Mukherjee, Amit

We introduce a stronger form of quantum nonlocality, termed local subset unidentifiability, that arises from the limitation of spatially separated parties to perfectly identify a subset of mutually orthogonal multipartite quantum states, randomly chosen from a larger known set, using local operations and classical communication (LOCC). We show that this nonlocality is stronger than other existing forms of quantum nonlocality, such as local indistinguishability and local unmarkability. If more than one multipartite states from a locally indistinguishable set are distributed between spatially separated parties in a sequentially ordered fashion, then they may or may not mark which state is which using LOCC. However, we show that even when the parties cannot mark the states, they may still locally identify the particular states given to them, though not their order - i.e., they can identify the elements of the given subset of states. Then we prove the existence of such subsets that are not even locally identifiable, thereby manifesting a stronger nonlocality. We also present the genuine version of this nonlocality - genuine subset unidentifiability - where the provided subset remains unidentifiable unless all the parties come together in a common location and perform global measurements. We anticipate potential applications of this nonlocality for future quantum technologies. We discuss one such application in a certain secret password distribution protocol, where this nonlocality outperforms its predecessors as a resource.

The extension of point-to-point communication to multi-node configurations has significant applications in internet and telecommunication networks. Quantum resources promise notable advantages in such settings. Here, we demonstrate a novel quantum advantage in simulating multiple access channels (MACs)-a common network configuration where multiple distant senders transmit messages to a single receiver (e.g. the uplink from several mobile phones to a server). Specifically, we show that qubit transmission outperforms its classical counterpart, even when the latter is supplemented with classical shared randomness. Remarkably, unlike the seminal quantum superdense coding protocol, this advantage is achieved without any pre-shared entanglement between the senders and the receiver-a feat prohibited by Holevo and Frenkel-Weiner no-go theorems in the one-sender-one-receiver scenario. The receiver’s ability to simultaneously decode quantum systems from multiple senders underpins this distinct advantage in the MAC setup. Some of our MAC designs are inspired by constructs in quantum foundations, such as the Pusey-Barrett-Rudolph theorem and ‘quantum nonlocality without entanglement’. Beyond network applications, this quantum advantage reveals a deeper connection to ‘quantum nonlocality without inputs’ phenomenon and suggests potential for semi-device-independent certification of entangled measurements. © 2025 The Author(s). Published by IOP Publishing Ltd on behalf of the Institute of Physics and Deutsche Physikalische Gesellschaft.