Probability-valued probes with compositional renormalization for local and compositional measurements in quantum field theory

In the local positive formalism — a spacetime-local, compositional generalization of quantum theory in which measurement processes are probes associated with spacetime regions — the modulus-square construction of Oeckl and Zampeli yields probes that measure expectation values of quadratic field observables with a renormalization scheme that commutes with spacetime composition. The paper's acknowledged open questions were: full outcome statistics rather than expectation values only; the restriction of the construction to a special class of observables; the correspondence with apparatus-based measurement models; and the scope of the treatment — free bosonic field theory in globally hyperbolic spacetimes, without gauge symmetries. The authors' own follow-up (Oeckl, Phys. Rev. D 113, 085002 (2026)) resolved the linear-observable part of the outcome-statistics question: it constructed probability-valued, spacetime-compositional, causally transparent probes via a regularized spectral decomposition, and showed that the spectral regulator cannot be removed at the level of probes while physical probability quotients do converge. The open core that survives is the marriage of probability-valued probes with renormalization for singular observables such as the point-localized $\phi^2(x)$ and the stress-energy tensor, together with the acknowledged directions of extending beyond the free bosonic, gauge-free setting and deriving the probes from explicit ancilla or detector models.

Background

How should a measurement be described in relativistic quantum field theory (QFT)? In the standard formulation of quantum theory, a general measurement is represented by an instrument, that is, a family of Kraus operators $K_i$ (operator factors whose weighted norms provide the outcome probabilities $tr(K_i\sigma K_i^\dagger)$ for a state $\sigma$) applied to the whole system at a single instant. Sorkin showed in 1993 that this picture is incompatible with relativity: measurements localized in a spacetime region, if described by the nonrelativistic rules, can be arranged to signal faster than light (the 'impossible measurements' problem). Since then, several routes to a local measurement theory for QFT have been pursued: detector models of the Unruh-DeWitt type (a pointlike quantum system following a timelike worldline and linearly coupled to the field along it), apparatus-based algebraic frameworks such as that of Fewster and Verch, in which a separate probe field interacts with the system field in a localized spacetime region and is read out later, and histories-based approaches using quantum temporal probabilities; the landscape is reviewed by Papageorgiou and Fraser.

The source paper, by Oeckl and Zampeli, takes a different route: the local positive formalism (PF), an axiomatic spacetime-local generalization of quantum theory in which the central object is the probe — the spacetime generalization of a quantum operation, associated with a spacetime region rather than an instant. Probes compose whenever their spacetime regions are disjoint (no causal orderability or separating hypersurfaces are required), and the formalism retains a consistent probability interpretation. What the PF by itself does not supply is a prescription for constructing the probe that implements the measurement of a given classical observable in a given region; this is a quantization problem for measurement, and it had been open.

Oeckl and Zampeli's modulus-square construction addresses it for an important class of observables. If an observable can be written as a modulus square $\Lambda=A\bar A$ of a simpler (e.g. linear) observable $A$ on the space of field configurations, one inserts $A$ into one branch and $\bar A$ into the other branch of the double path integral (the Schwinger-Keldysh, or closed-time-path, formalism used for expectation values in nonequilibrium QFT). The resulting probe $\mathcal P[A|A]$ is positive, spacetime-local, compositional, and reproduces standard single-measurement results. For quadratic observables such as the point-localized $\phi^2(x)$, which are singular (their vacuum two-point functions diverge at coinciding points), the construction needs renormalization: the prescription subtracts $r(D,D)$ — the observable propagator evaluated on the same observable, the divergent coincidence limit — times the null probe (free evolution). This subtraction is uniquely fixed by requiring semiclassical behavior: a single measurement on a coherent state (the field state corresponding to a classical solution) must reproduce the classical expectation value. It reduces to normal ordering for a single instantaneous measurement, and it 'commutes with composition': renormalizing a composite measurement equals composing the renormalized individual probes. Relativistic causality was demonstrated in an explicit example — a non-selective Sorkin-type 'kick' probe composed with the renormalized quadratic probe satisfies the required locality identity outside the kick's causal future — but because these probes encode expectation values and no corresponding non-selective probe was constructed, a general causal-transparency theorem (no superluminal signaling) was not available for the scheme at that stage; it was later proved in the follow-up for regularized spectral probes of linear observables. All explicit computations are for free bosonic (Klein-Gordon) fields in globally hyperbolic spacetimes (spacetimes admitting Cauchy hypersurfaces) with a chosen global vacuum.

The construction has one built-in limitation: the probes encode expectation values and their multiplicative composites, not outcome probabilities, and the latter cannot be recovered from the former without something like a spectral decomposition. Explicit spectral decompositions of field operators were obtained only recently by Oeckl in the operator setting (J. Math. Phys. 66, 042302 (2025)), with a Gaussian regularization. The paper suggested transferring this to the local PF through sharply peaked Gaussian observables,

$$H_D^\epsilon(q)(\phi)=\frac{1}{\sqrt{\pi}\epsilon}\exp\left(-\frac{1}{\epsilon^{2}}(D(\phi)-q)^{2}\right),$$

whose probe $\mathcal P[H_D^\epsilon(q)|H_D^\epsilon(q)]$ approximately asks whether the linear observable $D$ takes the value $q$, with the exact spectral correspondence obtained in the limit $\epsilon\to0$; integrating over $q$ with weight functions (the value $q$, or characteristic functions of outcome sets) was proposed to recover expectation values and probabilities, including a proper non-selective probe (a probe in which the outcome is not recorded, obtained by summing over all outcomes). The paper left the systematic development of this route, its convergence, and its compatibility with compositionality and causality open. Its acknowledged limitations were, in addition, the restriction to a special class of observables, the absence of a connection with apparatus-based models, and the scope of the treatment — free bosonic field theory in globally hyperbolic spacetimes, without gauge symmetries.

Problem Statement

Complete the local and compositional measurement framework for quantum field theory so that it yields full joint outcome statistics — not only expectation values — for the observables that require renormalization. The objective is a measurement theory that is simultaneously spacetime-local, compositional, causally consistent, and probabilistically complete. The central open problem, explicitly deferred in both source papers, is:

1. Construct, in the local positive formalism, probability-valued probes for singular, renormalization-requiring observables — at minimum the point-localized quadratic observables $\phi^2(x)$ and the stress-energy tensor $T_{\mu\nu}$ in free scalar field theory (these two constitute the required minimum; treating further observables in the class is a welcome but non-required extension). The construction should combine the regularized spectral (equivalently, Gaussian modulus-square) probe construction — or an equivalent outcome-probe construction — with a renormalization scheme such that: (a) positivity and the probability interpretation are preserved, with outcome probabilities and expectation values obtained by quotient normalization against a proper non-selective probe — that is, by dividing the outcome-selective probe by the non-selective probe that sums over all outcomes, which yields the conditional probability of an outcome given that the measurement was performed; (b) renormalization commutes with spacetime composition; (c) causal transparency (no superluminal signaling) holds; and (d) the joint limits removing the spectral regulator ($\epsilon\to0$ in the Gaussian construction) and the ultraviolet point-splitting regulator (in which products of fields are evaluated at slightly separated spacetime points before the coincidence limit is taken) are characterized — either proven to exist for the physical probabilities, or the obstruction precisely identified.

The source papers acknowledge two further open directions, which are noted here as context rather than as required components and are not needed to resolve the task:

2. Extend the probability-valued probe construction beyond the scope of the current treatment: free bosonic field theory in globally hyperbolic spacetimes, without gauge symmetries, and assuming a chosen global vacuum.

3. Provide the systematic correspondence with apparatus-based measurement models: derive the probes from explicit ancilla or detector field models, and characterize which apparatus models realize a given probe.

Two further questions are raised by the construction but not posed in the source papers; they are noted here as context rather than as required components: whether the renormalization prescription that semiclassicality uniquely fixes for quadratic observables extends uniquely and without ambiguity to more general observable classes (higher polynomials, composite operators), or whether inequivalent schemes satisfying the same requirements (semiclassicality, compositionality of renormalization, relativistic causality) can be exhibited; and rigorous functional-analytic or algebraic foundations for the probes, including control of point-splitting limits and convergence of the coherent-state resolutions and basis sums on which the formal constructions rest.

Boundary note: probability-valued probes for linear observables and functions of several linear observables in free bosonic theory — including spacetime compositionality, causal transparency, and the analysis of the $\epsilon\to0$ regulator limit — have already been constructed in the literature (Oeckl, Phys. Rev. D 113, 085002 (2026)) and are not part of the open problem; an answer restricted to that resolved case does not address the problem.

References

1. R. Oeckl, A. Zampeli, Towards local and compositional measurements in quantum field theory, arXiv:2505.10968 (2025), DOI 10.48550/arXiv.2505.10968, https://arxiv.org/abs/2505.10968
2. R. Oeckl, Causal measurement in quantum field theory: spacetime, Phys. Rev. D 113, 085002 (2026), arXiv:2511.06566, DOI 10.1103/5gl8-3j6p, https://arxiv.org/abs/2511.06566
3. R. Oeckl, Spectral decomposition of field operators and causal measurement in quantum field theory, J. Math. Phys. 66, 042302 (2025), arXiv:2409.08748, DOI 10.1063/5.0245368, https://arxiv.org/abs/2409.08748
4. R. Oeckl, A positive formalism for quantum theory in the general boundary formulation, Found. Phys. 43, 1206-1232 (2013), arXiv:1212.5571, DOI 10.1007/s10701-013-9741-5, https://arxiv.org/abs/1212.5571
5. R. Oeckl, A local and operational framework for the foundations of physics, Adv. Theor. Math. Phys. 23, 437-592 (2019), arXiv:1610.09052, DOI 10.4310/ATMP.2019.v23.n2.a4, https://arxiv.org/abs/1610.09052
6. R. Sorkin, Impossible measurements on quantum fields, in Directions in General Relativity, Vol. 2 (eds. B. L. Hu, T. A. Jacobson), Cambridge University Press, Cambridge, 1993, pp. 293-305, DOI 10.1017/cbo9780511524653.024, https://doi.org/10.1017/cbo9780511524653.024
7. C. J. Fewster, R. Verch, Quantum fields and local measurements, Commun. Math. Phys. 378, 851-889 (2020), arXiv:1810.06512, DOI 10.1007/s00220-020-03800-6, https://arxiv.org/abs/1810.06512
8. M. Papageorgiou, D. Fraser, Eliminating the 'impossible': Recent progress on local measurement theory for quantum field theory, Found. Phys. 54, 26 (2024), arXiv:2307.08524, DOI 10.1007/s10701-024-00756-8, https://arxiv.org/abs/2307.08524
9. C. Anastopoulos, B.-L. Hu, K. Savvidou, Quantum field theory based quantum information: Measurements and correlations, Ann. Phys. 450, 169239 (2023), arXiv:2208.03696, DOI 10.1016/j.aop.2023.169239, https://arxiv.org/abs/2208.03696
10. F. Daem, L. Ballesteros Ferraz, A. Zampeli, A. Matzpin, Sensing relativistic quantum fields with minimally perturbing local measurements, arXiv:2607.17920 (2026), https://arxiv.org/abs/2607.17920
11. R. Simmons, M. Papageorgiou, M. Christodoulou, C. Brukner, Factorisation conditions and causality for local measurements in QFT, arXiv:2511.21644 (2025), https://arxiv.org/abs/2511.21644
12. R. Simmons, Causality is rare: some topological properties of causal quantum channels, arXiv:2603.25315 (2026), https://arxiv.org/abs/2603.25315
13. J. Polo-Gomez, T. Rick Perche, E. Martin-Martinez, State updates and useful qubits in relativistic quantum information, arXiv:2506.18906 (2025), https://arxiv.org/abs/2506.18906
14. P. Dall'Olio, J. A. Zapata, Towards relational foundations for spacetime quantum physics, arXiv:2506.17321 (2025), https://arxiv.org/abs/2506.17321
15. N. Koliopoulou, C. Anastopoulos, N. Savvidou, Quantum field theory measurements for relativistic particles, Phys. Rev. D 114, 045011 (2026), arXiv:2602.14175, DOI 10.1103/tvd2-c113, https://arxiv.org/abs/2602.14175
16. J. Mandrysch, M. Navascues, Quantum field measurements in the Fewster-Verch framework, arXiv:2411.13605 (2024), https://arxiv.org/abs/2411.13605
17. J. Mandrysch, R. Simmons, M. Navascues, Causality and realizability of local operations in quantum field theory, arXiv:2607.12976 (2026), https://arxiv.org/abs/2607.12976