Multipartite entanglement from ditstrings for 1+1D systems
Zane Ozzello, Yannick Meurice · 2025
Read Multipartite entanglement from ditstrings for 1+1D systems by Zane Ozzello, Yannick Meurice (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-2507-14422.txt.
Introduction
Zane Ozzello and Yannick Meurice use entanglement to pinpoint phase transitions in a paper from two thousand twenty-five. Multipartite entanglement measures how quantum parts share information beyond any pair. The authors test whether certain combinations of such measures flag critical points in one-dimensional systems.
They study the quantum Ising model, a lattice field theory approximated with qutrits, and arrays of Rydberg atoms. Single-part entanglement already shifts at phase boundaries, but combined quantities sharpen and localize the signal. A mutual information stand-in gives a lower bound that tracks the same changes. A filtering step removes weak probabilities and tightens that bound. The approach offers a cheaper route to mapping phase diagrams.
Catalog record
| CREATOR | Zane Ozzello, Yannick Meurice |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-14422
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Multipartite entanglement from ditstrings for 1+1D systems Zane Ozzello, Yannick Meurice Abstract We show that multipartite entanglement can be used as an efficient way of identifying the critical points of 1+1D systems. We demonstrate this with the quantum Ising model, lattice λ ϕ 4 \lambda\phi^{4} approximated with qutrits, and arrays of Rydberg atoms. To do so we make use of multipartite compositions of entanglement quantities for different parts combined to form the strong subadditivity, weak monotonicity, and a convex combination of these with conformal properties. These quantities display some remarkable properties. We will demonstrate how the entanglement of individual parts together displays behavior at phase boundaries, but the combination of these in the aforementioned quantities sharpens and localizes this behavior to the boundaries even better. We will show that we can extend a scheme for approximating the entanglement with the mutual information, and that this acts as a lower bound which will also follow the changes in the entanglement for the above quantities, despite the additional contributions of different signs. This mutual information approximation to the identifying quantities can have its lower probabilities removed in a process we call filtering, and despite the combination of terms will respond well to the filtering and offer improvements to the lower bound. I Introduction Entanglement entropy is a uniquely quantum property that allows for the quantification of the dependence of partitions of a system upon each other. Such a property naturally has roles throughout many areas of current research. In this regard, the bipartite entanglement entropy is invoked, a quantity specifically looking at the entanglement between two parts of a system Eisert et al. 2010 ; Ghosh et al. 2015 ; Van Acoleyen et al. 2016 ; Kharzeev and Levin 2017 ; Bañuls et al. 2017 ; Baker and Kharzeev 2018 ; Abanin et al. 2019 ; Cirac et al. 2021 ; Robin et al. 2021 ; Zhang et al. 2022 ; Knaute et al. 2024 . A more general, but often less discussed version of this is the multipartite entanglement entropy, which concerns the entanglement of multiple parts of a system with one another Horodecki et al. 2024 . This potentially opens additional doors for the application of entanglement entropy across current research avenues Blasone et al. 2009 ; Amico et al. 2008 ; Brökemeier et al. 2025 . Entanglement entropy, as it involves the interconnectedness of a system, can be very difficult to experimentally measure, causing problems not just in the multipartite case, but the bipartite case as well Islam et al. 2015 ; Pompili et al. 2021 . In the age of quantum computers, a work around to these limitations is of premier interest. Potentially, by having a means of identifying entanglement with a quantum computer, there could be measurement of the quantum properties of a quantum system directly. As a result, looking for means of approximating the entanglement while its direct measurement is difficult is paramount Hauke et al. 2016 ; Hamilton and Leditzky 2024 . Recently, methods have been proposed to utilize classical quantities, Shannon entropy and mutual information, to approximate bipartite entanglement entropy Meurice 2025 . This is done by using bitstring results from a quantum computer, along with their associated counts divided by shot totals, to create probabilities for the aforementioned quantities. In this work, we take the methods a step further, generalizing and showing that the methodology is applicable to multipartite entanglement entropy. Despite the fact that this filtering is applied to the multiple mutual information quantities at once, which introduce even more Shannon entropies at once, the filtering still yields plateaus that could be used as potential improvements to the bound. Two of the chief areas for the application of entanglement entropy is in conformal field theory (CFT) and looking at phase transitions Vidal et al. 2003 ; Calabrese and Cardy 2009 ; Hauke et al. 2016 . To further show the efficacy of our ditstring approach to multipartite entanglement, we take the quantities strong subadditivity, weak monotonicity, and a convex combination of these with conformal properties, S_{\Delta} which are outlined in Lin and McGreevy 2023 in the context of conformal field theory and identify critical points using entanglement entropy and its approximation. These quantities are guaranteed to be positive and are a set of entanglement entropies of different parts with one another that are added and subtracted to form the quantities. Despite these combinations of positive and negative terms, these quantities manage to isolate a peak in a parameter range by canceling off values to either side. Not only does the peak come about in this way, but these cancellations also better isolate these quantities to define the phase boundaries for their respective models. We will further highlight these behaviors in Sec. III . The article is organized as follows. In Sec. II we discuss the structure of multipartite Hilbert space and its application to von Neumann entanglement entropy and mutual information. In Sec. III , we discuss how we attempt to find critical points with the ideas of Lin and McGreevy 2023 as our starting point. We then apply our multipartite scheme and ditstring approximation to three 1+1D models: quantum Ising, lattice \phi^{4} with qutrits, and a chain of Rydberg atoms. For this we work exclusively in periodic boundary conditions. When we loosen this restriction, we look at the impact of switching to open boundary conditions in Sec. IV . We then utilize methods laid out in Kaufman et al. 2024 ; Kaufman et al. 2025 in Sec. V and apply a filtering scheme that involves the removal of low probabilities that go into the composite entropic quantities. In the conclusions we touch on future directions, looking at the validity of this method for identifying critical points, further pushing the mutual information lower bound, and discussion as to how these can be combined for actual quantum computer implementation. In the appendices we discuss how choice of basis impacts the value for the mutual information. II Multipartite Qudit Hilbert Space In the following we consider N qudits with d states in the local Hilbert space. The computational basis then consists of d^{N} elements of the form \left|{\{n\}}\right\rangle\equiv\left|{n_{0},n_{1},\dots,n_{N_{q}-1}}\right\rangle, (1) with n_{j}=0,\dots,d-1 . Any element of this basis can be factored in a multipartite way \left|{\{n\}}\right\rangle=\left|{\{n\}_{A}}\right\rangle\left|{\{n\}_{B}}\right\rangle\left|{\{n\}_{C}}\right\rangle\dots, (2) with \displaystyle\left|{\{n\}_{A}}\right\rangle \displaystyle\equiv \displaystyle\left|{n_{0},n_{1},\dots,n_{|A|-1}}\right\rangle\ {\rm and} \displaystyle\left|{\{n\}_{B}}\right\rangle \displaystyle\equiv \displaystyle\left|{n_{|A|},\dots,n_{|A|+|B|-1}}\right\rangle (3) \displaystyle\dots with |A| , |B|\dots the number of qudits in A,\ B,\dots An arbitrary state in the computational basis is of the form \left|{\psi}\right\rangle=\sum_{\{n\}}c_{\{n\}}\left|{\{n\}}\right\rangle, (4) and the state \left|{\{n\}}\right\rangle will be observed with a probability p_{\{n\}}=|c_{\{n\}}|^{2}. (5) Such probabilities can be estimated by ditstring returns from a quantum computer by taking the counts of each state’s occurrence divided by the total shots on the system. Then, a Shannon entropy of the entire system is H_{ABC\dots}\equiv-\sum_{\{n\}}p_{\{n\}}\ln(p_{\{n\}}), (6) associated with the state \left|{\psi}\right\rangle . When dealing with a pure state, your entanglement entropy will always be 0. Practically speaking, one will typically not be dealing with a pure state, but rather be using a system partitioned in some way. Working with a bipartition is natural, you just need to split the system using the basis factoring laid out in Eq. ( 2 ). When dealing with the multipartite case,