Abstract: Mixed-precision computation, which uses multiple different precision in a single code, is being studied to increase computational speed and energy efficiency. It typically uses the IEEE ...

Abstract: Numerous studies have proposed hardware architectures to accelerate sparse matrix multiplication, but these approaches often incur substantial area and power overhead, significantly ...

This is a benchmarking tool for Qdrant's sparse vector implementation using the NeurIPS 2023 datasets. This task is based on the common MSMARCO passage retrieval dataset, which has 8,841,823 text ...

Data centers face a conundrum: how to power increasingly dense server racks using equipment that relies on century-old technology. Traditional transformers are bulky and hot, but a new generation of ...