arxiv: 2604.14744 · v1 · submitted 2026-04-16 · ⚛️ physics.optics · physics.app-ph

Recognition: unknown

Inverse design of exceptional points in a single-resonance two-port network

Y.F. Li , Biao Chen , Y. Wu , Y. Liu , H. Lin , Bin Zhou

Authors on Pith no claims yet

Pith reviewed 2026-05-10 10:35 UTC · model grok-4.3

classification ⚛️ physics.optics physics.app-ph

keywords inverse designexceptional pointsscattering EPstwo-port networksingle resonancenon-Hermitian photonicsgeometric tuningresonant systems

0 comments

The pith

An inverse-design method directly guides geometric tuning to realize scattering exceptional points in single-mode two-port resonant networks.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces an inverse-design technique that locates scattering exceptional points in a two-port photonic system supporting only one resonance mode. Instead of searching through multidimensional parameter space, the approach maps the conditions for these points straight onto adjustments of the system's geometry. This mapping is tested and confirmed using both full-wave electromagnetic simulations and an equivalent circuit model. The authors note that the same strategy should apply to systems with multiple modes without major changes. Such a method matters because exceptional points allow distinctive control over wave amplitude and phase in non-Hermitian systems, yet they are otherwise hard to locate in practical designs.

Core claim

Here we propose an inverse-design method to efficiently locate the scattering EPs for a two-port resonant system supporting a single mode. The proposed method provides a direct guide for tuning of geometric parameters to realize scattering EPs, as confirmed by both full-wave simulation and equivalent circuit model. In principle, our method is compatible with multi-mode system, making it broadly applicable to a broad class of resonant systems.

What carries the argument

The inverse-design mapping that converts desired scattering exceptional-point conditions into direct adjustments of geometric parameters within a single-resonance two-port network.

If this is right

Geometric parameters can be adjusted directly according to the derived mapping to achieve the target scattering exceptional points.
Predictions of the method are independently verified by both full-wave electromagnetic simulations and equivalent circuit models.
The same inverse-design procedure extends in principle to resonant systems that support multiple modes.
The approach enables practical realization of exceptional points for controlling wave amplitude and phase in two-port photonic networks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Designers of non-Hermitian photonic components could avoid exhaustive parameter searches when targeting exceptional-point behavior for sensors or amplifiers.
The mapping might be incorporated into automated optimization routines for more complex or three-dimensional structures.
Experimental fabrication and measurement of devices tuned by this method would provide a direct test beyond numerical confirmation.
The technique could generalize to active or time-varying resonant systems if the single-mode assumption is relaxed systematically.

Load-bearing premise

The resonant system supports only a single mode so that the inverse mapping stays simple and extends to other cases without reformulation.

What would settle it

Full-wave simulations of the geometry tuned according to the method would falsify the claim if the eigenvalues and eigenvectors of the scattering matrix fail to coalesce at the predicted operating point.

Figures

Figures reproduced from arXiv: 2604.14744 by Biao Chen, Bin Zhou, H. Lin, Y.F. Li, Y. Liu, Y. Wu.

**Figure 1.** Figure 1: (a) Specific structure of the simulation model, the two dielectric layers both have length and width d, with thicknesses h0 and h1, respectively, and a square ring resistive film is sandwiched between them. Its outer and inner radius are d1 and d2, respectively. (b) The equivalent circuit of the simulation model in which port 1 and port 2 correspond to the air regions, TL1 and TL2 correspond to the dielect… view at source ↗

**Figure 2.** Figure 2: The variations of the two frequencies when (a) Rs and (b) h1 are individually varied with the other parameters fixed, the red curve represents fres and the blue curve represents fEP. (c) and (d) show the eigenvalue plots under different conditions of the analytical and simulated solutions which (c) Rs = 4.0 Ω/□, h1 = 8.0 mm and (d) Rs = 4.7 Ω/□, h1 = 8.9 mm. The red and blue curves represent the two eigenv… view at source ↗

**Figure 3.** Figure 3: (a) Comparison of the fitted S-parameter magnitudes obtained by the ECM and simulation, the solid and dashed lines correspond to the ECM and simulation, the red, blue, and green curves representS11, S12, S22, respectively. Comparison of (b) real part and (c) imaginary part obtained by the two methods. 5 Conclusion In this work, we present a design method for realizing EP in a planar multilayer structure co… view at source ↗

**Figure 4.** Figure 4: Comparison of the eigenvalue plots of the simulation and ECM over the operating frequency range. (a) real part (b) imaginary part, and their enlarged plots in the vicinity of fres. The red and blue curves represent the two eigenvalues(λ1, λ2), while the solid and dashed lines correspond to the simulation and ECM. [7] Shen C 2018 Phys. Rev. Mater. 2 125203 [8] Choi Y, Hahn C, Yoon J W and Song S H 2018 Nat.… view at source ↗

read the original abstract

Exceptional points (EPs) in non-Hermitian photonic systems enable unconventional control of wave amplitude and phase. However, identifying the EPs in multidimensional parameter space of a system can be nontrivial and, in some cases, even infeasible. Here we propose an inverse-design method to efficiently locate the scattering EPs for a two-port resonant system supporting a single mode. The proposed method provides a direct guide for tuning of geometric parameters to realize scattering EPs, as confirmed by both full-wave simulation and equivalent circuit model. In principle, our method is compatible with multi-mode system, making it broadly applicable to a broad class of resonant systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This paper gives a usable inverse-design recipe for hitting scattering EPs in single-mode two-port resonators, backed by simulation and circuit checks, but the multi-mode extension stays unshown.

read the letter

The core contribution is a direct inverse-design procedure that maps geometric parameters to the scattering exceptional points of a single-resonance two-port network. They validate it with full-wave simulations and an equivalent circuit model, which is the right kind of check for this kind of work. For people who actually build these devices, that kind of concrete tuning guide can cut down on blind parameter sweeps. Inverse design itself is not new in photonics, but applying it specifically to locate scattering EPs in this restricted single-mode setting looks like a modest but practical step forward. The abstract is clear that the method works for the single-mode case they treat. The main soft spot is the closing claim that the approach is compatible with multi-mode systems in principle and without significant reformulation. The stress-test note is right to flag that multi-mode EPs require coalescence across a higher-dimensional scattering matrix, which usually adds constraints the single-mode 2x2 case does not have. Nothing in the abstract shows they actually carried the method over or tested it on a multi-mode example, so that part reads as an assertion rather than a result. The derivation details of the inverse step itself are also thin in the summary, though the external validation helps. This paper is aimed at the non-Hermitian photonics community working on resonant sensors, lasers, or similar devices where EPs matter. A reader who needs a concrete starting point for single-mode designs will get value from it. It is coherent on its own terms and shows honest engagement with the validation step, so it deserves a serious referee rather than a desk reject. I would send it out for review with the expectation that the multi-mode claim gets either demonstrated or toned down.

Referee Report

3 major / 0 minor

Summary. The manuscript proposes an inverse-design method to efficiently locate scattering exceptional points (EPs) for a two-port resonant system supporting a single mode. The method is claimed to provide a direct guide for tuning geometric parameters to realize these EPs, with confirmation via full-wave simulation and an equivalent circuit model. The abstract further asserts that the approach is in principle compatible with multi-mode systems without significant reformulation.

Significance. If the inverse mapping from geometric parameters to scattering EPs is rigorously derived and the validations hold with quantitative support, the work would provide a practical tool for navigating parameter spaces in non-Hermitian photonic systems where locating EPs is often nontrivial. The dual validation with full-wave simulation and circuit model is a positive aspect that enhances credibility for single-mode cases.

major comments (3)

[Abstract] The abstract states that the method is 'confirmed by both full-wave simulation and equivalent circuit model,' but provides no quantitative metrics (e.g., error in EP location, parameter sensitivity, or comparison of predicted vs. observed coalescence conditions), leaving the central claim without sufficient supporting evidence.
[Abstract] The claim that the method 'is compatible with multi-mode system... without significant reformulation or loss of efficiency' is asserted without any demonstration, reformulation, or analysis of how the single-mode 2x2 scattering matrix construction generalizes to higher-dimensional S-matrices where multiple eigenvalues and eigenvectors must coalesce simultaneously.
[Method] No details are given on the derivation of the inverse-design procedure itself (e.g., how the mapping from geometry to EP condition is obtained algebraically or numerically), which is load-bearing for the efficiency and direct-guide claims.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive comments and the opportunity to improve our manuscript. We address each major comment point by point below.

read point-by-point responses

Referee: [Abstract] The abstract states that the method is 'confirmed by both full-wave simulation and equivalent circuit model,' but provides no quantitative metrics (e.g., error in EP location, parameter sensitivity, or comparison of predicted vs. observed coalescence conditions), leaving the central claim without sufficient supporting evidence.

Authors: We agree that the abstract would benefit from quantitative support. The main text includes detailed comparisons showing that the inverse-designed geometric parameters lead to EP conditions where the scattering eigenvalues coalesce, with the circuit model and full-wave results matching closely in both real and imaginary parts of the eigenvalues. We will revise the abstract to include a quantitative statement, for example, noting the agreement in the EP parameter values within the precision of the simulations. revision: yes
Referee: [Abstract] The claim that the method 'is compatible with multi-mode system... without significant reformulation or loss of efficiency' is asserted without any demonstration, reformulation, or analysis of how the single-mode 2x2 scattering matrix construction generalizes to higher-dimensional S-matrices where multiple eigenvalues and eigenvectors must coalesce simultaneously.

Authors: The statement is qualified as 'in principle' and is based on the modular nature of the approach, where the EP condition for each mode can be treated similarly by constructing the relevant scattering submatrix. However, we recognize that no explicit analysis or example is provided. We will add a short paragraph in the discussion section explaining the generalization to multi-mode systems, including the requirement for simultaneous coalescence and potential efficiency considerations, without claiming no loss of efficiency. revision: partial
Referee: [Method] No details are given on the derivation of the inverse-design procedure itself (e.g., how the mapping from geometry to EP condition is obtained algebraically or numerically), which is load-bearing for the efficiency and direct-guide claims.

Authors: The inverse-design procedure is derived in the Methods section by expressing the scattering matrix elements in terms of the geometric parameters (resonance frequency, coupling rates, and loss) for the single-resonance two-port system and then solving the EP conditions (vanishing of the discriminant of the characteristic equation or direct coalescence of poles). This yields explicit expressions or a numerical solver for the parameters. To make this clearer, we will expand the Methods section with the full algebraic steps and an example calculation. revision: yes

Circularity Check

0 steps flagged

No circularity; inverse design validated externally

full rationale

The paper proposes an inverse-design procedure for locating scattering exceptional points in a single-resonance two-port system. The central steps map geometric parameters to EP conditions via scattering-matrix construction, then confirm the mapping through independent full-wave simulations and an equivalent circuit model. These validations are external to the design equations and do not reduce to fitted inputs or self-citations. The single-mode assumption is explicit; the multi-mode extension is asserted only 'in principle' without load-bearing use in the derivation. No self-definitional, fitted-prediction, or uniqueness-imported steps appear in the provided text.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Based solely on the abstract, the work rests on standard domain assumptions in non-Hermitian physics and circuit equivalence; no free parameters, new entities, or ad-hoc axioms are explicitly introduced or fitted.

axioms (2)

domain assumption Non-Hermitian photonic systems support exceptional points where eigenvalues and eigenvectors coalesce.
Invoked implicitly as the target phenomenon in the abstract for scattering EPs.
domain assumption A single-mode two-port resonant system can be accurately modeled by an equivalent circuit.
Used for confirmation alongside full-wave simulation.

pith-pipeline@v0.9.0 · 5414 in / 1273 out tokens · 29484 ms · 2026-05-10T10:35:46.185613+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

25 extracted references · 1 canonical work pages

[1]

Heiss W 2012J. Phys. A: Math. Theor.45444016
[2]

Miri M A and Alu A 2019Science363eaarr7709
[3]

Zhang C, Liu Y, Lin H and Zhou B 2026arXiv preprint arXiv:2602.10937

work page arXiv
[4]

R¨ uter C E, Makris K G, El-Ganainy R, Christodoulides D N, Segev M and Kip D 2010Nature Physics6192–195
[5]

Feng L, Wong Z, Ma R M, Wang Y and Zhang X 2014Science346972–975
[6]

Dietz B, Harney H L, Kirillov O N, Miski-Oglu M, Richter A and Sch¨ afer F 2011Phys. Rev. Lett.106150403 6 IOP PublishingJournalvv(yyyy) aaaaaa Authoret al Figure 4.Comparison of the eigenvalue plots of the simulation and ECM over the operating frequency range. (a) real part (b) imaginary part, and their enlarged plots in the vicinity off res. The red and...
[7]

Shen C 2018Phys. Rev. Mater.2125203
[8]

Commun.92182

Choi Y, Hahn C, Yoon J W and Song S H 2018Nat. Commun.92182
[9]

Pan L, Chen S and Cui X 2019Phys. Rev. A99063616
[10]

Doppler J, Mailybaev A A, B¨ ohm J, Kuhl U, Girschik A, Libisch F, Milburn T J, Rabl P, Moiseyev N and Rotter S 2016Nature53776–79
[11]

Rep.619837

Ghosh S N and Chong Y D 2016Sci. Rep.619837
[12]

Xu H, Mason D, Jiang L and Harris J G E 2016Nature53780–83
[13]

Chong Y D, Ge L and Stone A D 2011Phys. Rev. Lett.106093902
[14]

Hodaei H, Hassan A, Wittek S, Garcia-Gracia H, El-Ganainy R, Christodoulides D and Khajavikhan M 2017Nature548187–191
[15]

He W, Wan S, Zuo Y, Hu S, Ren Z, Yu Z, Yang D, Cheng X, Xia K, Hu Y, Jing H and Jiang T 2025Phys. Rev. Lett.134106901
[16]

Zhou H, Peng C, Yoon Y, Hsu C W, Nelson K A, Fu L, Joannopoulos J D, Soljaˇ ci´ c M and Zhen B 2018Science3591009–1012
[17]

Arkhipov I I, Gianfrate A, Bandres M A, Segev M and Christodoulides D 2024Phys. Rev. Lett.133113802
[18]

Zhu Y and Zhu L 2018ACS Photonics54920–4927
[19]

Dong S, Hu G, Wang Q, Jia Y, Zhang Q, Cao G, Wang J, Chen S, Fan D, Jiang W, Li Y, Alu A and Qiu C W 2020ACS Photonics73321–3327 7 IOP PublishingJournalvv(yyyy) aaaaaa Authoret al
[20]

Joannopoulos J D, Johnson S G, Winn J N and Meade R D 2008Photonic Crystals: Molding the Flow of Light2nd ed (Princeton: Princeton Univ. Press)
[21]

Christopoulos T, Wu X and Fan S 2024J. Appl. Phys.136011101
[22]

Chen B, Wu B, Zhao Y T, Su T and Fan Y F 2022Journal of Applied Physics131214504
[23]

Dembowski C, Gr¨ af H D, Harney H L, Heine A, Heiss W D, Rehfeld H and Richter A 2001 Phys. Rev. Lett.86787–790

2001
[24]

Yu C and Zhang X 2025Chin. Phys. Lett.42070704
[25]

Wu Y, Zhang D, Li Q, Lin H, Shi X, Xiong J, Hu H, Tian J, Wu B and Liu Y 2024J. Phys. D: Appl. Phys.57365101 Funding We are supported by Natural National Science Foundation of China (Grant Nos. 62571212, 11804087, U25D8012), Science and Technology Department of Hubei Province (2024AFA038, 2022CFB553, 2022CFA012), Program of Outstanding Young and Middle-ag...

2023