Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media

arXiv:2607.26406 · cond-mat.stat-mech · Submitted 2026-07-29 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media".

Mira: Here's the gist of the paper. Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media develops a theoretical framework for active jammed systems based on non-Hermitian random matrix theory to quantify linear-response compliance and establish a scaling law at the jamming transition Theoretical Foundation The work formulates the linearized dynamical matrix as a non-Hermitian perturbation of a Wishart ensemble describing the passive contact network <ref:2607.26406#pg2> Starting from a microscopic model of active particles with nonreciprocal interactions, it uses Girko’s Hermitization together with the self-consistent Born approximation to derive a self-consistent equation for the low-frequency resolvent based on the full Marchenko–Pastur distribution <ref:2607.26406#pg2> This approach shows that active nonreciprocity regularizes the soft-mode divergence at the jamming transition, leading to a scaling law for mechanical compliance <ref:2607.26406#pg2> Microscopic Modeling The microscopic model considers a D-dimensional disordered assembly of N active spherical particles governed by both symmetric mutual contact mechanics and asymmetric, non-equilibrium self-propulsion forces <ref:2607.26406#pg3> The passive structural Hessian matrix H0 is modeled as a shifted Wishart matrix ensemble whose baseline spectral density obeys the Marchenko–Pastur law <ref:2607.26406#pg3> Active self-propulsion or RTP is introduced by assigning a directional unit vector ni to each spherical grain, which acts as an explicit non-reciprocal perturbation atop the passive, Wishart-regularized elastic matrix <ref:2607.26406#pg3> The active driving term Fa ̸= 0 therefore acts as an explicit non-reciprocal perturbation atop this passive, Wishart-regularized elastic matrix <ref:2607.26406#pg3> Non-Hermitian RMT Analysis The linearized equations of motion result in a global stability operator M that is explicitly non-Hermitian due to the violation of detailed balance by active driving <ref:2607.26406#pg4> The operator M is decomposed into a structural passive component and an asymmetric active driving matrix M:= −H0 + α0A <ref:2607.26406#pg4> Modeling A via the Ginibre ensemble serves as a minimal, universality-based mean-field framework to analytically evaluate the regularized complex spectrum and universal compliance scaling without relying on fine-tuned spatial details <ref:2607.26406#pg5> The self-consistent resolvent equation is derived using Girko’s Hermitization method combined with the Self-Consistent Born Approximation (SCBA) <ref:2607.26406#pg5> Universal Scaling and Crossover The analysis yields a true, self-consistently derived universal asymptotic scaling law for active compliance at the jamming threshold: J ∼ α −6/7 0 <ref:2607.26406#pg7> This scaling arises because active non-reciprocity breaks the directional symmetry of action-reaction pairs between adjacent cells, generating complex, imaginary parts in the eigenvalue spectrum that cut off and regularize the low-frequency resolvent divergence <ref:2607.26406#pg7> A crossover scaling form J(∆ϕ, α0) = α −6/7 0 f±(α −4/5 0 ∆ϕ) is postulated to unify the regular perturbative behavior deep in the jammed phase with the singular scaling at marginal stability <ref:2607.26406#pg9> Numerical Validation and Interpretation Numerical evaluations using finite-size non-Hermitian random matrices confirm that the singular Eq.

Kai: (22) non-perturbative behavior governs the actual physical response in non-normal active media <ref:2607.26406#pg10> The comparison between the naive eigenvalue compliance Jeig and the singular value sum JSVD shows remarkable agreement with the theoretical SCBA scaling Eq.

Mira: (22) for α0 ≤ 0.03, confirming that active compliance exhibits this singular behavior <ref:2607.26406#pg11> The physical intuition underlying the striking non-integer exponent −6/7 at the jamming point represents a fundamental feature of non-equilibrium collective systems <ref:2607.26406#pg7> Future Directions The framework suggests several promising avenues for future research, including extending the formulation into the non-linear regime to capture largedeformation shear-thinning and plastic yielding in active solids <ref:2607.26406#pg13> Further work is desirable to incorporate spatial correlations, heterogeneous contact networks, and finite-dimensional fluctuations beyond the mean-field random matrix approximation <ref:2607.26406#pg13> The framework also opens avenues for applying this approach to frictional granular packings to isolate the distinct roles of thermal/active driving versus static frictional asymmetry <ref:2607.26406#pg13> The present work establishes a non-Hermitian RMT framework for active jammed systems, demonstrating how non-reciprocal microscopic activity regularizes soft-mode mechanical divergences <ref:2607.26406#pg12> The analysis provides a microscopic interpretation of how nonequilibrium activity modifies mechanical stability near jamming <ref:2607.26406#pg12> This work should be regarded as a first step toward a statistical-mechanical theory of active compliance, the inverse of rigidity <ref:2607.26406#pg7> Several important issues remain open, such as quantitative comparisons with numerical simulations or extending the theory beyond the mean-field random matrix approximation <ref:2607.26406#pg7> Acknowledgments This paper was partially supported by the JSPS KAKENHI Grant No.

Kai: JP26K06960 <ref:2607.26406#pg14> This work was inspired by Suravi Pal’s presentation at the mini-workshop on “the rheology of dense materials” <ref:2607.26406#pg14> The author appreciates stimulating discussions with Suravi Pal and Takeshi Kawasaki <ref:2607.26406#pg14> This paper is dedicated to Professor Hajime Tanaka to celebrate his scientific achievements <ref:2607.26406#pg14> Appendix A The derivation illustrates how microscopic non-reciprocal active fluxes and steric constraints generate non-Hermitian mode coupling in a lattice-based Active Exclusion Process <ref:2607.26406#pg15> The coarse-graining leads to coupled non-linear hydrodynamic equations for density and polarization fields <ref:2607.26406#pg15> The mapping to random matrix ensembles serves as an “effective mean-field representation” valid in strongly disordered, jammed states <ref:2607.26406#pg15> Appendix B The derivation provides a microscopic derivation of the activity α starting from a standard vertex-model description of tissues <ref:2607.26406#pg17> This derivation shows that the effective stiffness keff = k0 − αA, where α is a coarse-grained susceptibility encoding multiple microscopic mechanisms <ref:2607.26406#pg17> Appendix C The systematic, diagrammatic derivation of the self-consistent equation for the regularized compliance loop function J(z, z∗; ϵ) is provided using Girko’s Hermitization technique <ref:2607.26406#pg18> The Self-Consistent Born Approximation (SCBA) truncates the Dyson expansion at 1-loop order to yield the SCBA <ref:2607.26406#pg18> Appendix D The derivation connects the zero-frequency limit of the regularized low-frequency resolvent J to the macroscopic mechanical compliance matrix from linearized non-equilibrium dynamics <ref:2607.26406#pg20> This result explicitly proves that J serves as the macroscopic scalar compliance of the active system <ref:2607.26406#pg20> Appendix E The asymptotic solution of the self-consistent equation yields the scaling law J ∼ α −6/7 0 in the limit ∆ → 0 <ref:2607.26406#pg22> The derivation relies solely on the structure of the integral equation and shows that the exponent in Eq.

Title and authors: Kai: So we're looking at this paper, "Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media." It sets up a framework for understanding active jammed systems using non-Hermitian random matrix theory. Mira It starts by taking a microscopic model of active particles with nonreciprocal interactions and turns the linearized dynamical matrix into a non-Hermitian perturbation of a Wishart ensemble describing the passive contact network.

Lev: So, what does this framework actually help us build? We're dealing with active systems here, where the particles are moving on their own, not just sitting still. Kai Exactly. The paper suggests that this non-Hermitian structure lets them quantify how compliant these jammed systems are and establish a specific scaling law at the jamming transition.

Mira: It uses Girko’s Hermitization together with the self-consistent Born approximation to derive a self-consistent equation for the low-frequency resolvent, which is based on the full Marchenko–Pastur distribution. Lev That means they're trying to get a picture of how soft modes behave when things jam up, and they are using this non-Hermitian approach to regularize those divergences.

Kai: It's about taking something that usually blows up at the jamming transition—the soft modes—and showing that active nonreciprocity actually regularizes it, leading to a specific scaling law for mechanical compliance. Mira That’s the big claim, linking non-reciprocal activity directly to the mechanical response near jamming.

The paper's summary: Kai: So we're looking at this paper, "Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media." It starts by taking a microscopic model of active particles with nonreciprocal interactions and turns the linearized dynamical matrix into a non-Hermitian perturbation of a Wishart ensemble describing the passive contact network. Mira It uses Girko’s Hermitization together with the self-consistent Born approximation to derive a self-consistent equation for the low-frequency resolvent, which is based on the full Marchenko–Pastur distribution.

Lev: The core idea here is that they model D-dimensional disordered assembly of particles, like Janus grains, where you have both symmetric mutual contact mechanics and asymmetric, nonreciprocal self-propulsion forces. Kai So the active driving term A is explicitly nonreciprocal because it violates Newton’s third law, which means the coupling matrix A is strictly asymmetric.

Mira: Because of that asymmetry, they decompose the global stability operator M into a structural passive component and an asymmetric active driving matrix, M equals minus H0 plus alpha zero A. Lev And they model that active force transmission coefficient A using the Ginibre ensemble as a minimal mean-field framework to analyze the regularized complex spectrum.

Kai: The self-consistent resolvent equation is derived using Girko’s Hermitization combined with the Self-Consistent Born Approximation, which helps them find that universal compliance scaling without needing every tiny detail of the spatial arrangement. Lev So, what they actually show is that active nonreciprocity breaks the directional symmetry of action-reaction pairs between adjacent cells, which generates complex parts in the eigenvalue spectrum that cut off and regularize the low-frequency resolvent divergence.

The paper's improvements: Kai: The paper suggests a few things to improve this framework. They propose a true, self-consistently derived universal asymptotic scaling law for active compliance at the jamming threshold, which is J equals alpha minus six over seven zero. Mira That scaling arises because that active nonreciprocity breaks the directional symmetry of action-reaction pairs between adjacent cells, generating complex parts in the eigenvalue spectrum that cut off and regularize the low-frequency resolvent divergence.

Lev: They also propose a crossover scaling form for compliance, J equals alpha minus six over seven zero times a function of alpha minus four over five zero times delta phi. Kai That crossover form is meant to unify the perturbative behavior deep in the jammed phase with that singular scaling we see at marginal stability near the active jamming point.

Mira: They also suggest numerical evaluations using finite-size non-Hermitian random matrices M equals minus H0 plus alpha zero A confirm that this singular value sum, JSVD, follows the theoretical self-consistent Born approximation scaling, J equals alpha minus six over seven zero for activity alpha zero less than or equal to zero point three <ref:2607.26406#pg3>. Lev So they are using numerical non-Hermitian random matrices to check if their analytical SCBA scaling holds up against the singular value sum.

Kai: The paper notes that they compare the naive eigenvalue compliance, Jeig, with the singular value sum JSVD, and they show remarkable agreement for activity alpha zero less than or equal to zero point three <ref:2607.26406#pg3>. Mira That numerical confirmation suggests that active compliance does indeed exhibit this singular behavior near the jamming point.

Conclusion: Kai: So to wrap up, this paper establishes a non-Hermitian random matrix theory framework for active jammed systems, showing how nonreciprocal microscopic activity regularizes soft-mode mechanical divergences. Mira This analysis provides a microscopic interpretation of how nonequilibrium activity modifies mechanical stability near jamming.

Lev: The big implication is that this work should be seen as a first step toward a statistical mechanics theory of active compliance, which is the inverse of rigidity. Kai It’s about showing how non-reciprocal microscopic activity modifies mechanical stability near jamming, and it's a starting point for that.

Mira: But they also flag some things. They suggest extending the formulation into the non-linear regime to capture largedeformation shear-thinning and plastic yielding in active solids. Lev And they also point out that further work is needed to incorporate spatial correlations, heterogeneous contact networks, and finite dimensional fluctuations beyond this mean-field random matrix approximation.

Kai: It’s a solid first step, but they need to do more to get into the messy physics of real materials where things are spatially correlated. Mira And applying this approach to frictional granular packings could also help isolate the distinct roles of thermal or active driving versus static frictional asymmetry. Lev Yeah, that sounds like a direction we should definitely explore next when we look at these active systems.

Yukawa Institute for Theoretical Physics, Kyoto University

cond-mat.stat-mech

Submitted: 2026-07-29

Updated: 2026-11-11

Comments: 42 pages, 3 figures

License: http://creativecommons.org/publicdomain/zero/1.0/

Importance score: 87/100

The gist: Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media develops a theoretical framework for active jammed systems based on non-Hermitian random matrix theory to quantify…

Key concepts

Non-Hermitian Random Matrix Theory (RMT)
This mathematical framework is used to analyze systems where the governing equations are not symmetric, which is common in active media. It helps quantify complex spectral properties, such as eigenvalues and singular values, which directly relate to the mechanical stability and response of the jammed material.
Active Non-reciprocity
This refers to particles moving with self-propulsion forces that are not equal in both directions (asymmetric). This non-reciprocal driving acts as a perturbation on the passive structure, fundamentally changing how the system reacts to stress and preventing certain mechanical instabilities from becoming infinite.
Jamming Transition
This is the critical point where a disordered material transitions from a fluid-like state to an arrested, rigid solid. The paper investigates how activity modifies this transition, showing that non-reciprocity regularizes the soft modes (low-frequency vibrations) that usually cause divergence at this point.

Terminology

Summary

Non-Hermitian Random Matrix Theory of Jamming in Active Disordered Media develops a theoretical framework for active jammed systems based on non-Hermitian random matrix theory to quantify linear-response compliance and establish a scaling law at the jamming transition

Theoretical Foundation

The work formulates the linearized dynamical matrix as a non-Hermitian perturbation of a Wishart ensemble describing the passive contact network <ref:2607.26406#pg2> Starting from a microscopic model of active particles with nonreciprocal interactions, it uses Girko’s Hermitization together with the self-consistent Born approximation to derive a self-consistent equation for the low-frequency resolvent based on the full Marchenko–Pastur distribution <ref:2607. p. 0

Where the paper says it

Starting from a microscopic model of active particles with nonreciprocal interactions, we formulate the linearized dynamical matrix as a non-Hermitian perturbation of a Wishart ensemble describing the passive contact network. Using Girko’s Hermitization together with the self-consistent Born approximation, we derive a self-consistent equation for the low-frequency resolvent based on the full Marchenko–Pastur distribution.

26406#pg2> This approach shows that active nonreciprocity regularizes the soft-mode divergence at the jamming transition, leading to a scaling law for mechanical compliance <ref:2607. p. 0
Where the paper says it

Using Girko’s Hermitization together with the self-consistent Born approximation, we derive a self-consistent equation for the low-frequency resolvent based on the full Marchenko–Pastur distribution. We show that active nonreciprocity regularizes the soft-mode divergence at the jamming transition, leading to a scaling law for the mechanical compliance.

26406#pg2>

Microscopic Modeling

The microscopic model considers a D-dimensional disordered assembly of N active spherical particles governed by both symmetric mutual contact mechanics and asymmetric, non-equilibrium self-propulsion forces <ref:2607.26406#pg3> The passive structural Hessian matrix H0 is modeled as a shifted Wishart matrix ensemble whose baseline spectral density obeys the Marchenko–Pastur law <ref:2607. p. 1

Where the paper says it

Because E0 relies strictly on pairwise mutual contacts, the matrix elements are strictly non-zero only when the global packing fraction ϕ exceeds the conventional jamming point ϕJ as ϕ > ϕJ . Following standard granular RMT [7, 31–33], H0 can be modeled as a shifted Wishart matrix ensemble whose baseline spectral density ρ0(λ; ϕ) obeys the Marchenko–Pastur law.

26406#pg3> Active self-propulsion or RTP is introduced by assigning a directional unit vector ni to each spherical grain, which acts as an explicit non-reciprocal perturbation atop the passive, Wishart-regularized elastic matrix <ref:2607. p. 1
Where the paper says it

The distance from the isostatic threshold dictates the lower band edge position λmin(ϕ) ∝ ∆Z ∝ (ϕ − ϕJ ) 1/2 , where ∆Z is the excess coordination number. Active Driving and Non-Hermitian Formulations Active self-propulsion or RTP is introduced by assigning a directional unit vector ni to each spherical grain, mimicking Janus particles.

26406#pg3> The active driving term Fa ̸= 0 therefore acts as an explicit non-reciprocal perturbation atop this passive, Wishart-regularized elastic matrix <ref:2607. p. 1
Where the paper says it

(4) for self-consistency. The active self-propulsion term Fa ̸= 0 therefore acts as an explicit non-reciprocal perturbation atop this passive, Wishart-regularized elastic matrix, driving the dynamical spectrum into the complex plane as detailed in Appendix A.

26406#pg3>

Non-Hermitian RMT Analysis

The linearized equations of motion result in a global stability operator M that is explicitly non-Hermitian due to the violation of detailed balance by active driving <ref:2607.26406#pg4> The operator M is decomposed into a structural passive component and an asymmetric active driving matrix M:= −H0 + α0A <ref:2607. p. 2

Where the paper says it

To quantify the linear response from a statistical mechanical perspective, we consider the linearized equations of motion for the displacement fields δr := {δri} N′ i=1 as ζδr˙ = Mδr, (6) where r˙ := {dri/dt} N′ i=1, and the global stability operator M is explicitly non-Hermitian due to the violation of detailed balance by active driving. We decompose M into a structural passive component and an asymmetric active driving matrix: M := −H0 + α0A.

26406#pg4> Modeling A via the Ginibre ensemble serves as a minimal, universality-based mean-field framework to analytically evaluate the regularized complex spectrum and universal compliance scaling without relying on fine-tuned spatial details <ref:2607. p. 3
Where the paper says it

4 mapping allows us to analytically evaluate the regularized complex spectrum and universal compliance scaling without relying on fine-tuned spatial details. In this paper, thus, H0 := RT R in Eq.

26406#pg5> The self-consistent resolvent equation is derived using Girko’s Hermitization method combined with the Self-Consistent Born Approximation (SCBA) <ref:2607. p. 3
Where the paper says it

It should be noted that the microscopic activity α0 is related to the macroscopic activity α introduced in the continuum or hydrodynamic field descriptions, such as in Appendix B. Resolvent Regularization via the Complete Marchenko–Pastur Law To evaluate the mechanical rigidity of the system under non-reciprocal active driving for Nc ≫ 1, we employ Girko’s Hermitization method combined with the self-consistent Born approximation (SCBA) [40].

26406#pg5>

Universal Scaling and Crossover

The analysis yields a true, self-consistently derived universal asymptotic scaling law for active compliance at the jamming threshold: J ∼ α −6/7 0 <ref:2607.26406#pg7> This scaling arises because active non-reciprocity breaks the directional symmetry of action-reaction pairs between adjacent cells, generating complex, imaginary parts in the eigenvalue spectrum that cut off and regularize the low-frequency resolvent divergence <ref:2607. p. 5

Where the paper says it

In an active tissue, the non-reciprocal forces introduced by A break the directional symmetry of action-reaction pairs between adjacent cells. This active non-reciprocity generates complex, imaginary parts in the eigenvalue spectrum that cut off and regularize the low-frequency resolvent divergence, effectively truncating the structural soft-mode cascade and restoring a finite, dynamically stabilized mechanical rigidity to the tissue matrix.

26406#pg7> A crossover scaling form J(∆ϕ, α0) = α −6/7 0 f±(α −4/5 0 ∆ϕ) is postulated to unify the regular perturbative behavior deep in the jammed phase with the singular scaling at marginal stability <ref:2607. p. 7
Where the paper says it

Scaling Form and Crossover Function To cleanly unify the regular perturbative behavior deep in the jammed phase with the singular scaling at marginal stability, we introduce a universal homogeneous scaling form. We treat the active jamming threshold (∆ϕ = 0) as a nonequilibrium critical point where the non-reciprocal active driving strength α0 behaves as a relevant symmetrybreaking field that regularizes the unmitigated structural soft-mode divergence.

26406#pg9>

Numerical Validation and Interpretation

Numerical evaluations using finite-size non-Hermitian random matrices confirm that the singular Eq. (22) non-perturbative behavior governs the actual physical response in non-normal active media <ref:2607. p. 8

Where the paper says it

Orange filled circles depict the median compliance JSVD = 1 N Tr[(M†M) −1 ] = 1 N P k σ −2 k calculated from singular value spectra, where error bars correspond to the standard error determined via bootstrap resampling. The SVD data show remarkable agreement with the analytical self-consistent Born approximation (SCBA) prediction J ∼ α −6/7 0 (dashed blue line), confirming that the non-perturbative scaling governs the actual physical response in non-normal active media.

26406#pg10> The comparison between the naive eigenvalue compliance Jeig and the singular value sum JSVD shows remarkable agreement with the theoretical SCBA scaling Eq. p. 8
Where the paper says it

Orange filled circles depict the median compliance JSVD = 1 N Tr[(M†M) −1 ] = 1 N P k σ −2 k calculated from singular value spectra, where error bars correspond to the standard error determined via bootstrap resampling. The SVD data show remarkable agreement with the analytical self-consistent Born approximation (SCBA) prediction J ∼ α −6/7 0 (dashed blue line), confirming that the non-perturbative scaling governs the actual physical response in non-normal active media.

(22) for α0 ≤ 0. p. 0
Where the paper says it

Modeling active non-reciprocity as an asymmetric Ginibre perturbation atop a passive Wishart structural Hessian, we employ Girko’s Hermitization and the self-consistent Born approximation (SCBA) over the Marchenko– Pastur spectrum. We demonstrate that non-reciprocal activity regularizes the infrared soft-mode divergence, establishing a scaling law J ∼ α −6/7 0 at the active jamming threshold, where α0 is the activity characterizing non-reciprocal force.

03, confirming that active compliance exhibits this singular behavior <ref:2607. p. 9
Where the paper says it

This excellent agreement confirms that the SCBA properly captures the energy spectrum of the non-Hermitian operator M†M, demonstrating that the active compliance indeed exhibits the singular Eq. (22) non-perturbative behavior.

26406#pg11> The physical intuition underlying the striking non-integer exponent −6/7 at the jamming point represents a fundamental feature of non-equilibrium collective systems <ref:2607. p. 5
Where the paper says it

Solving explicitly for J yields the true, self-consistently derived universal asymptotic scaling law for active compliance at the jamming threshold: J ∼ α −6/7 0 . (22) The physical intuition underlying the striking non-integer exponent −6/7 at the jamming point represents a fundamental feature of non-equilibrium collective systems.

26406#pg7>

Future Directions

The framework suggests several promising avenues for future research, including extending the formulation into the non-linear regime to capture largedeformation shear-thinning and plastic yielding in active solids <ref:2607.26406#pg13> Further work is desirable to incorporate spatial correlations, heterogeneous contact networks, and finite-dimensional fluctuations beyond the mean-field random matrix approximation <ref:2607. p. 2

Where the paper says it

Now, we adopt the mean-field approximation where there are all-to-all couplings among the particles. We regard A as a random matrix similar to H0 as in Refs.

26406#pg13> The framework also opens avenues for applying this approach to frictional granular packings to isolate the distinct roles of thermal/active driving versus static frictional asymmetry <ref:2607.26406#pg13>

Conclusion

The present work establishes a non-Hermitian RMT framework for active jammed systems, demonstrating how non-reciprocal microscopic activity regularizes soft-mode mechanical divergences <ref:2607.26406#pg12> The analysis provides a microscopic interpretation of how nonequilibrium activity modifies mechanical stability near jamming <ref:2607. p. 0

Where the paper says it

However, how microscopic activity-induced softening at the single-cell scale coexists with non-reciprocal force transmission to dictate the structural compliance of dense, active jammed tissues remains poorly understood. Conventional descriptions of mechanical stability rely on symmetric, real Hessian operators H0, where low-frequency soft vibrational modes lead to a divergent compliance at the marginal jamming threshold [7, 19, 31–33].

26406#pg12> This work should be regarded as a first step toward a statistical-mechanical theory of active compliance, the inverse of rigidity <ref:2607.26406#pg7> Several important issues remain open, such as quantitative comparisons with numerical simulations or extending the theory beyond the mean-field random matrix approximation <ref:2607. p. 2
Where the paper says it

Now, we adopt the mean-field approximation where there are all-to-all couplings among the particles. We regard A as a random matrix similar to H0 as in Refs.

26406#pg7>

Acknowledgments

This paper was partially supported by the JSPS KAKENHI Grant No. JP26K06960 <ref:2607. p. 0

Where the paper says it

∗ hisao@yukawa. jp arXiv:2607.

26406#pg14> This work was inspired by Suravi Pal’s presentation at the mini-workshop on “the rheology of dense materials” <ref:2607.26406#pg14> The author appreciates stimulating discussions with Suravi Pal and Takeshi Kawasaki <ref:2607. p. 0
Where the paper says it

∗ hisao@yukawa. jp arXiv:2607.

26406#pg14> This paper is dedicated to Professor Hajime Tanaka to celebrate his scientific achievements <ref:2607. p. 0
Where the paper says it

∗ hisao@yukawa. jp arXiv:2607.

26406#pg14>

Appendix A

The derivation illustrates how microscopic non-reciprocal active fluxes and steric constraints generate non-Hermitian mode coupling in a lattice-based Active Exclusion Process <ref:2607.26406#pg15> The coarse-graining leads to coupled non-linear hydrodynamic equations for density and polarization fields <ref:2607. p. 2

Where the paper says it

Linearized Active Dynamics and Non-Hermitian Operator Structure Recently, theoretical frameworks connecting microscopic particle dynamics to macroscopic fluctuating hydrodynamics have provided valuable insights into non-equilibrium-driven systems [38, 39]. Starting from RTPs or active exclusion processes, coarse-graining via Dean’s method and the adiabatic elimination of fast polarization degrees of freedom yields a continuous linearized hydrodynamic equation for density fluctuations, ∂tδρ = Lhydroδρ + noise.

26406#pg15> The mapping to random matrix ensembles serves as an “effective mean-field representation” valid in strongly disordered, jammed states <ref:2607. p. 2
Where the paper says it

Instead, modeling A via the Ginibre ensemble serves as a minimal, universality-based mean-field framework. By capturing the essential feature of active media, unbiased, non-reciprocal vector interactions in the large-N limit, this random-matrix

26406#pg15>

Appendix B

The derivation provides a microscopic derivation of the activity α starting from a standard vertex-model description of tissues <ref:2607.26406#pg17> This derivation shows that the effective stiffness keff = k0 − αA, where α is a coarse-grained susceptibility encoding multiple microscopic mechanisms <ref:2607.26406#pg17>

Appendix C

The systematic, diagrammatic derivation of the self-consistent equation for the regularized compliance loop function J(z, z∗; ϵ) is provided using Girko’s Hermitization technique <ref:2607.26406#pg18> The Self-Consistent Born Approximation (SCBA) truncates the Dyson expansion at 1-loop order to yield the SCBA <ref:2607. p. 8

Where the paper says it

The SVD data show remarkable agreement with the analytical self-consistent Born approximation (SCBA) prediction J ∼ α −6/7 0 (dashed blue line), confirming that the non-perturbative scaling governs the actual physical response in non-normal active media. The Regular Perturbative Limit (x → ∞): When moving deeply into the jammed phase or taking the weak-activity limit (α0 → 0 at finite ∆ϕ), the compliance must collapse back onto the unperturbed passive baseline compliance I0(q) ∼ ∆ϕ −15/14 along with a regular expansion series in powers of α 2 0 .

26406#pg18>

Appendix D

The derivation connects the zero-frequency limit of the regularized low-frequency resolvent J to the macroscopic mechanical compliance matrix from linearized non-equilibrium dynamics <ref:2607.26406#pg20> This result explicitly proves that J serves as the macroscopic scalar compliance of the active system <ref:2607.26406#pg20>

Appendix E

The asymptotic solution of the self-consistent equation yields the scaling law J ∼ α −6/7 0 in the limit ∆ → 0 <ref:2607.26406#pg22> The derivation relies solely on the structure of the integral equation and shows that the exponent in Eq. p. 6

Where the paper says it

Singular Regimes While active driving introduces singular, non-perturbative scaling behaviors directly at the critical jamming threshold, a system situated slightly away from the transition in the jammed state (ϕ > ϕJ ) can display distinct regimes depending on the magnitude of α0. To establish a quantitative criterion separating the regular perturbative expansion from the singular active jamming scaling, we analyze the competitive structure of the denominator in the self-consistent SCBA integral equation: J(0, 0) = Z λ+ λ− dλ ρ0(λ) λ2 + α 2 0 (1 + J(0, 0)).

(E11) is robust and does not depend on complex eigenvalue details <ref:2607. p. 5
Where the paper says it

This active non-reciprocity generates complex, imaginary parts in the eigenvalue spectrum that cut off and regularize the low-frequency resolvent divergence, effectively truncating the structural soft-mode cascade and restoring a finite, dynamically stabilized mechanical rigidity to the tissue matrix. To assess the robust nature of the scaling behavior, we evaluate how the compliance J depends on the low-frequency density of states of the unperturbed passive substrate.

26406#pg22>

Appendix F

The central assumption is that the linearized dynamical operator of an active jammed system can be represented as Eq. (7) <ref:2607. p. 0

Where the paper says it

∗ hisao@yukawa. jp arXiv:2607.

26406#pg23> The mapping of the Wishart eigenvalues to the passive baseline spectrum via the linear shift λ = Z/D(λMP − e) yields the structural density of states ρ0(λ) <ref:2607. p. 3
Where the paper says it

More explicitly, the baseline passive eigenvalue (λ) distribution ρ0(λ) of the structural matrix H0 is governed by the Marchenko–Pastur law [48, 49] in the limit Nc → ∞: ρ0(λ) = 1 2πqλ p (λ+ − λ)(λ − λ−), (9) where the upper and lower edges of the passive structural spectrum are given by λ± = (1 ± √q) 2 . Near the standard mechanical jamming point (q → 1), the lower bound approaches zero (λ− → 0), and the density of states exhibits the classic low-frequency divergence ρ0(λ) ∼ λ −1/2 , which corresponds to the emergence of soft, marginally stable structural modes.

26406#pg24> This leads to the characteristic scaling ρ0(λ) ∼ λ −1/2 for low-frequency soft modes <ref:2607. p. 4
Where the paper says it

See Appendix F for the behavior near the jamming point. Asymptotic Scaling at the Jamming Point (q → 1, λ− → 0) (see Appendix E): At the mechanical jamming threshold (q = 1), the lower bound of the passive structural spectrum vanishes (λ− = 0), and the low-frequency density of states exhibits the characteristic marginal divergence: ρ0(λ) ≈ 1 πσ0 λ −1/2 (for λ → 0).

26406#pg24> The effective active coupling matrix A is modeled by the Ginibre ensemble as a minimal universality-class model capturing dense, spatially uncorrelated non-Hermitian driving <ref:2607. p. 2
Where the paper says it

Instead, modeling A via the Ginibre ensemble serves as a minimal, universality-based mean-field framework. By capturing the essential feature of active media, unbiased, non-reciprocal vector interactions in the large-N limit, this random-matrix

26406#pg24>

Appendix G

The analytical evaluation of the integral I1(q) yields an explicit, closed-form expression for P3(e, q) <ref:2607.26406#pg25> This expression is derived by defining the function in the numerator as f(x) = √((λ+−x)(x−λ−))/x and using Cauchy’s residue theorem <ref:2607. p. 6

Where the paper says it

Applying the same variable substitution gives: I1(q) = Z λ− λ+ ρMP(x) Z D (x − e) 4 dx = D Z 4 1 2πq Z λ+ λ− p (λ+ − x)(x − λ−) x(x − e) 4 dx. (28) Evaluating this integral requires computing the higher-order derivative contributions from the 4th-order pole at x = e using a residue expansion.

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 AI can model active jammed systems by representing non-reciprocal microscopic activity as an asymmetric Ginibre perturbation atop a passive Wishart structural Hessian, allowing for analytical evaluation of the regularized complex spectrum.

  2. Improved AI can predict the mechanical compliance J at the active jamming threshold using the derived scaling law J ∼ alpha - 6/70, providing a robust prediction for non-equilibrium collective systems.

  3. Improved AI can distinguish between different physical regimes by implementing a crossover criterion based on the comparison of competing scales, specifically when lambda squared equals alpha squared zero plus one plus J.

  4. Improved AI can predict the required excess packing fraction for the regular perturbative expansion to be valid, identifying a critical threshold where ϕ minus ϕJ exceeds alpha four fifth zero.

  5. Improved AI can perform numerical simulations of non-Hermitian random matrices M = -H0 + alpha0A to confirm that the singular value sum JSVD follows the theoretical SCBA scaling J ∼ alpha - 6/70 for active compliance.

Abstract

We develop a non-Hermitian random-matrix description of active jammed systems in which a passive Wishart operator is perturbed by non-reciprocal Ginibre couplings. Within a Hermitized self-consistent Born approximation (SCBA), we derive a scalar susceptibility equation predicting J aboutα 0-6/7 at the marginal point. Direct random-matrix calculations confirm that the mechanically relevant response is governed by singular values rather than complex eigenvalues. They further reveal strong finite-size fluctuations: near marginality, a large fraction of the total susceptibility is typically carried by the smallest singular value. These results distinguish the smooth large- N c SCBA response from the extreme singular-value statistics of finite non-Hermitian systems.

Sources

Related papers