Bell non-locality is a fundamentally quantum phenomenon, arising when we measure two quantum systems in a space-like separated manner. The non-relativistic description of such an experiment prescribes the tensor product of two Hilbert spaces corresponding to the two systems. The relativistic description, however, prescribes a single Hilbert space with the measurements of the two systems commuting. Tsirelson thought that these two descriptions lead to equivalent experimental predictions in Bell non-locality. However, a recent breakthrough result showed that this is not the case: the relativistic description is strictly more general in principle [1]. The proof of this result is not constructive, leaving open the question: is the relativistic description more general in practice? That is, are the predictions showing this discrepancy actually physical? If so, how could we test which predictions are the correct ones? A possible resolution to these questions may arise from algebraic quantum field theory or group algebra representations.
[1] Ji, Natarajan, Vidick, Wright and Yuen, Communications of the ACM, 64(11), 131 - 138 (2021)
In this talk I will illustrate the "FV" theory of measurement in QFT, focussing on a specific model in which the polarisation state of a QFT can be measured. Measurements can be made for two polarisation channels and at multiple locations in spacetime. In particular, this model allows one to compute probabilities for joint measurements at spacelike separated locations, allowing discussion of Bell inequalities and the interpretation of state updates following measurement. No previous acquaintance with the measurement framework is required.
The talk is based on the preprint Coupled Proca theories: Green-hyperbolicity, quantization and applications to polarization measurement, CJ Fewster and CKM Klein, arXiv:2511.11348, and further work in progress. The measurement framework was introduced in Quantum fields and local measurements, CJ Fewster and R Verch, Commun. Math. Phys. 378 (2020) 851-889 arXiv:1810.06512.
The interest in local von Neumann algebras in relativistic quantum field theory arises naturally, as they correspond to the algebras of local observables represented on the vacuum Fock space. In particular, Haag duality ensures that these algebras are maximal in the sense that they contain all local observables. Within this framework, restricting the vacuum state to a local von Neumann algebra allows one to use modular theory and, in turn, introduce the notion of Araki–Uhlmann relative entropy. This provides a natural definition of relative entropy in QFT, where the absence of a trace—stemming from the structure of local von Neumann algebras—precludes the use of standard quantum mechanical tools. In this talk, I will review these concepts in the setting of a free real scalar quantum field theory. Time permitting, I will also present results on relative entropy in fermionic QFT and discuss how Haag duality is modified when the vacuum state is replaced by a KMS state.
This work is based on collaborations with Albert Much, Leonardo Sangaletti, and Rainer Verch.
We discuss limitations of local measurement schemes to describe measurements in quantum field theory in probing relational observables, which naturally arise from requirements such as diffeomorphism invariance. To overcome these limitations, we introduce notions of relative measurements performed in relation to operational quantum reference frames, and discuss their associated observable quantities and state updates.
In this talk, I present semiclassical gravity and explain its impact on unresolved RQI questions. I argue that, despite its complexity, it is worth understanding this theory in full conceptual and mathematical detail. How to make progress in this task is the second point I touch on.
I recall my 1998 matter-gravity entanglement hypothesis according to which the physical entropy of a quantum gravitational closed system is to be identified with its matter-gravity entanglement entropy and I explain how this is motivated by the so-called thermal atmosphere puzzle. Under the further identification of 'information' with negative entropy, one aspect of the information loss puzzle then becomes a special case (the case where the closed system consists initially of a collapsing ball of matter in an otherwise empty universe) of the second law puzzle: Why does the entropy of a closed system increase? Our proposed answer is that if (low-energy) quantum gravity is a conventional (unitary) quantum theory, then one would expect that, if it is initially low, the matter-gravity entanglement entropy of the quantum state of a closed system would increase for all time, due to interactions between matter and gravity. I also recall my more recent arguments (https://arxiv.org/abs/2206.07445) that, if this is on the right track, then one would expect the final total state of black hole evaporation for an initially large black hole formed by collapse to (be pure and) consist mainly of photons entangled with gravitons.
Bell inequalities are an important tool to distinguish between quantum mechanics and local hidden-variable theories. Since the first work by Bell, different experimental designs have been put forward to demonstrate the violation of Bell inequalities in way that is, under certain assumptions, impossible for hidden-variable theories. In this talk, I will describe how such an experimental design can be modelled within the "FV" measurement framework and under which circumstances the same violation of the corresponding Bell inequality as for the quantum-mechanical version can be observed.
The talk is based on the preprint Coupled Proca theories: Green-hyperbolicity, quantization and applications to polarization measurement, CJ Fewster and CKM Klein, arXiv:2511.11348, and further work in progress.
In this talk, I will present an ongoing work with D. Janssen and K. Rejzner on the gluing problem in quantum electromagnetism.
Put simply, the question is: can we reconstruct the global physics on a spacetime knowing the physics in its subregions?
For gauge theories and spacetimes with non-trivial topologies and boundaries, the problem is theoretically challenging; from an experimental point of view, phenomena such as the Aharonov-Bohm effect motivates an in depth analysis.
Using a relative differential cohomology approach, we construct a space of field configurations for electromagnetism that is compatible with gluing, and we discuss their dynamics. We derive an algebra of observables, roughly corresponding to smeared Wilson lines; in this setting, we prove that given a (nice enough) decomposition of spacetime, the algebra of observables on the full spacetime can be reconstructed from the algebras on its subregions. Eventually, we discuss quantisation and its superselection sectors, and their link with relativistic quantum information experiments.
TBA
At the heart of both Wigner's friend paradoxes and black hole puzzles lies the question of unitarity. In Wigner’s friend setups, sealed-lab measurements are modeled unitarily, probing the measurement problem. In black hole physics, the unitarity problem concerns information preservation in evaporation. In this talk, I discuss a refined version of the Frauchiger--Renner paradox and extend a recent analogy between Wigner's friend and black hole paradoxes exposed by Hausmann and Renner [arXiv:2504.03835v1], by constructing new paradoxes that merge black hole physics with extensions of Wigner’s friend scenarios into a unified argument. I shortly discuss implications of these works and highlight subtleties in black hole puzzles.