5 papers
UEPS: Robust and Efficient MRI Reconstruction
Xiang Zhou, Hong Shang, Zijian Zhan +3
Deep unrolled models (DUMs) have become the state of the art for accelerated MRI reconstruction, yet their robustness under domain shift remains a critical barrier to clinical adop…
(How) Can Transformers Predict Pseudo-Random Numbers?
Tao Tao, Darshil Doshi, Dayal Singh Kalra +2
Transformers excel at discovering patterns in sequential data, yet their fundamental limitations and learning mechanisms remain crucial topics of investigation. In this paper, we s…
Towards Distributed Neural Architectures
Aditya Cowsik, Tianyu He, Andrey Gromov
We introduce and train distributed neural architectures (DNA) in vision and language domains. DNAs are initialized with a proto-architecture that consists of (transformer, MLP, att…
3D-LLaVA: Towards Generalist 3D LMMs with Omni Superpoint Transformer
Jiajun Deng, Tianyu He, Li Jiang +3
Current 3D Large Multimodal Models (3D LMMs) have shown tremendous potential in 3D-vision-based dialogue and reasoning. However, how to further enhance 3D LMMs to achieve fine-grai…
Universal Sharpness Dynamics in Neural Network Training: Fixed Point Analysis, Edge of Stability, and Route to Chaos
Dayal Singh Kalra, Tianyu He, Maissam Barkeshli
In gradient descent dynamics of neural networks, the top eigenvalue of the loss Hessian (sharpness) displays a variety of robust phenomena throughout training. This includes early…