The Junction That Remembers: Infrared Response of Two Coupled SYK Majorana Systems

We study a dynamically opened Majorana junction formed by two coupled Sachdev–Ye–Kitaev (SYK) systems to investigate how opening an interacting fermionic junction changes many-body dynamics, revealing distinct signatures in transfer, entanglement, and infrared response.

The gist: The dynamical opening of a coupled SYK Majorana junction acts as a quench of many-body connectivity that produces three logically distinct channels—information transfer across the junction, entanglement growth measured by the reduced entropy SL, and long-wavelength response and operator scrambling—with infrared memory being amplified by soft frequencies.

How it works

The central physical idea is to regard the time-dependent bilinear coupling as a quench of many-body connectivity: two strongly interacting quantum systems begin almost independently, are brought into contact, and subsequently evolve as a coupled many-body system. This setting allows for the study of coherent transfer, entanglement growth, operator spreading, and infrared response to be studied within one microscopic model. The protocol is described by the Hamiltonian:

H(t) = HL + HR + Hlink(t), where Hlink(t) = ig(t) X L i χ L i χ R i (1).

The shutter protocol is defined by a smooth switch: g(t) = gcl + δg/2 (1 + tanh(t-t0)/τ), where τ is the switching time. This switch controls how rapidly the architecture changes, leading to three dynamical regimes: sudden, intermediate, and slow switching based on the comparison of microscopic scale J−1 and shutter time τ.

Key Diagnostics and Signatures

The protocol generates three logically different channels that must be distinguished:

1. Transfer: A cross-sector correlator provides a natural propagation diagnostic, T(t) = 1/N X N i (45). For the exact finite-N calculation, the left fermion parity is used as the primary transfer diagnostic.

2. Entanglement: The left-right von Neumann entropy, SL(t) = -Tr[ρL(t) ln ρL(t)], where ρL(t) = TrR Ψ(t)⟩⟨Ψ(t) (47), is the second diagnostic. This quantity makes the physical distinction between models especially transparent, as interacting dynamics can redistribute amplitude among many-body configurations, leading to an entropy that substantially exceeds the ln 2 ceiling of the restricted one-particle quadratic reference.

3. Operator spreading and scrambling: The infinite-temperature cross-sector OTOC F(t) = 4dim H Tr[V(t)W V(t)W] (48) is used as a direct numerical diagnostic to evaluate operator spreading and scrambling.

Infrared Response and Memory

The analytical part focuses on an infrared question concerning the driven inter-sector response. For the q=4 SYK saddle, the low-frequency spectral weight behaves as ρ(ω) ∼ ω−1/2 (20). This implies an infrared-enhanced source when the time-dependent link is inserted into the conformal theory. The paper carefully distinguishes this controlled source scaling from an effective Bogoliubov kernel, noting that a previously invoked kernel proportional to (ων)−1/4(ω + ν)−1 yields a conditional 1/µ Hilbert–Schmidt scaling, but this exponent is not established here as a theorem of the full SYK Keldysh problem. The crucial clarification is that the exact interacting SYK evolution is not a Bogoliubov transformation, and the symbol βeff refers only to the anomalous part of a linearized quasiparticle response about the large-N conformal saddle.

Finite-N Benchmarks and Interpretation

Exact finite-dimensional simulations of two N=8 SYK clusters show enhanced inter-sector entanglement, parity transfer, and cross-sector operator sensitivity after the opening. These results demonstrate that the quartic dynamics have converted a simple transfer problem into a genuine many-body redistribution problem. The finite-size physics shows that the opening does not merely transfer the initial excitation; it exposes that excitation to a new interacting environment, allowing for state-space growth and entropy exceeding the quadratic ln 2 ceiling. While exact diagonalization cannot reproduce the continuum limit, it serves as a benchmark demonstrating that the physical channels identified analytically—transfer, entanglement, and cross-sector operator sensitivity—are present in an explicitly interacting finite Hilbert space. The results clarify that infrared memory concerns the persistence of sensitivity to the low-frequency structure of the drive.

Holographic Interpretation

In the nearly-AdS2 description, time-dependent couplings are interpreted as a "time-dependent boundary deformation coupled to the low-energy reparametrization dynamics.

Here are specific improvements to AI systems derived from the concepts presented in this scientific paper, focusing on leveraging its findings in areas where current models fall short:

)1. Enhanced Quantum State Characterization and Simulation (Leveraging Finite-N Results)

The paper's exact finite-N simulations (Section 8) demonstrate that opening an interacting junction leads to many-body operator growth and entanglement that exceeds the quadratic reference limit.

The interacting SYK dynamics can exceed this value because the excitation is dressed by the many-body degrees of freedom of both sectors.

This suggests AI systems should move beyond simple single-particle or Gaussian approximations when modeling complex, highly correlated quantum states.

Improving AI Systems:

1. Implement a Many-Body Dressing Module in Quantum Simulation (e.g., Tensor Networks or Qubit Architectures). This module would dynamically adjust the effective Hilbert space representation when connectivity is opened (i.e., when a coupling term like Eq. 1 becomes non-zero).

2. Improve Entanglement Metrics: Develop AI algorithms to efficiently calculate and characterize the resulting entanglement (like the left-right von Neumann entropy, SL(t)) in real-time during dynamic processes, rather than relying on post-hoc analysis of time evolution.

Improved AI System Capability:

The system can accurately predict how a simple local perturbation (the shutter opening) propagates through a strongly interacting environment, quantifying not just the transfer probability but the resulting growth in operator complexity and many-body correlations. This is crucial for designing robust quantum error correction codes or simulating complex materials where local excitations are coupled to global dynamics.

)2. Infrared Response Modeling (Leveraging Conformal Field Theory Insights)

The paper establishes that time-dependent coupling introduces an infrared-singular source for soft frequencies, and this enhancement depends critically on the projection method (the kernel).

The Fourier transform of τ −1/2 scales as ω−1/2, not ω−1/4.

This means standard continuum approximations might fail to capture the true low-energy behavior of driven systems.

Improving AI Systems:

1. Develop a Frequency-Dependent Response Predictor for non-equilibrium dynamics. Instead of using a fixed spectral density approximation, the AI should dynamically incorporate the dependence on the switching time scale (τJ) and temperature (T).

2. Implement Adaptive Regularization: Use machine learning to determine the optimal regulator (cutoff µ or T) based on which frequency sector is being probed, rather than applying a universal one.

Improved AI System Capability:

The system can perform Infrared Memory analysis in driven quantum processors or physical systems. It can distinguish between genuine, history-dependent low-frequency responses (memory) and simple thermal noise by analyzing the scaling of correlations as the probe frequency approaches the system's intrinsic low-energy cutoff (T or finite size).

)3. Protocol Analysis for Quantum Thermodynamics and Work Extraction (Leveraging Energy Balance)

The paper provides an exact energy balance relation derived from the Hamiltonian evolution:

The total work performed over the protocol is therefore obtained by integrating the response of the inter-sector bilinear to the prescribed drive. (Eq. 7)

This separates energy injection from information transfer, which is vital for quantum thermodynamics.

Improving AI Systems:

1. Integrate a Work/Energy Tracking Subsystem into control algorithms for quantum devices (e.g., superconducting circuits or trapped ions). This subsystem would monitor the work done by external driving fields in real-time and correlate it with entanglement growth and operator spreading metrics.

2. Develop a Thermodynamic Signature Analyzer. This AI would analyze time-series data from experimental protocols to determine if the observed dynamics are dominated by nonadiabatic energy injection, coherent transfer, or genuine scrambling.

Improved AI System Capability:

The system can optimize quantum control protocols not just for reaching a target state, but for achieving specific thermodynamic outcomes—for instance, maximizing work extracted while minimizing unwanted entanglement generation during a switching event.

)4. Holographic and Geometric Interpretation (Leveraging Nearly-AdS2 Mapping)

The paper maps the coupled SYK shutter to time-dependent boundary deformations in nearly-AdS2 gravity, suggesting that infrared memory is about the geometry retaining information about the boundary deformation history.

The phrase “infrared memory” can then be viewed as a question about whether the late-time soft sector retains information about the time profile of the boundary deformation.

Improving AI Systems:

1. Create a Geometric Dynamics Predictor for effective field theories derived from quantum systems. This AI would learn to map microscopic coupling functions (like g(t)) onto effective geometric actions (like Eq. 62, involving the Schwarzian derivative) and predict the resulting low-energy mode evolution based on boundary conditions.

2. Design simulation architectures that mirror this geometry, using Time-Dependent Boundary Condition Solvers to study how external driving functions perturb the infrared structure of a coupled system.

Improved AI System Capability:

The system can provide a geometric interpretation for chaotic quantum dynamics, allowing researchers to predict the long-term soft-mode behavior (infrared response) of complex systems by modeling them as evolving gravitational boundaries.