Hugging Face's logo

Evaluation datasets

community
Activity Feed

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
hynky  updated a dataset about 1 month ago
lighteval/QazUNTv2
hynky  updated a dataset about 2 months ago
lighteval/HAWP
View all activity

lighteval's activity

lewtun 
posted an update about 22 hours ago
view post
Post
1474
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
lewtun 
posted an update 8 days ago
view post
Post
1965
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
·
lewtun 
posted an update 21 days ago
view post
Post
6665
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 29 days ago
view post
Post
4572
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
hynky 
updated a dataset about 1 month ago
thomwolf 
posted an update about 1 month ago
SaylorTwift 
posted an update about 2 months ago
hynky 
updated a dataset about 2 months ago
albertvillanova 
posted an update about 2 months ago
view post
Post
1470
🚨 How green is your model? 🌱 Introducing a new feature in the Comparator tool: Environmental Impact for responsible #LLM research!
👉 open-llm-leaderboard/comparator
Now, you can not only compare models by performance, but also by their environmental footprint!

🌍 The Comparator calculates CO₂ emissions during evaluation and shows key model characteristics: evaluation score, number of parameters, architecture, precision, type... 🛠️
Make informed decisions about your model's impact on the planet and join the movement towards greener AI!
thomwolf 
posted an update about 2 months ago
hynky 
updated 2 datasets about 2 months ago
albertvillanova 
posted an update 2 months ago
view post
Post
1554
🚀 New feature of the Comparator of the 🤗 Open LLM Leaderboard: now compare models with their base versions & derivatives (finetunes, adapters, etc.). Perfect for tracking how adjustments affect performance & seeing innovations in action. Dive deeper into the leaderboard!

🛠️ Here's how to use it:
1. Select your model from the leaderboard.
2. Load its model tree.
3. Choose any base & derived models (adapters, finetunes, merges, quantizations) for comparison.
4. Press Load.
See side-by-side performance metrics instantly!

Ready to dive in? 🏆 Try the 🤗 Open LLM Leaderboard Comparator now! See how models stack up against their base versions and derivatives to understand fine-tuning and other adjustments. Easier model analysis for better insights! Check it out here: open-llm-leaderboard/comparator 🌐
albertvillanova 
posted an update 2 months ago
view post
Post
3135
🚀 Exciting update! You can now compare multiple models side-by-side with the Hugging Face Open LLM Comparator! 📊

open-llm-leaderboard/comparator

Dive into multi-model evaluations, pinpoint the best model for your needs, and explore insights across top open LLMs all in one place. Ready to level up your model comparison game?
thomwolf 
posted an update 2 months ago
view post
Post
4135
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
·
albertvillanova 
posted an update 2 months ago
view post
Post
1228
🚨 Instruct-tuning impacts models differently across families! Qwen2.5-72B-Instruct excels on IFEval but struggles with MATH-Hard, while Llama-3.1-70B-Instruct avoids MATH performance loss! Why? Can they follow the format in examples? 📊 Compare models: open-llm-leaderboard/comparator
albertvillanova 
posted an update 3 months ago
view post
Post
1919
Finding the Best SmolLM for Your Project

Need an LLM assistant but unsure which hashtag#smolLM to run locally? With so many models available, how can you decide which one suits your needs best? 🤔

If the model you’re interested in is evaluated on the Hugging Face Open LLM Leaderboard, there’s an easy way to compare them: use the model Comparator tool: open-llm-leaderboard/comparator
Let’s walk through an example👇

Let’s compare two solid options:
- Qwen2.5-1.5B-Instruct from Alibaba Cloud Qwen (1.5B params)
- gemma-2-2b-it from Google (2.5B params)

For an assistant, you want a model that’s great at instruction following. So, how do these two models stack up on the IFEval task?

What about other evaluations?
Both models are close in performance on many other tasks, showing minimal differences. Surprisingly, the 1.5B Qwen model performs just as well as the 2.5B Gemma in many areas, even though it's smaller in size! 📊

This is a great example of how parameter size isn’t everything. With efficient design and training, a smaller model like Qwen2.5-1.5B can match or even surpass larger models in certain tasks.

Looking for other comparisons? Drop your model suggestions below! 👇