Comment on "Geometry of the quantum set on no-signaling faces"

TL;DR

This paper corrects a previous claim by showing that the Almost Quantum set can reproduce certain boundary points of the quantum set, clarifying the reliability of numerical methods in this context.

Contribution

It provides analytical proofs demonstrating that the Almost Quantum set can indeed reproduce specific boundary points previously thought incompatible.

Findings

Almost Quantum set reproduces boundary points of the quantum set

Numerical issues can lead to incorrect conclusions about correlation sets

Analytical proofs confirm the capabilities of the Almost Quantum set

Abstract

In Ref. [1] the authors claim that the Almost Quantum set of correlations cannot reproduce two points on the boundary of the quantum set of correlations. This claim is incorrect. The underlying issue is that the associated SDP is not strictly feasible, which makes the numerical solvers give unreliable answers. We give analytical proofs that both points are indeed reproduced by Almost Quantum.