Some physical objects are hardly accessible to direct experimentation. It is then desirable to infer their properties based solely on the interactions they have with systems over which we have control. In this spirit, here we introduce schemes for assessing the non-classicality of the inaccessible objects as characterised by quantum discord. We consider two probes individually interacting with the inaccessible object, but not with each other. The schemes are based on monitoring entanglement dynamics between the probes. Our method is robust and experimentally friendly as it allows the probes and the object to be open systems, makes no assumptions about the initial state, dimensionality of involved Hilbert spaces and details of the probe-object Hamiltonian. We apply our scheme to a membrane-in-the-middle optomechanical system, to detect system-environment correlations in open system dynamics as well as non-classicality of the environment, and we foresee potential benefits for the inference of the non-classical nature of gravity.What should be known about an inaccessible object to conclude that it is "not classical"? Here we show, inspired by quantum communication scenarios, that it is sufficient to verify whether such object can be used to increase quantum entanglement between remote probeparticles that individually interact with it, but are not directly coupled to each other.Specifically, we prove that such gain in quantum entanglement is only possible if, during its evolution, the object shares with the probes quantum correlations in the form of quantum discord [1][2][3][4][5]. In turn, the presence of quantum discord between the probes and the object entails a non-classical feature of the object itself. According to the definition of discord, two or more subsystems share quantum correlations if there is no von Neumann measurement on one of them that keeps the total state unchanged. This can only happen when non-orthogonal (indistinguishable) states are involved in the description of the physical configuration of the measured subsystem. This indistinguishability is the non-classical feature that we aim to detect. We formulate analytical criteria revealing such non-classicality based on operations performed only on the probes, and without any detailed modelling of the inaccessible object in question.We emphasise that the non-classicality is revealed under a set of minimal assumptions. Namely: (i) The object may remain inaccessible at all times, i.e. it needs not be directly measured. In particular its quantum state and Hilbert space dimension can remain unknown throughout the whole assessment. Our method is thus valid when the object is an elementary system or an arbitrarily complex one; (ii) The details of the interaction between the object and the probes may also remain unspecified; (iii) Every party can be open to its own local environment. These properties make our method applicable to a large number of experimentally relevant situations.We demonstrate the revealing power of our criteria for non-classicality ...

The key requirement for quantum networking is the distribution of entanglement between nodes. Surprisingly, entanglement can be generated across a network without direct transfer-or communication-of entanglement. In contrast to information gain, which cannot exceed the communicated information, the entanglement gain is bounded by the communicated quantum discord, a more general measure of quantum correlation that includes but is not limited to entanglement. Here, we experimentally entangle two communicating parties sharing three initially separable photonic qubits by exchange of a carrier photon that is unentangled with either party at all times. We show that distributing entanglement with separable carriers is resilient to noise and in some cases becomes the only way of distributing entanglement through noisy environments.

We classify protocols of entanglement distribution as excessive and non-excessive ones. In a non-excessive protocol, the gain of entanglement is bounded by the amount of entanglement being communicated between the remote parties, while excessive protocols violate such bound. We first present examples of excessive protocols that achieve a significant entanglement gain. Next we consider their use in noisy scenarios, showing that they improve entanglement achieved in other ways and for some situations excessive distribution is the only possibility of gaining entanglement.