Pith. sign in

REVIEW 3 major objections 3 minor 49 references

A feedback-controlled lattice of paired mechanical oscillators implements a tunable SU(2) gauge field, and Wilson-loop measurements show the loop operators fail to commute.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 15:39 UTC pith:V6NPNFZA

load-bearing objection A genuinely new experimental platform and a mostly sound non-Abelian Wilson-loop measurement, but the headline claim lacks the error analysis needed to rule out loss-induced false positives. the 3 major comments →

arxiv 2607.18215 v1 pith:V6NPNFZA submitted 2026-07-20 cond-mat.mes-hall cond-mat.otherphysics.class-phquant-ph

Non-Abelian Gauge Field Mechanics

Ivan Velkovsky , Carlos Camacho , Tomoki Ozawa , Hannah Price , Bryce Gadway This is my paper
classification cond-mat.mes-hall cond-mat.otherphysics.class-phquant-ph
keywords non-Abelian gauge fieldWilson loopsynthetic gauge fieldactive mechanical latticefeedback controlnon-Hermitian skin effectSU(2) holonomytwo-oscillator pseudo-spin
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper tries to show that a purely classical, feedback-controlled mechanical lattice can carry a genuinely non-Abelian gauge field—a field whose holonomies depend on the order in which paths are traversed—and that the same platform can be used to explore non-Hermitian versions of such fields. Each lattice site is a pair of oscillators encoding a pseudo-spin, and real-time measurement and feedback impose spin-dependent hopping matrices. The authors extract Wilson-loop magnitudes and argue that a measured value |W4|≠2 for a specially chosen commutator loop proves the loop operators do not commute, hence the gauge field is genuinely non-Abelian. They further show that non-reciprocal couplings make the Wilson loop direction-dependent and that tuning gauge angles can switch skin-mode localization between the two ends of a one-dimensional chain. A sympathetic reader would care because this offers a programmable tabletop setting for gauge-field physics usually studied in quantum or photonic systems.

Core claim

On its own terms, the paper establishes that an active mechanical lattice—oscillators coupled only through real-time measurement and feedback—can implement a spin-1/2 hopping model with tunable non-Abelian link variables X=e^{iασ_y}, Y=e^{iβσ_x}. Using four initial pseudo-spin states and single-bond transfer sequences, the authors extract the magnitudes of Wilson loops for the elementary plaquette, a double plaquette, their product, and the commutator loop U4=U2†U1†U2U1. Because an SU(2) loop operator with |tr(U)|=2 must be ±identity, a measured |W4|≠2 directly implies U2U1≠U1U2, i.e., genuine non-Abelianness. They then break reciprocity and use the same apparatus to demonstrate direction-de

What carries the argument

The load-bearing device is the feedback-imposed link: each bond is engineered as a prescribed 2×2 SU(2) rotation on the two-oscillator pseudo-spin, switched on for one calibrated tunneling time. The diagnostic is the Wilson-loop magnitude criterion—in SU(2), |tr(U)|=2 forces U=±1, so the commutator loop U4=U2†U1†U2U1 has |W4|≠2 exactly when the elementary loop operators fail to commute. A spanning-tree argument shows that on the six-site lattice all loops are generated by U1, U2, and their inverses, so checking these traces decides whether the whole finite lattice is Abelian or genuinely non-Abelian. For the non-Hermitian models, non-reciprocal link matrices break the Hermitian relation betw

Load-bearing premise

The Wilson-loop extraction treats every single-bond feedback transfer as an ideal unitary rotation completed in one calibrated tunneling time, with losses and transfer errors small enough that a measured |W4|≠2 is read as noncommutation rather than as systematic suppression.

What would settle it

A controlled-loss test: choose an intentionally Abelian configuration where |W4| should equal 2, then deliberately increase oscillator damping and see whether the measured |W4| falls below 2. If it does, the magnitude-only threshold is not a clean non-Abelian signature. The decisive check is to reconstruct the full 2×2 matrices U1 and U2 (not just traces) from the four-state tomography and verify directly that [U2,U1]≠0.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Active mechanical lattices become a reconfigurable tabletop platform for non-Abelian gauge fields, with arbitrary rotation angles and non-reciprocal hopping strengths.
  • The Wilson-loop magnitude criterion provides a practical gauge-invariant test for genuine non-Abelianness in finite synthetic lattices.
  • Non-Hermitian non-Abelian gauge potentials can make Wilson loops direction-sensitive, a signature unavailable in Hermitian systems.
  • Tuning the gauge angles gives direct control over the direction and localization of energy transport via the non-Hermitian skin effect in one dimension.
  • The platform's programmability extends to nonlinear or amplitude-dependent couplings, opening a route toward dynamical gauge fields where the wave itself alters the holonomy.

Where Pith is reading between the lines

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

  • The same feedback protocol should transfer to other classical wave platforms, such as acoustic, photonic, or electronic arrays, because it requires only real-time measurement and actuation rather than quantum coherence.
  • The |W4|≠2 test is magnitude-only; reconstructing the full 2×2 loop matrices from the same four-state tomography would provide a stronger, direct check of U2U1≠U1U2 and would help separate loss-induced suppression from true noncommutation.
  • Direction-dependent Wilson loops in non-Hermitian settings may serve as a gauge-invariant order parameter for non-Hermitian topology, potentially connecting to point-gap winding in higher dimensions.
  • In larger lattices, iterating these feedback-imposed SU(2) links could simulate lattice gauge theories or topological band structures with position-dependent gauge fields, including dynamical back-action.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The manuscript reports an experimental realization of tunable SU(2) gauge fields in a measurement-and-feedback-coupled classical mechanical oscillator array. Each lattice site is encoded by two oscillators, and feedback imposes spin-dependent hopping matrices X=e^{iασ_y}, Y=e^{iβσ_x}. Wilson-loop magnitudes are extracted by a sequence of four initial-state/tomography-like measurements and used to infer a genuinely non-Abelian field from |W4|=|Tr(U2†U1†U2U1)|≠2. The same platform is then used to realize a non-Hermitian 2D model with direction-dependent Wilson loops and a 1D non-Abelian Hatano-Nelson model in which the gauge angles (θ_R,θ_L) switch the non-Hermitian skin effect between the two ends of an open 7-site chain.

Significance. If the measurement systematics are under control, this would be a notable experimental milestone: a programmable active-mechanical platform that realizes non-Abelian holonomies and, simultaneously, non-Hermitian effects, with the Wilson-loop magnitude as a gauge-invariant observable. The finite-size basis-loop reduction in the Supplemental Material is an elegant way to reduce the infinite set of loop operators to U1 and U2 and to justify the |W4|≠2 criterion. The independent calibration of J via two-site tunneling, the use of four tomographic initial states, and the direct time-resolved observation of skin-mode switching are all strengths. However, the central experimental evidence is Wilson-loop magnitudes, and these are currently presented without uncertainty bars, repetition statistics, or quantitative control measurements; the extraction formula assumes ideal unitary bond rotations. The non-Abelian claim therefore needs additional quantitative support before it can be considered established.

major comments (3)
  1. [SM 'Wilson Loop Measurement Protocol'; Eq. (S10); Fig. 1(d)] The load-bearing criterion for a genuinely non-Abelian field is the sharp threshold |W4|≠2 versus |W4|=2. The extraction formula (S10) is derived for ideal unitary bond rotations; any per-bond transfer loss, incomplete rotation, or spin-dependent damping enters the tomographic combinations A,B,C,D and suppresses the inferred |W|. The SM explicitly lists 'incomplete transfer of energy during a single rotation' and 'accumulated differences in energy loss' as error sources, but no error bars, repetition statistics, or per-bond efficiency calibration are reported. A configuration with true |W4|=2 (e.g. α=0 or β=0, where one link variable is the identity) could then be measured as |W4|<2, exactly the signature claimed for non-Abelianness. Please supply uncertainty estimates, state whether final states are renormalized before applying Eq. (S10), report control measurements on lines where |W4|
  2. [SM 'OVER VIEW', 'Wilson Loop Measurement Protocol', and 'Derivation of basis loops'] The Supplemental Material is not in a complete, checkable state. The Overview has empty section references ('In Section , ...'), the derivation of the central extraction formula Eq. (S10) is not given, and the numerical-simulation model is not fully specified (ideal model versus model including dissipation/feedback). An unedited thesis fragment remains in the SM (the caption 'Figure 5.8: Selected dynamics... used to produce Fig. 5.7' and the text '5.3 Conclusions ... Here I have shown ...'). The basis-loop reduction is presented, but because the Wilson-loop extraction is the basis for the main non-Abelian claim, the absence of a complete derivation and the presence of unedited fragments compromise reproducibility. This must be fixed in revision.
  3. [SM Eq. (8); Fig. 3(e,f)] The 1D skin-effect demonstration relies on long-time bias measurements while a nonlinear energy-stabilizing feedback of the form Hstab=-igD Σ c†c is applied, with D depending on the total squared amplitude. The text states that this feedback does not affect the normalized energy distribution and that numerical simulations confirm this, but no supporting data or derivation are shown. The experimental bias map in Fig. 3(f) is also presented without error bars and is systematically offset from the numerics, an effect attributed to damping. Please quantify the sensitivity of the measured bias to the stabilizing gain g, provide repetition statistics, and either justify more rigorously or demonstrate numerically that the normalized spatial distribution is independent of this feedback.
minor comments (3)
  1. [General] Typos: 'geninuely' appears repeatedly and should be 'genuinely'; 'tuneable' is used inconsistently.
  2. [References] Ref. [43] and Ref. [49] are the same work (Chuang and Nielsen, J. Mod. Opt. 44, 2455 (1997)); one duplicate should be removed.
  3. [SM Figure 6] The caption of the supplemental dynamics figure still uses thesis-chapter numbering ('Figure 5.8') and refers to 'Fig. 5.7'; renumber and remove the fragment beginning '5.3 Conclusions'.

Circularity Check

0 steps flagged

No circularity: the central Wilson-loop claim rests on an independent tomographic measurement, not on a fitted or self-citational reduction; acknowledged loss errors are a validity concern, not a circular step.

full rationale

The central claim that the feedback-imposed bonds implement non-Abelian SU(2) link variables is supported by an independent tomographic measurement, not by a fit or by definition. The Wilson-loop magnitudes in Fig. 1 are extracted from four separately prepared initial states (SM Eq. S9) and the tomographic combination of Eq. S10; this reconstructs |tr U| from measured final-state amplitudes and does not assume the value being claimed. The hopping amplitude J is calibrated from direct two-site tunneling dynamics ('The experimental hopping rates J are calibrated based on direct measurement of two-site tunneling dynamics'), not from the Wilson-loop data, so the non-Abelian signature is not a fitted parameter renamed as a prediction. The inference from |W4| != 2 to noncommuting loop operators is a theorem established in the SM via the spanning-tree basis, and the platform mapping is rederived in Eqs. S1-S8 rather than resting solely on self-citations. The 1D skin-effect result compares experimental dynamics with numerical eigenstate bias from Eq. 7; no parameter is adjusted to enforce agreement. The paper's acknowledged error sources ('incomplete transfer of energy during a single rotation as well as accumulated differences in energy loss') are a real measurement-validity concern — uncalibrated per-bond loss could suppress |W| and mimic a non-Abelian signal — but this is a systematic-error/correctness issue, not a circular reduction: the paper does not define any quantity in terms of the target result, nor fit a parameter to force the outcome. Reliance on the group's prior platform papers (Refs. 24-29) is background capability, and the present SM contains the derivation of the mapping. No circular step can be identified from the paper's own equations.

Axiom & Free-Parameter Ledger

3 free parameters · 6 axioms · 0 invented entities

The central claims rest on the classical-to-tight-binding mapping from prior work and on the RWA assumption. No new particles, forces, or conserved quantities are introduced; the pseudo-spin is an experimental encoding of two oscillator modes. The gauge angles α, β, θ are scanned experimental controls rather than fitted parameters. The genuinely fitted or hand-set quantities are calibration constants and the stabilization gain.

free parameters (3)
  • Empirical hopping rate J = J/2π ≈ 32 mHz
    Calibrated from two-site tunneling dynamics and used to set the single-bond transfer time in all Wilson-loop measurements and to parameterize numerical simulations. This is an independent calibration, not fitted to the Wilson-loop results, but central to the measurement protocol.
  • Energy-stabilization gain g = not specified
    Hand-tuned feedback strength introduced in Supplemental Eq. 8 to keep total oscillator energy near target during 60 s NHSE runs. The paper states it does not alter normalized spatial dynamics, but the value is unreported and its effect is only numerically asserted.
  • Per-oscillator calibration constants = not reported per oscillator
    Signal scale factors, x/p rotation angles, and damping-cancellation feedback are adjusted per oscillator via ringdown and self-feedback calibration. These are needed to transform acceleration and jerk signals into usable x and p proxies, but individual values are not listed.
axioms (6)
  • domain assumption Feedback forces are weak and the oscillator frequency is the largest frequency scale, so the rotating-wave approximation applies and Newton's equations map onto the target tight-binding Hamiltonian.
    SM 'Theoretical Mapping', Eqs. S6-S8: if feedback is not sufficiently weak or ω is not dominant, the co-rotating approximation fails and the realized dynamics differ from Eqs. 1, 4, and 5.
  • domain assumption Classical complex amplitudes α_iσ obey the same equations as bosonic annihilation operators for the feedback-coupled Hamiltonian.
    SM Eqs. S4-S8: the entire hopping and pseudo-spin picture relies on treating mechanical amplitude and phase as a bosonic wavefunction.
  • domain assumption For the 6-site Hermitian lattice, every loop operator based at site A is a product of U1, U2 and their inverses via the spanning-tree basis.
    SM 'Derivation of basis loops': this ensures the four Wilson-loop magnitudes in Fig. 1 are sufficient to diagnose Abelian versus non-Abelian behavior in this finite lattice.
  • standard math For an SU(2) loop operator, |Tr(U)|=2 iff U=±I, and |Tr(U2† U1† U2 U1)|≠2 implies [U2,U1]≠0.
    Used in the main text and SM to convert measured Wilson-loop magnitudes into a logical inference about genuine non-Abelianness.
  • domain assumption In non-Hermitian systems, net winding of the PBC spectrum around OBC eigenvalues indicates the presence and direction of the skin effect.
    Main text Fig. 3(b-d) and Refs [32,45]: used to connect the measured bias and spectra to the non-Hermitian skin effect.
  • domain assumption The uniform nonlinear energy-stabilizing feedback of Supplemental Eq. 8 leaves the normalized spatial population dynamics unchanged.
    SM '1D Non-Abelian Hatano-Nelson Dynamics': the authors state this is confirmed numerically; if false, the 60 s B_expt could reflect stabilization artifacts rather than intrinsic NHSE.

pith-pipeline@v1.3.0-alltime-deepseek · 18593 in / 16230 out tokens · 140615 ms · 2026-08-01T15:39:29.072313+00:00 · methodology

0 comments
read the original abstract

Non-Abelian gauge fields play a key role in describing the behavior of particles whose motion is coupled to internal degrees of freedom, such as their spin. Here, we experimentally realize a tuneable non-Abelian gauge field in an active mechanical lattice by using pairs of oscillators to encode a local pseudo-spin for each site, with inter-site spin-dependent couplings engineered via real-time measurement and feedback. We experimentally extract Wilson-loop observables in our set-up and hence demonstrate that we can create a genuinely non-Abelian gauge field. We then exploit the controllability of our mechanical lattice to engineer non-reciprocal hoppings to explore non-Hermitian non-Abelian gauge potentials. For a two-dimensional (2D) lattice, we demonstrate that the non-Hermiticity can manifest in direction-dependent Wilson loops for a single plaquette, while for a one-dimensional (1D) system, we show that a non-Abelian gauge potential can switch the localization of non-Hermitian skin modes between opposite ends of a chain. Our work establishes active mechanical lattices as a flexible and programmable platform for probing non-Abelian gauge fields and exploring their interplay with non-Hermitian dynamics.

Figures

Figures reproduced from arXiv: 2607.18215 by Bryce Gadway, Carlos Camacho, Hannah Price, Ivan Velkovsky, Tomoki Ozawa.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: ]. Hence, the Wilson loops need not satisfy WCW = W∗ CCW . Explicitly, they are given by WCW/CCW = 2e iθD (cos θL cos θR cos θU ± sin θL sin θR sin θU ) [33]; thus the CW/CCW contrast is controlled by θL, θR, θU , which are associated with the non-commuting Pauli rotations. To show this experimentally, we use the above Wilson￾loop measurement procedure to extract |WCW | and |WCCW | for a single-plaquette o… view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. ( [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. The energy spectrum of our 6-site non-Abelian Hermitian ladder [c.f. Eq. 1 in the main text], as extracted using a [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. A choice of spanning tree for our 6-site mechanical [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 5.8
Figure 5.8. Figure 5.8: Selected dynamics of the 1D non-Hermitian chain with non-reciprocal FIG. 6. Experimentally-measured dynamics in the 1D non-Abelian Hatano-Nelson chain, starting from an initial excitation of [PITH_FULL_IMAGE:figures/full_fig_p014_5_8.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

49 extracted references · 1 linked inside Pith

  1. [1]

    Demler, W

    E. Demler, W. Hanke, and S.-C. Zhang, Rev. Mod. Phys. 76, 909 (2004)

    2004

  2. [2]

    Nayak, S

    C. Nayak, S. H. Simon, A. Stern, M. Freedman, and S. Das Sarma, Rev. Mod. Phys.80, 1083 (2008)

    2008

  3. [3]

    C. N. Yang and R. L. Mills, Phys. Rev.96, 191 (1954)

    1954

  4. [4]

    Osterloh, M

    K. Osterloh, M. Baig, L. Santos, P. Zoller, and M. Lewenstein, Phys. Rev. Lett.95, 010403 (2005)

    2005

  5. [5]

    Ruseckas, G

    J. Ruseckas, G. Juzeli¯ unas, P. ¨Ohberg, and M. Fleis- chhauer, Phys. Rev. Lett.95, 010404 (2005)

    2005

  6. [6]

    Z. Wu, L. Zhang, W. Sun, X.-T. Xu, B.-Z. Wang, S.-C. Ji, Y. Deng, S. Chen, X.-J. Liu, and J.-W. Pan, Science 354, 83 (2016)

    2016

  7. [7]

    Huang, Z

    L. Huang, Z. Meng, P. Wang, P. Peng, S.-L. Zhang, L. Chen, D. Li, Q. Zhou, and J. Zhang, Nature Physics 12, 540 (2016)

    2016

  8. [8]

    Sugawa, F

    S. Sugawa, F. Salces-Carcoba, A. R. Perry, Y. Yue, and I. B. Spielman, Science360, 1429 (2018)

    2018

  9. [9]

    Goldman, A

    N. Goldman, A. Kubasiak, P. Gaspard, and M. Lewen- stein, Phys. Rev. A79, 023624 (2009)

    2009

  10. [10]

    Goldman, G

    N. Goldman, G. Juzeli¯ unas, P.¨Ohberg, and I. B. Spiel- man, Reports on Progress in Physics77, 126401 (2014)

    2014

  11. [11]

    Liang, Z

    Q. Liang, Z. Dong, J.-S. Pan, H. Wang, H. Li, Z. Yang, W. Yi, and B. Yan, Nature Physics20, 1738 (2024)

    2024

  12. [12]

    Y. Yang, B. Yang, G. Ma, J. Li, S. Zhang, and C. T. Chan, Science383, eadf9621 (2024)

    2024

  13. [13]

    Cheng, K

    D. Cheng, K. Wang, and S. Fan, Phys. Rev. Lett.130, 083601 (2023)

    2023

  14. [14]

    Zhang, F

    X.-L. Zhang, F. Yu, Z.-G. Chen, Z.-N. Tian, Q.-D. Chen, H.-B. Sun, and G. Ma, Nature Photonics16, 390 (2022)

    2022

  15. [15]

    Cheng, K

    D. Cheng, K. Wang, C. Roques-Carmes, E. Lustig, O. Y. Long, H. Wang, and S. Fan, Nature637, 52 (2025)

    2025

  16. [16]

    T. T. Wu and C. N. Yang, Phys. Rev. D12, 3845 (1975)

    1975

  17. [17]

    Aharonov and D

    Y. Aharonov and D. Bohm, Phys. Rev.115, 485 (1959)

    1959

  18. [18]

    Y. S. S. Patil, J. H¨ oller, P. A. Henry, C. Guria, Y. Zhang, L. Jiang, N. Kralj, N. Read, and J. G. E. Harris, Nature 607, 271 (2022)

    2022

  19. [19]

    Y. Wang, Y. Wu, X. Ye, C.-K. Duan, Y. Wang, H. Hu, X. Rong, and J. Du, Nature nanotechnology20, 873 (2025)

    2025

  20. [20]

    Q. Guo, T. Jiang, R.-Y. Zhang, L. Zhang, Z.-Q. Zhang, B. Yang, S. Zhang, and C. T. Chan, Nature594, 195 (2021)

    2021

  21. [21]

    Y. Yang, C. Peng, D. Zhu, H. Buljan, J. D. Joannopoulos, B. Zhen, and M. Soljaˇ ci´ c, Science365, 1021 (2019)

    2019

  22. [22]

    Y. Yang, B. Zhen, J. D. Joannopoulos, and M. Soljaˇ ci´ c, Light: Science & Applications9, 177 (2020)

    2020

  23. [23]

    Q. Yan, Z. Wang, D. Wang, R. Ma, C. Lu, G. Ma, X. Hu, and Q. Gong, Adv. Opt. Photon.15, 907 (2023)

    2023

  24. [24]

    Anandwade, Y

    R. Anandwade, Y. Singhal, S. N. M. Paladugu, E. Martello, M. Castle, S. Agrawal, E. Carlson, C. Battle- McDonald, T. Ozawa, H. M. Price, and B. Gadway, Phys. Rev. A108, 012221 (2023)

    2023

  25. [25]

    Martello, Y

    E. Martello, Y. Singhal, B. Gadway, T. Ozawa, and H. M. Price, Phys. Rev. E107, 064211 (2023)

    2023

  26. [26]

    Singhal, E

    Y. Singhal, E. Martello, S. Agrawal, T. Ozawa, H. Price, and B. Gadway, Phys. Rev. Res.5, L032026 (2023)

    2023

  27. [27]

    M. Tian, I. Velkovsky, T. Chen, F. Sun, Q. He, and B. Gadway, Phys. Rev. Lett.132, 126602 (2024)

    2024

  28. [28]

    Velkovsky, A

    I. Velkovsky, A. Abraham, E. Martello, J. Yu, Y. Sing- hal, A. Gonzalez, D. Lewis, H. Price, T. Ozawa, and B. Gadway, arXiv preprint arXiv:2406.01732 (2024)

    Pith/arXiv arXiv 2024

  29. [29]

    Rhyno, I

    B. Rhyno, I. Velkovsky, P. Adshead, B. Gadway, and S. Vishveshwara, Phys. Rev. Res.7, L022004 (2025)

    2025

  30. [30]

    F. K. Kunst, E. Edvardsson, J. C. Budich, and E. J. Bergholtz, Phys. Rev. Lett.121, 026808 (2018)

    2018

  31. [31]

    Ashida, Z

    Y. Ashida, Z. Gong, and M. Ueda, Advances in Physics 69, 249 (2020)

    2020

  32. [32]

    Zhang, T

    X. Zhang, T. Zhang, M.-H. Lu, and Y.-F. Chen, Ad- vances in Physics: X7, 2109431 (2022)

    2022

  33. [33]

    Z. Pang, B. T. T. Wong, J. Hu, and Y. Yang, Phys. Rev. Lett.132, 043804 (2024)

    2024

  34. [34]

    Salerno and I

    G. Salerno and I. Carusotto, EPL (Europhysics Letters) 106, 24002 (2014)

    2014

  35. [35]

    Salerno, T

    G. Salerno, T. Ozawa, H. M. Price, and I. Carusotto, Phys. Rev. B93, 085105 (2016)

    2016

  36. [36]

    Salerno, A

    G. Salerno, A. Berardo, T. Ozawa, H. M. Price, L. Taxis, N. M. Pugno, and I. Carusotto, New Journal of Physics 19, 055001 (2017). 7

    2017

  37. [37]

    Brandenbourger, X

    M. Brandenbourger, X. Locsin, E. Lerner, and C. Coulais, Nature Communications10, 4608 (2019)

    2019

  38. [38]

    Sirota, R

    L. Sirota, R. Ilan, Y. Shokef, and Y. Lahini, Phys. Rev. Lett.125, 256802 (2020)

    2020

  39. [39]

    W. Wang, X. Wang, and G. Ma, Nature608, 50 (2022)

    2022

  40. [40]

    Veenstra, O

    J. Veenstra, O. Gamayun, X. Guo, A. Sarvi, C. V. Mein- ersen, and C. Coulais, Nature627, 528 (2024)

    2024

  41. [41]

    See Supplementary Material for more details on the the- oretical mapping, experimental system, additional mea- surements, and the theoretical analysis

  42. [42]

    K. G. Wilson, Phys. Rev. D10, 2445 (1974)

    1974

  43. [43]

    I. L. Chuang and M. A. Nielsen, Journal of Modern Op- tics44, 2455 (1997)

    1997

  44. [44]

    Hatano and D

    N. Hatano and D. R. Nelson, Phys. Rev. Lett.77, 570 (1996)

    1996

  45. [45]

    R. Lin, T. Tai, L. Li, and C. H. Lee, Frontiers of Physics 18, 53605 (2023)

    2023

  46. [46]

    Greschner, D

    S. Greschner, D. Huerga, G. Sun, D. Poletti, and L. San- tos, Phys. Rev. B92, 115120 (2015)

    2015

  47. [47]

    G¨ org, K

    F. G¨ org, K. Sandholzer, J. Minguzzi, R. Desbuquois, M. Messer, and T. Esslinger, Nature Physics15, 1161 (2019)

    2019

  48. [48]

    Riechert, J

    H. Riechert, J. C. Halimeh, V. Kasper, L. Bretheau, E. Zohar, P. Hauke, and F. Jendrzejewski, Phys. Rev. B 105, 205141 (2022)

    2022

  49. [49]

    spin down

    I. L. Chuang and M. A. Nielsen, Journal of Modern Op- tics44, 2455 (1997). 8 Supplemental Material for: Non-Abelian Gauge Field Mechanics OVER VIEW In Section , we review the theoretical mapping from our classical experimental systems to the desired Hamil- tonian physics. In Section , we present details of the ex- perimental implementation, beginning with...

    1997