{"id":"arxiv/2610.07126","url":"https://arxiv.org/abs/2610.07126","tags":["math.NT","cs.DC","cs.MS"],"score":0,"title":"CUDA-MPQS: A GPU-Resident Self-Initializing Quadratic Sieve, and the Factorization of RSA-155","domain":"math.NT","source":"arxiv","authors":["Fabian Januszewski","Christoph Heinrichs"],"reviews":0,"subject":"math.NT","summary":"The quadratic sieve (QS) is an irregular, cache-hostile, branch-heavy integer factorization algorithm; prior GPU work accelerates individual stages of it. We present \\swname{}, a self-initializing quadratic sieve in which \\emph{every} stage --- polynomial initialization, sieving, relation post-processing, large-prime matching, \\GFtwo{} matrix construction, Block Wiedemann linear algebra and the square root --- executes on the GPU, the CPU restricted to orchestration, setup and I/O, the sieve window 99.9\\% GPU-busy with no device-wide or event synchronization and no synchronous copy. With it…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-05","reactions":{}}