[briankirby]

I believe they're referring to this proposal by Gottesman [0]. I will try to explain in layman's terms what is happening: the simplest telescope array you can imagine is just two telescopes separated by what is called the baseline. The telescopes could either just detect light from a star independently and then autocorrelate these signals afterwards, or they could interfere their signals before detection. The latter is called an interferometric telescope array and has a resolution advantage over the former, simpler, solution. The length of the baseline is an important factor in determining the size of the objects you can resolve with such a telescope, and the goal is generally to make it as large as possible. Unfortunately, as the baseline becomes larger the signals from the two telescopes are likely to become distorted or lost before you can interfere them due to the distance you have to send one to reach the other. The Gottesman paper offers one solution to this, which is to use a quantum repeater (essentially quantum teleportation) to send one of the telescope's signals to the other telescope without any loss or degradation, in principle allowing for arbitrarily long baselines in telescopes.

[0] https://arxiv.org/abs/1107.2939