Hugging Face's logo
Activity Feed

AI & ML interests

None defined yet.

Recent Activity
clefourrierย  authored a paper 24 days ago
Strategic Navigation or Stochastic Search? How Agents and Humans Reason Over Document Collections
lewtunย  submitted a paper about 2 months ago
Single-minus gluon tree amplitudes are nonzero
lewtunย  submitted a paper about 2 months ago
Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL
View all activity
qgallouedecย 
posted an update 6 days ago
view post
Post
2092
TRL v1.0 is out!

Hugging Face's TRL library is downloaded 3 million times a month. Over 130k models trained with it are public on the Hub, and major projects like @unsloth and @axolotl-ai-co build directly on top of it. v1.0 is the moment we acknowledged that responsibility explicitly, with a real stability contract.

The field hasn't settled. Building stable software in a domain that keeps invalidating its own assumptions is the actual problem we're solving. The answer is a design that can absorb the next shift without breaking what people rely on.

What's in v1.0:
Deep Hugging Face integration, low infrastructure burden
What's next: asynchronous GRPO, better scaling support, and making training legible enough that agents can inspect and steer it.

pip install --upgrade trl


Read more: hf.co/blog/trl-v1
clefourrierย 
authored a paper 24 days ago
qgallouedecย 
posted an update about 2 months ago
view post
Post
2969
@CohereLabs just released ๐ŸŒฟ Tiny Aya: a fully open-source 3B parameter model that speaks 70+ languages ๐ŸŒ! But thereโ€™s a catch:

Tiny Aya is just a language model. It doesnโ€™t support tool calling, the key capability that turns frontier models into powerful *agents*.
So the real question is:

How hard is it to turn Tiny Aya into an agent?

Turns outโ€ฆ itโ€™s simple, thanks to Hugging Face TRL.
Weโ€™re sharing a hands-on example showing how to train Tiny Aya to turn it into a tool-calling agent using TRL, unlocking what could become the first *massively multilingual open agent*.

Small model. Global reach. Agent capabilities.

๐Ÿ‘‰ https://github.com/huggingface/trl/blob/main/examples/notebooks/sft_tool_calling.ipynb
  • 1 reply
ยท
lewtunย 
submitted 2 papers to Daily Papers about 2 months ago
julien-cย 
submitted a paper to Daily Papers 2 months ago
qgallouedecย 
submitted a paper to Daily Papers 3 months ago
nouamanetaziย 
posted an update 5 months ago
view post
Post
4654
After training ๐’๐ฆ๐จ๐ฅ๐‹๐Œ๐Ÿ‘ on ๐Ÿ‘๐Ÿ–๐Ÿ’ ๐‡๐Ÿ๐ŸŽ๐ŸŽ๐ฌ for nearly a month, I've come to realize something most people overlook: ๐ข๐ง๐Ÿ๐ซ๐š๐ฌ๐ญ๐ซ๐ฎ๐œ๐ญ๐ฎ๐ซ๐ž ๐ข๐ฌ ๐ญ๐ก๐ž ๐ฆ๐š๐ค๐ž-๐จ๐ซ-๐›๐ซ๐ž๐š๐ค ๐Ÿ๐š๐œ๐ญ๐จ๐ซ ๐ข๐ง ๐‹๐‹๐Œ ๐ญ๐ซ๐š๐ข๐ง๐ข๐ง๐ . ๐Ÿ”ฅ

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 ๐๐‚๐‚๐‹ ๐ž๐ซ๐ซ๐จ๐ซ๐ฌ, or when your expensive GPU cluster is running at ๐Ÿ”๐ŸŽ% ๐ž๐Ÿ๐Ÿ๐ข๐œ๐ข๐ž๐ง๐œ๐ฒ, the problem isn't your model. It's most probably a ๐ฆ๐ข๐ฌ๐ฎ๐ฌ๐ž ๐จ๐Ÿ ๐ญ๐ก๐ž ๐ก๐š๐ซ๐๐ฐ๐š๐ซ๐ž. ๐Ÿ› ๏ธ

Questions that seemed simple but had no clear answers: Why is ๐Œ๐จ๐„ ๐ญ๐ซ๐š๐ข๐ง๐ข๐ง๐  ๐ฌ๐ฅ๐จ๐ฐ๐ž๐ซ ๐ญ๐ก๐š๐ง ๐๐ž๐ง๐ฌ๐ž ๐ฆ๐จ๐๐ž๐ฅ๐ฌ? Which ๐๐‚๐‚๐‹ ๐Ÿ๐ฅ๐š๐ ๐ฌ should we actually set? How often should we checkpoint without killing throughput?

That's why we built ๐“๐ก๐ž ๐’๐ฆ๐จ๐ฅ ๐“๐ซ๐š๐ข๐ง๐ข๐ง๐  ๐๐ฅ๐š๐ฒ๐›๐จ๐จ๐ค ๐Ÿ“–: a complete guide covering everything from model architecture and data curation to the SmolLM3 training marathon, post-training techniques, and crucially, the ๐ข๐ง๐Ÿ๐ซ๐š๐ฌ๐ญ๐ซ๐ฎ๐œ๐ญ๐ฎ๐ซ๐ž ๐ฅ๐š๐ฒ๐ž๐ซ that most teams get wrong.

We validated real vs theoretical bandwidth across the entire stack: ๐‡๐๐Œ๐Ÿ‘ ๐ก๐ข๐ญ๐ญ๐ข๐ง๐  ๐Ÿ‘ ๐“๐/๐ฌ, ๐๐•๐‹๐ข๐ง๐ค ๐Ÿ’.๐ŸŽ ๐ซ๐ž๐š๐œ๐ก๐ข๐ง๐  ๐Ÿ•๐Ÿ–๐Ÿ” ๐†๐/๐ฌ, ๐๐‚๐ˆ๐ž ๐†๐ž๐ง๐Ÿ’ ๐š๐ญ ๐Ÿ๐Ÿ’.๐Ÿ ๐†๐/๐ฌ. Then we ran collective operations across ๐Ÿ๐Ÿ๐Ÿ– ๐†๐๐”๐ฌ (16 nodes, 8xH100s each) and measured how performance degrades at scale: all-reduce drops from ๐Ÿ’๐Ÿ–๐ŸŽ ๐†๐/๐ฌ on a single node to ๐Ÿ‘๐Ÿ๐ŸŽ-๐Ÿ‘๐Ÿ“๐ŸŽ ๐†๐/๐ฌ 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.

๐“๐ก๐ž ๐’๐ฆ๐จ๐ฅ ๐“๐ซ๐š๐ข๐ง๐ข๐ง๐  ๐๐ฅ๐š๐ฒ๐›๐จ๐จ๐ค: https://lnkd.in/e5MKXUHS

Shared with โค๏ธ by the HuggingFace team
megย 
posted an update 5 months ago
view post
Post
4171
๐Ÿค– Did you know your voice might be cloned without your consent from just *one sentence* of audio?
That's not great. So with @frimelle , we brainstormed a new idea for developers who want to curb malicious use: โœจThe Voice Consent Gate.โœจ
Details, code, here: https://huggingface.co/blog/voice-consent-gate
  • 3 replies
ยท
anditoย 
authored a paper 6 months ago
anditoย 
posted an update 6 months ago
view post
Post
2507
Finally, our new paper is out! "๐—™๐—ถ๐—ป๐—ฒ๐—ฉ๐—ถ๐˜€๐—ถ๐—ผ๐—ป: ๐—ข๐—ฝ๐—ฒ๐—ป ๐——๐—ฎ๐˜๐—ฎ ๐—œ๐˜€ ๐—”๐—น๐—น ๐—ฌ๐—ผ๐˜‚ ๐—ก๐—ฒ๐—ฒ๐—ฑ"! ๐Ÿฅณ
FineVision: Open Data Is All You Need (2510.17269)

If you've ever trained a VLM, you know this problem: nobody shares their data mixtures. It's a black box, making replicating SOTA work impossible.
We wanted to change that.

FineVision unifies 200 sources into 24 million samples. With 17.3 million images and 9.5 billion answer tokens, it's the largest open resource of its kind.

In the paper, we share how we built it:
๐Ÿ” finding and cleaning data at scale
๐Ÿงน removing excessive duplicates across sources
๐Ÿค— decontaminating against 66 public benchmarks

My favorite part is Figure 6 (in the video!). It's our visual diversity analysis. It shows that FineVision isn't just bigger; it's more balanced and conceptually richer than other open datasets.
NVIDIA's Eagle 2 paper highlighted just how critical this visual diversity is, and our results confirm it: models trained on FineVision consistently outperform those trained on any other open dataset on 11 benchmarks!

๐ŸŽ‰ To celebrate the paper, Iโ€™m also releasing a concatenated and shuffled version of the full dataset! ๐Ÿ‘‰HuggingFaceM4/FineVision_full_shuffled

Itโ€™s ready to stream, so you can start training your own models right away:

from datasets import load_dataset
d = load_dataset("HuggingFaceM4/FineVision_full_shuffled", split="train", streaming=True)
print(next(iter(d)))

A big shoutout to the first authors: Luis Wiedmann and Orr Zohar. They are rockstars!
ariG23498ย 
authored a paper 6 months ago
thomwolfย 
authored a paper 6 months ago
lvwerraย 
authored a paper 6 months ago
megย 
posted an update 7 months ago
view post
Post
2958
๐Ÿค– As AI-generated content is shared in movies/TV/across the web, there's one simple low-hanging fruit ๐Ÿ‡ to help know what's real: Visible watermarks. With the Gradio team, I've made sure it's trivially easy to add this disclosure to images, video, chatbot text. See how: https://huggingface.co/blog/watermarking-with-gradio
Thanks to the code collab in particular from @abidlabs and Yuvraj Sharma.
clefourrierย 
updated a collection 7 months ago
ariG23498ย 
posted an update 7 months ago
view post
Post
2013
New post is live!

This time we cover some major updates to transformers.

๐Ÿค—
  • 2 replies
ยท
guipenedoย 
updated a collection 7 months ago
megย 
posted an update 8 months ago
BrigitteTousiย 
posted an update 8 months ago