Entanglement and magic transitions in an all-to-all non-Hermitian spin model

arXiv:2608.31064 · quant-ph · Submitted 2026-08-31 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: We're looking at the paper: "Entanglement and magic transitions in an all-to-all non-Hermitian spin model".

Mira: Here's the gist of the paper. The paper constructs an all-to-all non-Hermitian spin model where the dominant eigenvalue can be tracked analytically from a clean limit into a disordered regime, revealing how spectral outliers interact with the bulk spectrum and how this competition dictates long-time dynamics through entanglement and magic measures 1.

Kai: Model Construction The authors construct a non-Hermitian spin ensemble of L spins with exactly k-local all-to-all interactions, in which this dominant eigenvalue can be tracked analytically from the clean limit into the disordered regime 2.

Mira: The model is formulated by taking two components to be independent, exactly kR and kI –local spin Hamiltonians, with coupling coefficients drawn from two normal distributions with independently tunable means and variances 7.

Kai: The non-Hermitian model is defined as G(µR, µI, σR, σI) = 1/sqrt(2)

HR(µR, σ2 R) + iHI (µI, σ2 I): 8.

Title and authors: Kai: So, we’ve talked about what this paper is titled and who wrote it. The title itself points directly to the two main things they are studying: entanglement entropy and magic measures in a non-Hermitian system >

Mira: And they constructed this specific model where you have these L spins with k-local all-to-all interactions, but the key insight is that they can actually track the dominant eigenvalue analytically even when disorder kicks in >

Lev: It’s interesting because standard random matrix theory doesn't usually capture how structured non-Hermitian ensembles behave in this way; they are motivated by that because they find these constraints correlate the matrix elements in ways unstructured random-matrix models don't capture >

Kai: The authors are Sasanka Dowarah and Michael Kolodrubetz, and what’s significant is that they use this structure to formulate the problem as an outlier problem for a structured non-Hermitian random-matrix ensemble >

Mira: They are then testing three distinct disorder regimes: purely anti-Hermitian disorder, purely Hermitian disorder, and mixed disorder of equal strength >

Lev: For the engineer, having these fixed Pauli string weights—exactly k local—means they aren't just throwing random noise at the system; they are restricting it to specific interaction patterns >

Kai: So before we get into what the results actually show, let’s hear how this paper summarizes its main findings and what that means for us experimentally >

The paper's summary: Mira: Basically, the summary of this paper says they define two natural probes of the steady state: entanglement entropy and magic measures, because those tools are supposed to distinguish between a product-like state and a thermal state >

Lev: They show that as dissipation is tuned—which is what you’re doing when you tune that anti-Hermitian disorder—one can ask how these two notions of complexity change >

Kai: The core finding is the existence of an outlier-to-bulk transition, which they define as the point where the bulk overtakes the outlier in imaginary part >

Mira: When this happens, it causes a sharp switch in both entanglement and magic measures at that crossing point, leaving a distinct signature in both individual cases and on average >

Lev: This is important because it means you can use those two observables as diagnostics for non-generic transitions; if you see that specific abrupt change, you know what the system is doing >

Kai: It also shows that for the equal mixed-disorder case, the post-transition state has a compact statistical description where a random vector ansatz predicts a Haar-random magic density and a Page law entanglement slope >

The paper's improvements: Mira: They suggest some key improvements to this work. For instance, they explore how the distribution of sample dependent pseudocritical points converges toward analytical predictions when you look at large locality >

Lev: That’s useful because when we try to implement this on real hardware, knowing where those critical disorder thresholds are converging helps us predict how robust our system will be >

Kai: They also talk about a fully rigorous treatment of the outlier trajectory, which is something they haven't fully done yet and is a clear path forward for them >

Mira: And they suggest coupling a spatially local chaotic Hermitian Hamiltonian to an anti-Hermitian component, which seems like it could be a way to bridge the gap between their structured model and more general physical Hamiltonians >

Lev: From an error correction standpoint, if we can map these dynamics onto a canonical form of a conditional no-click generator, that gives us a concrete way to analyze the jump operators and quantify the trade-off between locality and postselection probability >

Conclusion: Kai: So wrapping up this paper, the main implication is that the outlier-to-bulk switching transition isn't generic; it only happens if you have this specific competition where the random matrix bulk outgrows a deterministic outlier along that imaginary direction >

Mira: That distinction between existence and dynamical dominance—which means having the largest imaginary part and controlling the long time state—is crucial because it explains why entanglement and magic signatures change so abruptly at that crossing point >

Lev: For us on the engineering side, this helps us understand how dissipation, represented by that anti-Hermitian disorder strength sigma I, has a direct experimental meaning as an engineered dissipation rate >

Kai: The paper suggests extending this to coupling with chaotic Hermitian Hamiltonians and understanding the scaling of these critical points is key to seeing where the physics is truly non-generic >

Mira: It connects these spectral mechanisms to measurement-induced transitions, which feeds right into the next area we’ll be looking at >

Lev: So, for this paper, it’s a strong link between the spectral structure and the dynamics of monitored systems that we need to keep in mind as we build our next generation of quantum processors >

Department of Physics, The University of Texas at Dallas

quant-ph

Submitted: 2026-08-31

Updated: 2026-11-10

Comments: 35 pages, 18 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 80/100

The gist: The paper constructs an all-to-all non-Hermitian spin model where the dominant eigenvalue can be tracked analytically from a clean limit into a disordered regime…

Key concepts

Dominant Eigenvalue ($z^ullet$)
This is the eigenvalue of the non-Hermitian model that has the largest imaginary part. It represents the state that controls the long-time behavior of the system. Its position relative to other eigenvalues determines whether a spectral outlier exists or if it becomes dynamically dominant.
Outlier vs. Dynamical Dominance
Outlier existence means an eigenvalue is isolated from the main spectrum (spectral isolation). Dynamical dominance means that eigenvalue possesses the largest imaginary part, giving it control over the system's long-time state. The transition between these two states is crucial for understanding system evolution.
Magic Measures
These are specific quantities, like entanglement and magic signatures, that change abruptly when the dominant eigenvalue switches from being an outlier to being part of the bulk spectrum. These sharp changes serve as diagnostic tools for identifying critical transitions in the system's state.
Postselected Quantum Trajectory
The paper shows that the system's dynamics can be mapped exactly onto a monitored quantum system where measurements are performed. The long-time states studied correspond to conditional steady states, interpreted as postselected trajectories of this monitored spin model.

Terminology

Summary

The paper constructs an all-to-all non-Hermitian spin model where the dominant eigenvalue can be tracked analytically from a clean limit into a disordered regime, revealing how spectral outliers interact with the bulk spectrum and how this competition dictates long-time dynamics through entanglement and magic measures 1. p. 0

Model Construction

The authors construct a non-Hermitian spin ensemble of L spins with exactly k-local all-to-all interactions, in which this dominant eigenvalue can be tracked analytically from the clean limit into the disordered regime 2. The model is formulated by taking two components to be independent, exactly kR and kI –local spin Hamiltonians, with coupling coefficients drawn from two normal distributions with independently tunable means and variances 7. p. 1 The non-Hermitian model is defined as G(µR, µI, σR, σI) = 1/sqrt(2) [HR(µR, σ2 R) + iHI (µI, σ2 I)] 8. p. 2

Spectral Structure and Universality

The deterministic limit yields a spectrum given in closed form by Krawtchouk polynomials 21. In the fully disordered limit, the ensemble falls into one of the standard non-Hermitian random-matrix classes, fixed by an antiunitary symmetry tied to the parities of L, kR, and kI 7. p. 1 The spectral bulk exhibits an elliptic law for the zero-mean part of the Hamiltonian 45, which is consistent with known Gaussian-to-semicircle crossover behavior depending on whether k is greater than or less than sqrt(L) 46. p. 6

Outlier Dynamics and Phase Transitions

The paper distinguishes between outlier existence, meaning spectral isolation from the bulk, and dynamical dominance, meaning possession of the largest imaginary part and hence control of the long-time state FIG. 1. The dominant eigenvalue z⋆, which is shown as a green circle in the figure, is defined as the eigenvalue with the largest imaginary part C. p. 3 Diagnostics of the long-time state>. p. 3 The transition between outlier existence and dynamical dominance occurs when the bulk overtakes the outlier along the imaginary direction IV. p. 1 PHASE DIAGRAM: EXISTENCE VERSUS DOMINANCE>. p. 9 This competition is present for purely anti-Hermitian and for mixed disorder, where it produces a sharp switch in the dominant eigenvalue and simultaneous jumps in entanglement and magic V. p. 1 OUTLIER-TO-BULK TRANSITIONS OF DOMINANT EIGENVALUE>. p. 0

Entanglement and Magic Signatures

The transition point where the dominant eigenvalue switches from the outlier to the bulk edge is exactly the point at which these two quantities undergo an abrupt change V. The long-time state changes abruptly at this crossing, leaving a sharp signature in both entanglement and magic III C 2. p. 1 For the equal mixed-disorder case, the post-transition state admits a compact statistical description where a randomvector ansatz predicts a Haar-random magic density and a Page-law entanglement slope VI. p. 1 RELATION TO POSTSELECTED MONITORED DYNAMICS>. p. 0

Relation to Monitored Dynamics

The normalized dynamics of the system are shown to admit an exact reinterpretation as the postselected quantum trajectory of a monitored system VI. The long-time states studied in this work can therefore be interpreted as the conditional steady states of an interacting monitored spin system, postselected on the absence of jumps VI. p. 0 The local unraveling maps the dynamics to a canonical form of a conditional no-click generator VI. p. 1

Conclusion and Outlook

The results show that the outlier-to-bulk switching transition is not generic but requires the random-matrix bulk to outgrow the deterministic outlier specifically along the imaginary direction VII DISCUSSION AND OUTLOOK>. The paper suggests extensions such as a fully rigorous treatment of the outlier trajectory and coupling a spatially local chaotic Hermitian Hamiltonian to an anti-Hermitian component VII. p. 1 The anti-Hermitian disorder strength σI has a direct experimental meaning as an engineered dissipation rate VII. p. 1 The results are consistent with the broader picture emerging from non-Hermitian many-body quantum chaos VII. p. 4

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved spectral analysis for non-Hermitian systems by tracking eigenvalues with largest imaginary parts allows for analytical tractability in interacting many-body models, enabling prediction of long-time states from a spectral one.

  2. Development of an analytically tractable model that interpolates between universal random-matrix predictions and microscopic physical Hamiltonians by independently specifying the statistical properties of Hermitian and anti-Hermitian components.

  3. Ability to distinguish between different disorder regimes—purely anti-Hermitian, purely Hermitian, and mixed—by analyzing how the spectral bulk competes with an isolated spectral outlier along the imaginary direction.

  4. System can predict sharp transitions in entanglement entropy and magic based on whether the random-matrix bulk outgrows a deterministic outlier, providing a diagnostic for non-generic transitions.

  5. Implementation of a postselected quantum trajectory framework that maps non-Hermitian evolution onto monitored spin systems, connecting spectral mechanisms to measurement-induced transitions.

  6. Capacity to analyze entanglement and magic signatures of the long-time state by identifying abrupt changes at the outlier-to-bulk dominance crossing, revealing how these observables change sharply in both individual realizations and on average.

  7. Application of a random vector ansatz to predict Haar-random magic density and Page-law entanglement slope above the dominance threshold, providing a statistical description of typical chaotic states.

  8. Capability to determine the finite size scaling behavior of critical disorder thresholds by analyzing the distribution of sample-dependent pseudocritical points, showing convergence toward analytical predictions for large locality.

  9. Creation of an exact kI-local unraveling for monitored dynamics, allowing explicit mapping of jump operators to local Pauli strings and quantifying the trade-off between locality and postselection probability.

Sources

Related papers