Ruling out Bipartite Nonsignaling Nonlocal Models for Tripartite Correlations

TL;DR

This paper develops a rigorous framework to distinguish genuine tripartite nonlocal correlations from those simulatable by bipartite nonsignaling models, confirming quantum violations of certain Bell-like inequalities.

Contribution

It introduces a new framework for analyzing tripartite correlations, enabling the derivation of Bell-type inequalities that quantum systems can violate, thus ruling out bipartite-only explanations.

Findings

Framework confirms no-signaling properties of bipartite-simulated correlations

Derives Bell-inequality constraints for tripartite correlations

Quantum systems can violate these inequalities, indicating genuine tripartite nonlocality

Abstract

Many three-party correlations, including some that are commonly described as genuinely tripartite nonlocal, can be simulated by a network of underlying subsystems that display only bipartite nonsignaling nonlocal behavior. Quantum mechanics predicts three-party correlations that admit no such simulation, suggesting there are versions of nonlocality in nature transcending the phenomenon of bipartite nonsignaling nonlocality. This paper introduces a rigorous framework for analyzing tripartite correlations that can be simulated by bipartite-only networks. We confirm that expected properties of so-obtained correlations, such as no-signaling, indeed hold, and show how to use the framework to derive Bell-inequality-type constraints on these correlations that can be robustly violated by tripartite quantum systems. In particular, we use this framework to rederive a version of one such constraint previously described in a paper of Chao and Reichardt (arXiv:1706.02008).