Six telescopes atop a mountain capture light from a distant galaxy. Each sees its own copy of that light, and to get a sharp image, the signals must be combined. But how? Usually, each telescope measures its own light separately, and then the results are added up. It's like an orchestra where musicians play individually and the recordings are spliced together: the overall rhythm is lost, and the music sounds worse.
Scientists compared two approaches: sequential measurement and collective measurement. It turned out that sequential measurement suffers from incompatibility: estimates of different light properties interfere with each other, and the more telescopes, the worse it gets. Collective measurement removes this problem. The gain in sensitivity grows as the square root of the number of stations: for six telescopes, that's nearly two and a half times. In practice, this means future networks could be several times more compact while seeing just as sharply. An orchestra that rehearses together hears every note more precisely, and telescopes that measure light jointly see farther and clearer.
The paper used GPT 5.6 Solar to formalize and prove key metrological bounds, indicating the growing role of AI in theoretical physics.
The idea of collectively measuring copies of a quantum state echoes the concept of a quantum computer processing many states simultaneously, as in Greg Egan's novel 'Quarantine'.