stats, so I made this app to show @hf.co hub stats in a tidy design!

Thanks @jfcalvo.hf.co and @ameeelie.bsky.social for the feature!

1 2 13

Clem Delangue 🤗 @clem.hf.co · Dec 16

This is the power of open research and open-source AI:
huggingface.co/spaces/Huggi...

Scaling test-time compute - a Hugging Face Space by HuggingFaceH4

Discover amazing ML apps made by the community

huggingface.co

1 11

Clem Delangue 🤗 @clem.hf.co · Dec 16

Just 10 days after o1's public debut, we’re thrilled to unveil the open-source version of the technique behind its success: scaling test-time compute

By giving models more "time to think," Llama 1B outperforms Llama 8B in math—beating a model 8x its size. The full recipe is open-source!

4 18 82

Clem Delangue 🤗 @clem.hf.co · Dec 13

Who’s at #neurips2024 and want to meet HF team members?

3

Reposted by Clem Delangue 🤗

Guilherme Penedo @guilherme.hf.co · Dec 8

Announcing 🥂 FineWeb2: A sparkling update with 1000s of 🗣️languages.

We applied the same data-driven approach that led to SOTA English performance in🍷 FineWeb to thousands of languages.

🥂 FineWeb2 has 8TB of compressed text data and outperforms other datasets.

1 19 75

Reposted by Clem Delangue 🤗

Thomas Wolf @thomwolf.bsky.social · Dec 8

The FineWeb team is happy to finally release "FineWeb2" 🥂🥳

FineWeb 2 extends the data driven approach to pre-training dataset design that was introduced in FineWeb 1 to now covers 1893 languages/scripts

Details: huggingface.co/datasets/Hug...

A detailed open-science tech report is coming soon

3 13 110

Reposted by Clem Delangue 🤗

Christopher Akiki @cakiki.bsky.social · Dec 8

The folks at Foursquare released a @hf.co dataset of 104.5 million places of interest and here's all of them plotted using datashader

2 3 16

Clem Delangue 🤗 @clem.hf.co · Dec 6

I weirdly love this! #5 trending on HF right now

9

Clem Delangue 🤗 @clem.hf.co · Dec 5

Excited to see more biology open-source models for real positive use-cases of AI!

Chai does structure predictions at AlphaFold3 levels of accuracy and able to handle multi-peptide or peptide-ligand complexes rather than just single chains.

Apache 2.0 on HF huggingface.co/chaidiscover...

3 3 19

Clem Delangue 🤗 @clem.hf.co · Dec 4

Very cool! Will share it in the coming days!

2

Reposted by Clem Delangue 🤗

Cosmico @cosmico.org · Dec 4

🤖 6 AI Predictions for 2025 by Hugging Face CEO

#HuggingFace
#ClementDelangue
#ArtificialIntelligence

www.cosmico.org/6-ai-predict...

6 AI Predictions for 2025 by Hugging Face CEO | Cosmico

Hugging Face CEO predicts AI protests, market disruptions, personal robots, and breakthroughs in science, as China leads the AI race in 2025.

www.cosmico.org

3 7

Reposted by Clem Delangue 🤗

Daniel Vila @dvilasuero.hf.co · Nov 26

Let's make AI more inclusive.

At @huggingface.bsky.social we'll launch a huge community sprint soon to build high-quality training datasets for many languages.

We're looking for Language Leads to help with outreach.

Find your language and nominate yourself:
forms.gle/iAJVauUQ3FN8...

8 21 52

Clem Delangue 🤗 @clem.hf.co · Dec 2

How my predictions for 2024 turned out:

3 1 15

Clem Delangue 🤗 @clem.hf.co · Dec 2

Six predictions for AI in 2025 (and a review of how my 2024 predictions turned out):

4 9 55

Reposted by Clem Delangue 🤗

Ted Underwood @tedunderwood.com · Nov 29

I would put this even more strongly: open source AI is probably our only realistic chance to avoid a terrifying increase in concentration of power. I do not want to live in a world where the people with all the money also have all the intellectual power.

Nathan Lambert @natolambert.bsky.social · Nov 29

The most realistic reason to be pro open source AI is to reduce concentration of power.

Alondra Nelson @alondra.bsky.social · Nov 29

"money has flowed to tech giants and others in their orbit... [and] raises an uncomfortable prospect: that this supposedly revolutionary technology might never deliver on its promise of broad economic transformation, but instead just concentrate more wealth" www.bloomberg.com/opinion/arti...

23 85 410

Clem Delangue 🤗 @clem.hf.co · Dec 1

Good list if you want to understand AI! go.bsky.app/Nik64nt

6 18 58

Clem Delangue 🤗 @clem.hf.co · Nov 29

QwQ is #1 trending on @hf.co!

3 1 13

Reposted by Clem Delangue 🤗

Simon Willison @simonwillison.net · Nov 29

This demo of structured data extraction running on an LLM that executes entirely in the browser (Chrome only for the moment since it uses WebGPU) is amazing

My notes here: simonwillison.net/2024/Nov/29/...

4 23 180

Reposted by Clem Delangue 🤗

Ishan Khatri @ishan.khatri.io · Nov 29

Wasn't going to wade into this but... as long as the internet exists in an open form, scraping will exist, and that's kind of the whole point... APIs exist because interacting with data is valuable and scraping (the old fashioned way) is annoying.

1 1 14

Clem Delangue 🤗 @clem.hf.co · Nov 29

For me, the biggest risk in AI is centralization of power and benefits in the hands of a few. This is why at @hf.co, we've been investing on building online courses for as many people as possible to understand and learn to build AI themselves. You can find some of them here: huggingface.co/learn

2 6 39

Reposted by Clem Delangue 🤗

Nathan Lambert @natolambert.bsky.social · Nov 29

The most realistic reason to be pro open source AI is to reduce concentration of power.

Alondra Nelson @alondra.bsky.social · Nov 29

"money has flowed to tech giants and others in their orbit... [and] raises an uncomfortable prospect: that this supposedly revolutionary technology might never deliver on its promise of broad economic transformation, but instead just concentrate more wealth" www.bloomberg.com/opinion/arti...

ChatGPT’s $8 Trillion Birthday Gift to Big Tech

Two years in, generative AI’s value to the world is still unclear. But these charts show that it’s been a bonanza for the largest tech firms.

www.bloomberg.com

6 36 190

Reposted by Clem Delangue 🤗

Jay 🦋 @jay.bsky.team · Nov 28

I’m thankful for everyone using Bluesky, everyone building on atproto, everyone listening to our message and sharing our dream of a better social media ecosystem that puts people first. We’re going to do this together — thanks for joining us on this journey!

1.5K 4.2K 54K

Reposted by Clem Delangue 🤗

Stella Biderman @stellaathena.bsky.social · Nov 28

A dataset of 1 million or 2 million Bluesky posts is completely irrelevant to training large language models.

The primary usecase for the datasets that people are losing their shit over isn't ChatGPT, it's social science research and developing systems that improve Bluesky.

Jeremy Howard @howard.fm · Nov 28

Did you know that 99% of email today is spam? Your inbox isn’t 99% spam because AI is used to filter it.

The same 99% will happen here too, but if AI researchers continue to get perma-banned for making available the datasets needed to filter it, it’s going to make this platform unusable.

8 39 250

Reposted by Clem Delangue 🤗

Jeremy Howard @howard.fm · Nov 28

Did you know that 99% of email today is spam? Your inbox isn’t 99% spam because AI is used to filter it.

The same 99% will happen here too, but if AI researchers continue to get perma-banned for making available the datasets needed to filter it, it’s going to make this platform unusable.

42 64 520