{"id":"arxiv/2607.21584","url":"https://arxiv.org/abs/2607.21584","tags":["quant-ph"],"score":0,"title":"Flow-based Phase-space Tomography of Continuous-variable Quantum States","domain":"quant-ph","source":"arxiv","authors":["Owen Dugan","Rumen Dangovski","Peter Y. Lu","Di Luo"],"reviews":0,"subject":"quant-ph","summary":"Continuous-variable quantum state tomography is limited by the cost of resolving non-Gaussian structure in high-dimensional phase space. We introduce QST-Flow, a quantum state tomography framework via flow-based generative modeling that represents experimentally accessible phase-space quasiprobability distributions with normalized, samplable neural densities rather than a truncated density matrix. The framework has two variants: QST-QFlow models the positive Husimi-$Q$ function with a single normalizing flow, while QST-WFlow models sign-changing Wigner functions as a trainable difference of…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-07-23","reactions":{}}