Thank you so much and appreciate you taking time to check the paper. We trained a model(
https://
huggingface.co/neulab/Cultura
lPangea-7B
…) on the subset of the dataset and were able to improve performance on many multilingual/cultural multimodal benchmarks. Yeahhh, would be great too if you could also
RESEARCH
-
Training multilingual cultural model improves multimodal benchmark performance
By
–
-
Stable Audio 2.5 live chat with ComfyUI demonstrates inpainting
By
–
We joined our friends at @ComfyUI for a live chat about what's new with Stable Audio 2.5 🥁
— Stability AI (@StabilityAI) 21 octobre 2025
Check out CJ Carr from our Audio Research team demoing the model’s inpainting capability.
You can watch the full episode here for more insights and examples 👉 https://t.co/430lf6kve3 pic.twitter.com/2A0KHZieQBWe joined our friends at @ComfyUI for a live chat about what's new with Stable Audio 2.5 🥁 Check out CJ Carr from our Audio Research team demoing the model’s inpainting capability. You can watch the full episode here for more insights and examples 👉 bit.ly/43nRa5x
→ View original post on X — @stabilityai, 2025-10-21 19:31 UTC
-
Scaling 2.8M Images and 3M Entities with Flexible Data Curation Pipeline
By
–
We have around 2.8M unique images and 3M unique entities and the data curation pipeline allow scaling that up(more images per entity, adding new countries/languages, questions per entity, etc…). Very glad to hear people were asking about this 🙂
-
Major VQA Dataset Release: 30M Samples Across 42 Countries
By
–
Congrats! Big props unifying vision language datasets and making it easy to access them. We also recently released a massive VQA dataset of 30M samples(22M filtered), spanning 42 countries and 39 languages. The dataset contains factual visual questions and answers about world
-
Physics-Informed AI Improves Molecular Structure Predictions
By
–
Physics-informed learning is essential for AI+Science. Check out our latest paper in @PNASNews special edition on AI+Chemistry on incorporating simple rules of physics that two atoms in a molecule cannot get too close to one another (due to repulsive forces) vastly improves
-
Majority of Americans Have Never Used ChatGPT
By
–
RT @KaitlynZhou: As of June 2025, 66% of Americans have never used ChatGPT. Our new position paper, Attention to Non-Adopters, explores… [Translated from EN to English]
-
Complexity vs Feature Engineering Trade-offs in Machine Learning
By
–
The less complexity the more capacity to focus on learning other things. But the more handcrafted features the less opportunity to learn better representations. It all kind of depends on how big the model is, how much data you have, and what your compute resources are
-
New Tokenization Approach Challenges BPE-Style Methods
By
–
TLDR: surprised that it works, skeptical whether this will replace BPE-style tokenizers, delighted about this refreshing approach
-

LLM Research Neglects Non-Adopters of ChatGPT
By
–
As of June 2025, 66% of Americans have never used ChatGPT. Our new position paper, Attention to Non-Adopters, explores why this matters: LLM research is being shaped around adopters, leaving non-adopters’ needs and key research opportunities behind. arxiv.org/abs/2510.15951
-
Inference scaling costs exceed RL training for large-scale deployments
By
–
Depends. If you are serving millions of customers this type of inference scaling is going to cost you more than training with RL.