Justin McDaniel has developed a cult following for getting his students to read — as long as they follow his rules.

Abstract: Accelerating matrix multiplication is crucial to achieve high performance in many application domains, including neural networks, graph analytics, and scientific computing. These ...

CUDA-L2 is a system that combines large language models (LLMs) and reinforcement learning (RL) to automatically optimize Half-precision General Matrix Multiply (HGEMM) CUDA kernels. CUDA-L2 ...

Abstract: This work presents a mathematical procedure for solving coupling matrix synthesis problems with arbitrary topologies. A discussion on the properties of arbitrary response-preserving ...

The Long Multiplication Benchmark evaluates Large Language Models (LLMs) on their ability to handle and utilize long contexts to solve multiplication problems. Despite long multiplication requiring ...

In the year after surgery for primary congenital glaucoma, rigid gas-permeable contact lenses led to significantly better visual outcomes than spectacles, according to a study published in JAMA ...