{"id":"arxiv/2611.16995","url":"https://arxiv.org/abs/2611.16995","tags":["math.PR","stat.ML"],"score":0,"title":"Entire sig-jump-diffusions, their expected signature and applications to score-based generative modeling","domain":"math.PR","source":"arxiv","authors":["Christa Cuchiero","Francesca Primavera","Sara Svaluto-Ferro"],"reviews":0,"subject":"math.PR","summary":"We develop score-based generative models for time series and standard finite-dimensional data, building on the analytical computation of the expected signature of a new class of jump-diffusions, called entire sig-jump-diffusions. These are semimartingales whose characteristics are given by linear combinations of potentially infinitely many components of their own signature. We show that their expected signature can be computed by solving linear ODEs in the extended tensor algebra, providing a tractable foundation for learning procedures. Specifically, the learning of a generalized score…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-11-12","reactions":{}}