Blog-explorers

Activity Feed Request to join this org

AI & ML interests

None defined yet.

Recent Activity

ccocks-deca new activity 13 days ago

blog-explorers/README:Card forensics using VLM localized prompt approach

suayptalha authored a paper about 1 month ago

Superpositional Gradient Descent: Harnessing Quantum Principles for Model Training

ccocks-deca new activity about 1 month ago

blog-explorers/README:Pending Blog-Explorers Access Request

View all activity

Jofthomas

posted an update 8 days ago

Post

3362

The new Mistral 3 models are here !

Today, we announce Mistral 3, the next generation of Mistral models. Mistral 3 includes three state-of-the-art small, dense models (14B, 8B, and 3B) and Mistral Large 3 – our most capable model to date – a sparse mixture-of-experts trained with 41B active and 675B total parameters.

All models are released under the Apache 2.0 license.

Ministrals :
https://huggingface.co/collections/mistralai/ministral-3

Mistral Large 3:
https://huggingface.co/collections/mistralai/mistral-large-3

2 replies

juhoinkinen

posted an update 13 days ago

Post

226

**AI4LAM’s annual conference, AI Everywhere, All at Once
December 3 – 5, 2025, British Library, London**

See the conference programme: 👉 https://www.conftool.org/fantastic-futures-2025/sessions.php

Some program items related to NatLibFi/Annif:
• Workshop:
• Evaluating Automated Subject Indexing Methods, Maximilian Kähler
• Presentations:
• Autocat Cataloguing Assistant
• The usage of hardware resources for automatic subject cataloguing at the German National Library – an analysis and outlook for future challenges, Christoph Poley
• Posters:
• AI-Powered Subject Indexing in the Archives – Piloting Finto AI at the Finnish Literature Society, Milla Eräsaari and Teemu Hirvonen
• From Annotation to Insight: Human-in-the-Loop Machine Learning for Historical Archives in HAICu WP2, C.A. Romein and others

reach-vb

authored a paper 13 days ago

Open ASR Leaderboard: Towards Reproducible and Transparent Multilingual and Long-Form Speech Recognition Evaluation

Paper • 2510.06961 • Published Oct 8 • 10

ccocks-deca

in blog-explorers/README 13 days ago

Card forensics using VLM localized prompt approach

#11 opened 14 days ago by

vijayksagi

nroggendorff

posted an update 15 days ago

Post

2811

I am now being charged for paused and unstarted spaces out of the blue.
I think this is it, folks. o7

The unstarted spaces I can get behind. I would've appreciated a warning email first, but whatever. However, every time I restart the active usage goes up, despite all of my spaces being moved to CPU (free), and being paused.

13 replies

nroggendorff

posted an update 25 days ago

Post

2343

Developing with ZeroGPU without a PRO account is painful. They give you so many requests at once, but then have like a 24 hour cooldown. I vote less requests in a batch, but then a shorter cooldown.

or just less of a cooldown, but i understand if that is not allowed

3 replies

GeorgeBredis

authored a paper 30 days ago

ESSA: Evolutionary Strategies for Scalable Alignment

Paper • 2507.04453 • Published Jul 6 • 3

xianbao

authored a paper about 1 month ago

RoboChallenge: Large-scale Real-robot Evaluation of Embodied Policies

Paper • 2510.17950 • Published Oct 20 • 7

suayptalha

authored a paper about 1 month ago

Superpositional Gradient Descent: Harnessing Quantum Principles for Model Training

Paper • 2511.01918 • Published Nov 1 • 11

ccocks-deca

in blog-explorers/README about 1 month ago

Pending Blog-Explorers Access Request

#10 opened about 1 month ago by

KarthikAvinash

nouamanetazi

posted an update about 1 month ago

Post

3971

After training 𝐒𝐦𝐨𝐥𝐋𝐌𝟑 on 𝟑𝟖𝟒 𝐇𝟏𝟎𝟎𝐬 for nearly a month, I've come to realize something most people overlook: 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐢𝐬 𝐭𝐡𝐞 𝐦𝐚𝐤𝐞-𝐨𝐫-𝐛𝐫𝐞𝐚𝐤 𝐟𝐚𝐜𝐭𝐨𝐫 𝐢𝐧 𝐋𝐋𝐌 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠. 🔥

Everyone talks about model architecture and data quality. And yes, those matter immensely. But here's what nobody tells you: when your training run fails at 2 AM because of mysterious 𝐍𝐂𝐂𝐋 𝐞𝐫𝐫𝐨𝐫𝐬, or when your expensive GPU cluster is running at 𝟔𝟎% 𝐞𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲, the problem isn't your model. It's most probably a 𝐦𝐢𝐬𝐮𝐬𝐞 𝐨𝐟 𝐭𝐡𝐞 𝐡𝐚𝐫𝐝𝐰𝐚𝐫𝐞. 🛠️

Questions that seemed simple but had no clear answers: Why is 𝐌𝐨𝐄 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐬𝐥𝐨𝐰𝐞𝐫 𝐭𝐡𝐚𝐧 𝐝𝐞𝐧𝐬𝐞 𝐦𝐨𝐝𝐞𝐥𝐬? Which 𝐍𝐂𝐂𝐋 𝐟𝐥𝐚𝐠𝐬 should we actually set? How often should we checkpoint without killing throughput?

That's why we built 𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤 📖: a complete guide covering everything from model architecture and data curation to the SmolLM3 training marathon, post-training techniques, and crucially, the 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐥𝐚𝐲𝐞𝐫 that most teams get wrong.

We validated real vs theoretical bandwidth across the entire stack: 𝐇𝐁𝐌𝟑 𝐡𝐢𝐭𝐭𝐢𝐧𝐠 𝟑 𝐓𝐁/𝐬, 𝐍𝐕𝐋𝐢𝐧𝐤 𝟒.𝟎 𝐫𝐞𝐚𝐜𝐡𝐢𝐧𝐠 𝟕𝟖𝟔 𝐆𝐁/𝐬, 𝐏𝐂𝐈𝐞 𝐆𝐞𝐧𝟒 𝐚𝐭 𝟏𝟒.𝟐 𝐆𝐁/𝐬. Then we ran collective operations across 𝟏𝟐𝟖 𝐆𝐏𝐔𝐬 (16 nodes, 8xH100s each) and measured how performance degrades at scale: all-reduce drops from 𝟒𝟖𝟎 𝐆𝐁/𝐬 on a single node to 𝟑𝟐𝟎-𝟑𝟓𝟎 𝐆𝐁/𝐬 across 16 nodes.

If you've ever wondered why your training runs are slower than they should be, or you're planning to scale up and want to avoid expensive mistakes, this guide might save you weeks of debugging.

𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤: https://lnkd.in/e5MKXUHS

Shared with ❤️ by the HuggingFace team

nroggendorff

posted an update about 2 months ago

Post

3708

Is it hot in here, or is it just me?

5 replies

huybery

authored a paper about 2 months ago

VideoAgentTrek: Computer Use Pretraining from Unlabeled Videos

Paper • 2510.19488 • Published Oct 22 • 19

nroggendorff

posted an update about 2 months ago

Post

3343

I love getting emails telling me when there's somebody else's active access token in one of my commit SHAs. HF should really only tell you if it is your token, otherwise I could just make a dataset with a bunch of random strings and wait for a valid token.

user,permission,token
nroggendorff,write,hf_...
pepper13,finegrained,hf_...
...,...,...
...

Also, don't comment about how unlikely this is. I've gotten a warning email about a token I 'leaked' at least four times.
In all cases, it has been in the digest hash.