Towards relativistic generalization of collapse models
Anirudh Gundhi, Lajos Diósi, Matteo Carlesso · 2025
Read Towards relativistic generalization of collapse models by Anirudh Gundhi, Lajos Diósi, Matteo Carlesso (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-06954.txt.
Introduction
Anirudh Gundhi, Lajos Diósi, and Matteo Carlesso push collapse models toward relativity in this two thousand twenty-five paper. Collapse models say quantum superpositions die on their own, solving the measurement puzzle in a testable way. Experiments keep tightening their allowed settings. No full relativistic version exists yet. Past tries broke causality, leaked endless energy, or spawned particles from empty space.
The authors reshape the collapse equation with a local field operator as the collapsing agent. The driving noise remembers its past and respects Lorentz symmetry. Their build dodges the old traps and matches the known theory at low speeds. A concrete choice for the noise pattern is discussed. Foundations of physics readers are the audience.
Catalog record
| CREATOR | Anirudh Gundhi, Lajos Diósi, Matteo Carlesso |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-06954
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Towards relativistic generalization of collapse models Anirudh Gundhi, Lajos Diósi, Matteo Carlesso Abstract Spontaneous collapse models provide a possible, testable solution to the quantum measurement problem. While experiments are providing increasingly stronger bounds on their parameters, a full-fledged relativistic extension is still missing. Previous attempts have encoutered different obstacles, such as violation of microcausality, infinite energy rate, and particle production from vacuum. Here, we propose generalization of the collapse master equation that is characterized by a local field collapse operator and a non-Markovian noise with a Lorentz invariant correlation. Our construction is able to overcome previously encountered problems and has the desirable properties in the non relativistic limit. A specific choice of the noise correlation function is also introduced and discussed. Non-relativistic spontaneous wavefunction collapse (or simply, collapse) models provide a possible coherent solution to the well-known quantum measurement problem Bassi and Ghirardi 2003 ; Bassi et al. 2013 . They modify the standard, linear unitary Schrödinger equation by adding non-linear and stochastic terms, which impose the statevector localization (or collapse) into one eigenstate of a suitable collapse operator. The outcome of a subsequent measurement would provide the corresponding eigenvalue. Different realizations of the stochastic process impose different outcomes, which are distributed according to the Born rule. While some collapse models, such as the Continuous Spontaneous Localization (CSL) Pearle 1989 ; Ghirardi et al. 1990 and the Diosi-Penrose (DP) Diósi 1987 models, are subject to continuous experimental and theoretical investigation Arndt and Hornberger 2014 ; Bilardello et al. 2016 ; Carlesso et al. 2016 ; Vinante et al. 2017 ; Toroš and Bassi 2018 ; Vinante et al. 2020 ; Donadi et al. 2021 ; Carlesso et al. 2022 ; Figurato et al. 2024 ; Piscicchia et al. 2024 ; Altamura et al. 2025 ; Piccione and Bassi 2025 , there is still an ongoing debate concerning their possible relativistic generalization Dove and Squires 1996 ; Breuer and Petruccione 1998 ; Breuer and Petruccione 1999 ; Dowker and Henson 2004 ; Roderich 2006 ; Tumulka 2006 ; Bedingham 2011 ; Bedingham et al. 2014 ; Tilloy 2017 ; Bedingham and Pearle 2019 ; Tumulka 2021 . Previous attempts have resulted in unwanted side effects Pearle 2007 ; Pearle 2009 such as violation of microcausality Jones et al. 2021 , infinite rate of energy Pearle 1990 , particle production from the vacuum Bedingham and Pearle 2019 , and tachyon-like dynamics Pearle 1999 . In addition, a more general conceptual issue surrounds the possible compatibility between the wavefunction localization and relativity principles Bassi and Ghirardi 2003 . Here, we propose a mathematically consistent relativistic generalization of the collapse dynamics, which is characterized by a local field collapse operator and a non-Markovian stochastic noise with a Lorentz invariant correlation. We show that locality saves microcausality, while non-Markovianity and a normal ordering prescription lead to a finite rate of energy, without having particle production from the vacuum or a tachyonic behavior. Non-linear unraveling. – We construct of our model, starting from the non-relativistic (NR), colored, non-linear collapse equation for the statevector Adler and Bassi 2007 ; Diósi and Ferialdi 2014 . To second order in \sqrt{\gamma} , the equation takes a closed form in \ket{\psi(t)} and reads Adler and Bassi 2007 \displaystyle\frac{\!\operatorname{d}\!}{\!\operatorname{d}\!t}\ket{\psi(t)}=\left[-\frac{i}{\hbar}\hat{H}^{\prime}+\sqrt{\gamma}\sum_{i=1}^{N}\left(\hat{A}_{i}-\braket{\hat{A}_{i}}_{t}\right)\xi_{i}(t)\right. (1) \displaystyle+\left.\gamma\left(\hat{O}_{+}(t)-\braket{\hat{O}_{+}(t)}\right)\right]\ket{\psi(t)}. Here, \hat{H}^{\prime}=\hat{H}_{0}+i\hbar\gamma\hat{O}_{-}(t) , with \hat{H}_{0} being the standard quantum mechanical Hamiltonian, \gamma is the common coupling with the N collapse noises \xi_{i}(t) , which are real Gaussian random processes having zero mean and correlations \mathbb{E}\left[\xi_{i}(t)\xi_{j}(s)\right]=D_{ij}(t,s). (2) Here D_{ij}(t,s) is the noise correlation whose explicit form is to be determined, \mathbb{E} denotes the average over different realizations of the noises \xi_{i}(t) . Further, \hat{A}_{i} in Eq. ( 1 ) are a set of commuting self-adjoint operators that describe how the collapse occurs, and are known as collapse operators. The operators \hat{O}_{-} and \hat{O}_{+} are, respectively, the anti-self-adjoint and self-adjoint parts of \hat{O} . They read \hat{O}_{\pm}(t)=-\sum_{i,j=1}^{N}\int_{0}^{t}\!\operatorname{d}\!s\,D_{ij}(t,s)\left[\hat{A}_{i},\hat{A}_{j}(s-t)\right]_{\pm}, (3) where \left[\,\cdot\,,\,\cdot\,\right]_{-}=\left[\,\cdot\,,\,\cdot\,\right] and \left[\,\cdot\,,\,\cdot\,\right]_{+}=\left\{\,\cdot\,,\,\cdot\,\right\} denote respectively the commutator and anticommutator, while \hat{A}_{j}(s-t)=\hat{U}_{0}^{\dagger}(s-t)\hat{A}_{j}\hat{U}_{0}(s-t) is evolved with respect to the free Hamiltonian \hat{H}_{0} , i.e. \hat{U}_{0}(t)=\exp(-i\hat{H}_{0}t/\hbar) . Notice that Eq. ( 1 ) is only valid up to second order in \sqrt{\gamma} . The prescription to obtain the equation for higher orders can be found in Adler and Bassi 2007 . We aim at generalizing Eq. ( 1 ) to the relativistic regime so that the corresponding master equation is relativistically covariant. The master equation can be equivalently obtained from the linear unraveling of Eq. ( 1 ) for the non-normalized vector \ket{\phi(t)} . It reads Adler and Bassi 2007 \displaystyle\frac{\!\operatorname{d}\!}{\!\operatorname{d}\!t}\ket{\phi(t)} \displaystyle=-\frac{i}{\hbar}{\hat{H}_{0}}\ket{\phi(t)}+\left[\sqrt{\gamma}\sum_{i=1}^{N}\hat{A}_{i}\xi_{i}(t)\right. (4) \displaystyle\left.-\gamma\sum_{i,j=1}^{N}\int_{0}^{t}\!\operatorname{d}\!s\,D_{ij}(t,s)\hat{A}_{i}\hat{A}_{j}(s-t)\right]\ket{\phi_{0}(t)}, where, to second order in \sqrt{\gamma} , we have \ket{\phi(t)}=\ket{\phi^{(0)}(t)}+\sqrt{\gamma}\ket{\phi^{(1)}(t)}+\gamma\ket{\phi^{(2)}(t)}+\mathcal{O}(\gamma^{3/2}) . The standard prescription gives the corresponding master equation in the Schrödinger picture \!\operatorname{d}\!\hat{\rho}^{(2)}(t)/{\!\operatorname{d}\!t}=-\frac{i}{\hbar}\left[{\hat{H}_{0}},\hat{\rho}^{(2)}(t)\right]+\mathcal{D}[\hat{\rho}^{(0)}(t)] , where \mathcal{D}[\hat{\rho}^{(0)}(t)]\!\!=\!\!-\gamma\!\!\sum_{i,j=1}^{N}\!\!\int_{0}^{t}\!\!\!\operatorname{d}\!s\,D_{ij}(t,s)\left[\hat{A}_{i},\left[\hat{A}_{j}(s-t),\hat{\rho}^{(0)}(t)\right]\right], (5) and \hat{\rho}^{(\alpha)}(t):=\mathbb{E}[\ket{\phi^{(\alpha)}(t)}\bra{\phi^{(\alpha)}(t)}] with \alpha=0,1,2 . We move to the interaction picture by using the standard relation \dot{\hat{\rho}}^{(2)}_{I}(t)=\hat{U}_{0}(t)\left(\frac{i}{\hbar}\left[{\hat{H}_{0}},\hat{\rho}^{(2)}(t)\right]+\dot{\hat{\rho}}^{(2)}(t)\right)\hat{U}_{0}^{\dagger}(t) , such that \frac{\!\operatorname{d}\!}{\!\operatorname{d}\!t}{\hat{\rho}}^{(2)}_{\text{\tiny I}}(t)\!\!=\!\!-\gamma\!\!\sum_{i,j=1}^{N}\!\!\int_{0}^{t}\!\!\!\operatorname{d}\!s\,D_{ij}(t,s)\left[\hat{A}_{i}(t),\left[\hat{A}_{j}(s),\hat{\rho}^{(0)}_{\text{\tiny I}}(t)\right]\right], (6) whose solution reads \displaystyle\hat{\rho}^{(2)}_{\text{\tiny I}}(t)=\hat{\rho}^{(0)}_{\text{\tiny I}}(t) (7) \displaystyle-\gamma\sum_{i,j=1}^{N}\int_{0}^{t}\!\operatorname{d}\!s\int_{0}^{s}\!\operatorname{d}\!s^{\prime}\,D_{ij}(s,s^{\prime})\left[\hat{A}_{i}(s),\left[\hat{A}_{j}(s^{\prime}),\hat{\rho}_{\text{\tiny I}}^{(0)}(t)\right]\right]. Further, we move from the discrete label i to its continuous version \mathbf{x} . In doing so, we identify \hat{A}_{i}(t)\leftrightarrow\hat{Q}(\mathbf{x},t) and \sum_{i,j=1}^{N}D_{ij}(t,s)\leftrightarrow\iint\!\operatorname{d}\!\mathbf{x}_{2}\!\operatorname{d}\!\mathbf{x}_{1}\,\mathbb{E}[\xi(t,\mathbf{x}_{2})\xi(s,\mathbf{x}_{1})]=\iint\!\operatorname{d}\!\mathbf{x}_{2}\!\operatorname{d}\!\mathbf{x}_{1}\,G(x_{2},x_{1}) , where x_{2}=(t,\mathbf{x}_{2}) and x_{1}=(s,\mathbf{x}_{1}) . Then, the expectation value of any generic operator \hat{O} , to the second order in \sqrt{\gamma} , can be expressed as \braket{\hat{O}(z)}^{(2)}=\operatorname{Tr}\left[\hat{O}_{\text{\tiny I}}(z)\hat{\rho}^{(2)}_{\text{\tiny I}}(z)\right] . Its time derivative reads \displaystyle\frac{\!\operatorname{d}\!}{\!\operatorname{d}\!z^{0}}\braket{\hat{O}(z)}^{(2)} \displaystyle=\frac{\!\operatorname{d}\!}{\!\operatorname{d}\!z^{0}}\operatorname{Tr}\left[\hat{O}_{\text{\tiny I}}(z)\hat{\rho}(0)\right]-\gamma\iint_{x_{2}^{0}\leq x_{1}^{0}=z^{0}}\!\operatorname{d}\!^{4}x_{2}\,\!\operatorname{d}\!^{3}\mathbf{x}_{1}\,G(x_{2},x_{1})\operatorname{Tr}\left[\left[\hat{Q}(x_{2}),\left[\hat{Q}(x_{1}),\hat{O}_{\text{\tiny I}}(z)\right]\right]\hat{\rho}(0)\right] (8) \displaystyle-\gamma\fint\,G(x_{2},x_{1})\operatorname{Tr}\left[\left[\hat{Q}(x_{2}),\left[\hat{Q}(x_{1}),\frac{\!\operatorname{d}\!}{\!\operatorname{d}\!z^{0}}\hat{O}_{\text{\tiny I}}(z)\right]\right]\hat{\rho}(0)\right], where \fint:=\iint_{{x_{2}^{0}\leq x_{1}^{0}\leq z^{0}}}\!\operatorname{d}\!^{4}x_{2}\,\!\operatorname{d}\!^{4}x_{1} . Eq. ( 8 ) can also be obtained by deriving \hat{O}^{(2)}(z)\!=\!\hat{O}^{(0)}(z)\!-\!\gamma\!\fint\!\!\,G(x_{2},x_{1})\!\left[\hat{Q}(x_{2}),\!\left[\hat{Q}(x_{1}),\hat{O}^{(0)}(z)\right]\right], (9) in the Heisenberg picture with respect to z^{0} , and taking the expectation value with respect to the initial state. Here, we have used \hat{O}_{\text{\tiny I}}(z)=\hat{O}^{(0)}(z) . From a mathematical standpoint, the time evolved second order expression for \hat{O}^{(2)}(z) in Eq. ( 9 ) can be also be obtained within the Heisenberg picture of a linear and unitary unraveling, as shown in Appendix A . In the unitary unraveling, all statistical effects of the collapse dynamics can be captured by adding a stochastic term \hat{H}_{\text{st}}(t)=\hbar\sqrt{\gamma}\int\!\operatorname{d}\!\mathbf{z}\,\hat{Q}(\mathbf{z})\xi(t,\mathbf{z}) to the standard quantum mechanical evolution, i.e. \hat{H}=\hat{H}_{0}+\hat{H}_{\text{st}}(t) . Conceptually, however, such an unraveling is in stark contrast with the non-linear and non-unitary unraveling of collapse models. For instance, if one insists that an observed outcome in an experiment is explained objectively only by the localization of the wavefunction around a specific eigenvector, the unitary unraveling would fail to provide such an explanation. One might argue instead that even though the unitary unraveling does not lead to the collapse, it still inevitably turns a pure state into a mixed one for a macroscopic system, consistently with the Born rule. The distiction between the two unravelings is certainly important from theoretical considerations. However, since experimentally one only has access to the density matrix Diósi 2015 ; Diósi 2018 , here we take a more pragmatic approach and focus only on the latter dynamics. We emphasize again that just like their discrete counterpart, \hat{Q}(x) appearing in Eq. ( 9 ) is the freely evolved operator in the Heisenberg picture and G(x,y) some generic function of x and y . Next we study the requirements on \hat{Q}(x) and G(x,y) to have that Eq. ( 9 ) relativistically consistent. Lorentz covariance.– The Lorentz covariance of Eq. ( 9 ) follows from standard quantum field theory (QFT). For the