trends.barnas.net

method · first seen 6 Oct 2026

low-precision floating-point arithmetic

Also: mixed precision, reduced precision, mixed-precision computing, multi-precision computing, mixed-precision arithmetic

Number formats and computation using fewer bits than standard floats to speed up and shrink numerical workloads.

Mentions per week

0319 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

signal-to-noise ratio precision analysisspatial predictionGaussian conditional simulationvision-language modelbfloat16conjugate gradientvectorized programmingdouble-float emulationWENO reconstruction