IEEE Spectrum on MSN

Better hardware could turn zeros into AI heroes

Sparse computing enables leaner, faster AI ...
Hosted on MSN

Stanford chip and Apple tools signal AI efficiency shift

Stanford researchers unveiled Onyx, a programmable chip that accelerates both sparse and dense AI computations, promising major energy and speed gains. Apple is reportedly adding three AI-powered ...
Scientific Research Publishing

Edge-Centric Generative AI: A Survey on Efficient Inference for Large Language Models in Resource-Constrained Environments ()

Edge-Centric Generative AI: A Survey on Efficient Inference for Large Language Models in Resource-Constrained Environments ...
IEEE

Scalable Sparse Transformer Accelerator with In-Memory Butterfly Zero Skipper and Local Attention Reusable Engine for Semi-Structured-Pruned NN

Abstract: Transformer-based large language models (LLMs) have achieved unprecedented advances across diverse AI tasks. However, their execution remains power-hungry, primarily due to the rapidly ...
IEEE

A Sparse Deep Neural Network Inference Accelerator

Abstract: This paper proposes a sparse Deep Neural Network (DNN) inference accelerator architecture that can be used for a reconfigurable edge computing platform that improves computational efficiency ...