Hugging Face's logo

Hugging Face Smol Cluster

community
Activity Feed Request to join this org

AI & ML interests

None defined yet.

Recent Activity

clefourrier  authored a paper about 1 month ago
Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation
lvwerra  authored a paper 2 months ago
SelfCodeAlign: Self-Alignment for Code Generation
clefourrier  authored a paper 6 months ago
The Hallucinations Leaderboard -- An Open Effort to Measure Hallucinations in Large Language Models
View all activity

HFSmolCluster's activity

lewtun 
posted an update about 5 hours ago
view post
Post
233
I was initially pretty sceptical about Meta's Coconut paper [1] because the largest perf gains were reported on toy linguistic problems. However, these results on machine translation are pretty impressive!

https://x.com/casper_hansen_/status/1875872309996855343

Together with the recent PRIME method [2] for scaling RL, reasoning for open models is looking pretty exciting for 2025!

[1] Training Large Language Models to Reason in a Continuous Latent Space (2412.06769)
[2] https://huggingface.co/blog/ganqu/prime
clem 
posted an update 3 days ago
lewtun 
posted an update 7 days ago
view post
Post
1922
This paper ( HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs (2412.18925)) has a really interesting recipe for inducing o1-like behaviour in Llama models:

* Iteratively sample CoTs from the model, using a mix of different search strategies. This gives you something like Stream of Search via prompting.
* Verify correctness of each CoT using GPT-4o (needed because exact match doesn't work well in medicine where there are lots of aliases)
* Use GPT-4o to reformat the concatenated CoTs into a single stream that includes smooth transitions like "hmm, wait" etc that one sees in o1
* Use the resulting data for SFT & RL
* Use sparse rewards from GPT-4o to guide RL training. They find RL gives an average ~3 point boost across medical benchmarks and SFT on this data already gives a strong improvement.

Applying this strategy to other domains could be quite promising, provided the training data can be formulated with verifiable problems!
  • 1 reply
·
clem 
posted an update 19 days ago
view post
Post
1720
Coming back to Paris Friday to open our new Hugging Face office!

We're at capacity for the party but add your name in the waiting list as we're trying to privatize the passage du Caire for extra space for robots 🤖🦾🦿

https://t.co/enkFXjWndJ
  • 1 reply
·
lewtun 
posted an update 21 days ago
view post
Post
6656
We outperform Llama 70B with Llama 3B on hard math by scaling test-time compute 🔥

How? By combining step-wise reward models with tree search algorithms :)

We show that smol models can match or exceed the performance of their much larger siblings when given enough "time to think"

We're open sourcing the full recipe and sharing a detailed blog post.

In our blog post we cover:

📈 Compute-optimal scaling: How we implemented DeepMind's recipe to boost the mathematical capabilities of open models at test-time.

🎄 Diverse Verifier Tree Search (DVTS): An unpublished extension we developed to the verifier-guided tree search technique. This simple yet effective method improves diversity and delivers better performance, particularly at large test-time compute budgets.

🧭 Search and Learn: A lightweight toolkit for implementing search strategies with LLMs and built for speed with vLLM

Here's the links:

- Blog post: HuggingFaceH4/blogpost-scaling-test-time-compute

- Code: https://github.com/huggingface/search-and-learn

Enjoy!
  • 2 replies
·
thomwolf 
posted an update 28 days ago
view post
Post
4561
We are proud to announce HuggingFaceFW/fineweb-2: A sparkling update to HuggingFaceFW/fineweb 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 multilingual datasets in our experiments.

The dataset is released under the permissive 📜 ODC-By 1.0 license, and the 💻 code to reproduce it and our evaluations is public.

We will very soon announce a big community project, and are working on a 📝 blogpost walking you through the entire dataset creation process. Stay tuned!

In the mean time come ask us question on our chat place: HuggingFaceFW/discussion

H/t @guipenedo @hynky @lvwerra as well as @vsabolcec Bettina Messmer @negar-foroutan and @mjaggi
  • 2 replies
·
clefourrier 
authored a paper about 1 month ago
thomwolf 
posted an update about 1 month ago
thomwolf 
posted an update about 1 month ago
clem 
posted an update about 1 month ago
view post
Post
4417
Six predictions for AI in 2025 (and a review of how my 2024 predictions turned out):

- There will be the first major public protest related to AI
- A big company will see its market cap divided by two or more because of AI
- At least 100,000 personal AI robots will be pre-ordered
- China will start to lead the AI race (as a consequence of leading the open-source AI race).
- There will be big breakthroughs in AI for biology and chemistry.
- We will begin to see the economic and employment growth potential of AI, with 15M AI builders on Hugging Face.

How my predictions for 2024 turned out:

- A hyped AI company will go bankrupt or get acquired for a ridiculously low price
✅ (Inflexion, AdeptAI,...)

- Open-source LLMs will reach the level of the best closed-source LLMs
✅ with QwQ and dozens of others

- Big breakthroughs in AI for video, time-series, biology and chemistry
✅ for video 🔴for time-series, biology and chemistry

- We will talk much more about the cost (monetary and environmental) of AI
✅Monetary 🔴Environmental (😢)

- A popular media will be mostly AI-generated
✅ with NotebookLM by Google

- 10 millions AI builders on Hugging Face leading to no increase of unemployment
🔜currently 7M of AI builders on Hugging Face
·
clem 
posted an update about 1 month ago
view post
Post
4373
Hugging Face is becoming the best place to share the most viral AI apps with spaces.

Kolors Virtual Try-on just crossed 6,000,000 unique visitors & is now the #5 most popular space. Congrats to the Kwai Kolors team!

Kwai-Kolors/Kolors-Virtual-Try-On
  • 2 replies
·
clem 
posted an update about 1 month ago
view post
Post
1979
I've been in Brazil for 10 days now 🇧🇷🇧🇷🇧🇷

I've been surprised by the gap between the massive number of people interested in AI (chatgpt adoption is crazy here) and the relatively low number of real AI builders - aka people and companies building their own AI models, datasets and apps.

Lots of efforts needed across the world for everyone to participate, control and benefit this foundational technology, starting with open-source & multi-lingual AI, more access to GPUs & AI builder training for all!
loubnabnl 
posted an update about 1 month ago
view post
Post
1747
Making SmolLM2 reproducible: open-sourcing our training & evaluation toolkit 🛠️ https://github.com/huggingface/smollm/

- Pre-training code with nanotron
- Evaluation suite with lighteval
- Synthetic data generation using distilabel (powers our new SFT dataset HuggingFaceTB/smoltalk)
- Post-training scripts with TRL & the alignment handbook
- On-device tools with llama.cpp for summarization, rewriting & agents

Apache 2.0 licensed. V2 pre-training data mix coming soon!

Which other tools should we add next?
thomwolf 
posted an update about 1 month ago
SaylorTwift 
posted an update about 2 months ago
thomwolf 
posted an update about 2 months ago
clem 
posted an update about 2 months ago
view post
Post
855
I'm in Belo Horizonte for a few days. Any HF community members in Brazil? What are the coolest orgs, models, datasets, spaces from here?
  • 2 replies
·
lvwerra 
authored a paper 2 months ago
thomwolf 
posted an update 2 months ago
view post
Post
4133
Parents in the 1990: Teach the kids to code
Parents now: Teach the kids to fix the code when it starts walking around 🤖✨
  • 2 replies
·
clem 
posted an update 2 months ago
view post
Post
4441
This is no Woodstock AI but will be fun nonetheless haha. I’ll be hosting a live workshop with team members next week about the Enterprise Hugging Face hub.

1,000 spots available first-come first serve with some surprises during the stream!

You can register and add to your calendar here: https://streamyard.com/watch/JS2jHsUP3NDM
·