Early-Signal Detector

method · first seen 10 Oct 2026

weight-only quantization

Reducing model size by quantizing only the weights of a neural network to low bit widths while keeping activations in higher precision.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

vector quantizationultra-low-bit quantizationmixed-precision quantizationGPTQFP-INT GEMM accelerator