LLMs are trained to compress data by mapping sequences to high-dim representations!
How does the complexity of this mapping change across LLM training? How does it relate to the model’s capabilities? 🤔
Announcing our #NeurIPS2025 📄 that dives into this.

🧵below
#AIResearch #MachineLearning #LLM

Yala lab's @kumarkagrawal.bsky.social introduces Atlas — a new AI model, inspired by need for better cancer detection, that uses multi-scale attention to analyze large images. At 4K resolution, it’s 7× faster than Vision Transformers and 30% more accurate than MambaVision. buff.ly/6wYexRm

Are you training self-supervised/foundation models, and worried if they are learning good representations? We got you covered! 💪
🦖Introducing Reptrix, a #Python library to evaluate representation quality metrics for neural nets: github.com/BARL-SSL/rep...
🧵👇[1/6]
#DeepLearning

Will multimodal models systematically generalize if trained on enough data? In a controlled VQA setting, we find it’s not data quantity, but data DIVERSITY that matters! 🧵

Joint w/ @ab-carrell.bsky.social @kumarkagrawal.bsky.social Yash Sharma @nsaphra.bsky.social
www.cs.toronto.edu/~ianberlot/d...