Coincidence postselection for genuine multipartite nonlocality: Causal diagrams and threshold efficiencies

• DOI: 10.48550/arxiv.2207.13579
• Publication Date: 2022-01-01
• Authors: Valentin Gebhart; Augusto Smerzi
• Keywords: Postselection; Multipartite

Genuine multipartite nonlocality (GMN), the strongest form of that describes fully collective nonlocal correlations among all experimental parties, can be observed when different distant parties each locally measure a particle from shared entangled many-particle state. For demonstration GMN, experimentally statistics are typically postselected: Events for which some do not detect must discarded. This coincidence postselection generally leads to detection loophole invalidates proper demonstration. In this work, we address how close in demonstrations and GMN. We first show if number detected particles is conserved, i.e., using ideal noiseless devices, one employ causal diagrams no-signalling principle prove cannot create any loophole. Furthermore, realistic devices with finite efficiencies, general Bell inequality sharpened such its new version still valid after measurement data. case, there threshold efficiencies that, surpassed experiment, lead possibility demonstrate GMN without opening Our results imply genuine $N$-partite generated $N$ independent sources even allowing non-ideal detectors.