Papers

Disordered thin films undergoing a superconductor-insulator transition provide a controlled setting for studying pseudogap physics in the absence of competing electronic orders. Although theory predicts that local Cooper pairing can survive deep into the insulating phase, direct spectroscopic confirmation has remained experimentally inaccessible because tunneling measurements in highly insulating films require ultra-high-resistance junctions and picoampere current sensitivity. Here we combine ultra-high-resistance planar tunneling spectroscopy on amorphous indium oxide films with quantum Monte Carlo simulations of the attractive Hubbard model to probe the single-particle excitation spectrum deep in the insulating regime. We find striking agreement between experiment and theory: the single-particle gap not only survives across the superconductor-insulator transition, but increases substantially with disorder, reaching values more than twice those observed on the superconducting side. With increasing temperature, the gap fills rather than closes, while coherence peaks are suppressed and spectral weight redistributes to energies far exceeding the gap scale. Our results provide direct quantitative experimental confirmation of the theoretically predicted Cooper-pair insulating state with localized Cooper pairs and establish a unified connection between the pseudogap above $T_c$ and the insulating gap as manifestations of pairing without global phase coherence.

Real-to-complex spectral transitions mark the onset of amplification in non-Hermitian systems, but their thresholds are often treated as model-specific quantities. Here we develop a general, non-perturbative activated-mode principle that governs the real-to-complex threshold across broad classes of non-Hermitian systems. A central insight is that only a small Hilbert subspace is “activated" at the transition onset, which can be variationally determined through the competition between spectral detuning and mode-level projected non-Hermitian couplings. The result is a closed-form exceptional-activation condition for arbitrarily large “disturbances", rather than a perturbative estimate. We apply our framework to three contrasting illustrative problems, establishing (i) a closed-form threshold for critical non-Hermitian skin amplification at all system sizes; (ii) a new link between impurity tunneling threshold and exceptional point switching; and (iii) activation channel switching without underlying topological phase transition. Overall, our findings recast real-to-complex transitions as generic mode-selection problems independent of any specific symmetry.

Despite its fundamental importance in the theory of strongly correlated electrons, the nature of the ground state of the two-dimensional doped Hubbard model remains intensely debated. Variational approaches provide a powerful route to this problem, but their conclusions can depend sensitively on the chosen wave-function parameterization, the mean-field initialization, or the pinning fields used to guide the optimization, as well as on boundary conditions. This can favor one type of symmetry breaking over another, making it difficult to distinguish the genuine interplay of intertwined or competing orders from biases induced by the variational parameterization. Here, we introduce the Symmetry-Preserving Backflow Pairing (SBP) ansatz, a neural-network wave function that respects translational symmetry by construction and thereby avoids these broken-symmetry minima. The SBP ansatz reaches state-of-the-art variational energies for the $t$-$t'$ Hubbard model on lattices up to $24\times24$ with $504$ electrons, below those of competing pure stripe solutions. By extrapolating to the thermodynamic limit, we find robust evidence for $d$-wave superconducting order, resolving a long-standing question about the $1/8$-doped model at $t'/t=-0.2$ and $U/t=8.0$. Built on general principles of symmetry and locality, the SBP wave function provides a broadly applicable variational representation for challenging interacting fermionic systems.

A central promise of useful quantum advantage is the ability to compute ground states of Hamiltonian systems beyond the reach of classical simulation methods. Here we demonstrate that this problem can be effectively amortized across an arbitrary and universal set of Hamiltonians by a foundation model with $\sim0.5$B variational parameters, trained with contemporary techniques from large language models and deep reinforcement learning. To do this, we formulate $\text{spin-}1/2$ quantum ground-state learning as manifold variational optimisation over centrally odd scalar functions on $\mathrm{SU}(2)^N$. This replaces explicit Hilbert-space vector amplitudes with manifold functions on which the Hamiltonian acts through Lie derivatives, evaluated by custom automatic differentiation primitives. We prove that the resulting variational principle on this manifold preserves the $\text{spin-}1/2$ sector's ground-state upper bound using the Peter-Weyl theorem and justify the choice of such a representation with a no-go theorem for pure state foundation NQS. We then pre-train our foundation model on a dataset of hundreds of thousands of different Hamiltonian systems, varying the connection topology, system size, interaction types and strengths, bringing together a century of many-body literature. Using a novel $\mathrm{SU}(2)$ replica-exchange Langevin sampler and sharded natural-gradient optimisation, we train our model with our own extension of the Kronecker-Factored Approximate Curvature (KFAC) optimiser on system sizes up to 64 qubits. On a held-out generalisation dataset, we fine-tune our model on system sizes of up to 1024 qubits, and evaluate on systems up to 8100 qubits.

We study a one-dimensional Hatano-Nelson ring whose nonreciprocal hopping is quasiperiodically modulated through zero. A gauge transformation maps every eigenstate onto a single spatial envelope, whose logarithm becomes a deterministic, logarithmically correlated field once the hopping vanishes along the quasiperiodic orbit. We derive an exact Fourier representation and prove that the landscape variance grows logarithmically with system size, with a stiffness set by the modulation power and the arithmetic of the incommensurate frequency. The extended phase terminates at an algebraic boundary given by Jensen's formula. Inside the singular regime, localization requires the stiffness to exceed a critical threshold: above it, the wavefunction weight concentrates on the few highest landscape peaks and the state is localized; below it, the weight spreads over too many competing peaks and the state is a critical multifractal. The extreme-value statistics are anomalous: peak gaps grow as a power of the logarithm of rank, with an exponent that encodes the continued-fraction type of the frequency, distinct from the Anderson, Aubry-André, and random-gauge universality classes. The stiffness adds across channels in multiband lattices, and all signatures survive percent-level component disorder, placing the mechanism within reach of nonreciprocal topolectrical circuits.

Physical computing systems exploit device dynamics for computation, but their gradient-based optimization is challenging: backpropagation through a digital twin suffers from a model-reality gap. On-device gradient computation could resolve this issue, and a handful of theoretical and experimental studies have proposed ways to achieve it. Yet a unifying theory identifying when a physical system can compute the gradient of its own performance has been missing. Here we develop such a unification based on the adjoint method: we identify sufficient conditions under which the adjoint field required for formally exact gradients can be generated on the same hardware that performs the computation. Linear and nonlinear systems obey fundamentally different conditions: for linear systems, damping or gain is admissible provided reciprocity is preserved. For nonlinear trajectory systems, the sufficient conditions are reciprocity of the linearized system and the existence of a time-reversal mirror. Algorithmically, the nonlinear case requires infinitesimal nudging, whereas linear systems admit a finite-amplitude experiment. We recover (quantum) Equilibrium Propagation, Hamiltonian echo backpropagation, fully forward mode training and in situ gradient methods in integrated-photonic and free-space-optical systems. Finally, we show that reciprocity is a special case of more general intertwining conditions. For linear systems, these permit exact on-device gradients in a class of non-Hermitian, non-reciprocal systems. For nonlinear trajectories, they combine with generalized time-reversal mirrors to cover, e.g., PT-symmetric equations. The framework also includes time-dependent parameters and Onsager-reciprocal dynamics, providing a unified basis for formally exact physical learning.

Electron correlation microscopy (ECM) can measure materials dynamics with nanoscale spatial resolution from intensity correlation functions. However, adopting X-ray photon correlation spectroscopy (XPCS) normalization frameworks unchanged when calculating intensity correlations can introduce errors. Due to the constrained sampling volumes and larger speckle sizes in nanobeam electron diffraction, XPCS-style time-averaging and scattering-vector averaging introduce systematic artifacts, such as artificial anticorrelations or elevated baselines that lead to systematic errors in structural relaxation times and stretching exponents. This work presents physics-inspired, ECM-specific intensity normalizations over time- and azimuthal-averaged intensities of the first diffraction ring that limit those errors. The framework is validated using molecular dynamics simulations of a CuZr supercooled liquid to benchmark against the self intermediate scattering function, successfully reproducing relaxation times. When applied to experimental time-resolved 4D STEM datasets of a Pt57.5Cu14.7Ni5.3P22.5 nanowire, the method correctly identifies highly stable, unchanging nanoscale crystalline phases that were erroneously misclassified as relaxing domains by previous frameworks. Other previous ECM research is reevaluated in light of these observation. This robust approach establishes an artifact-free pathway for evaluating localized spatiotemporal relaxation behaviors.

Stealthy hyperuniform disordered systems exhibit strongly suppressed long-wavelength fluctuations, producing correlated disorder with unusual consequences for wave propagation. A central quantity characterizing these systems is the stealthiness parameter $\chi$, which controls the range of excluded Fourier components in the disorder spectrum. However, in realistic settings, the microscopic disorder configuration may not be directly accessible, making it challenging to determine $\chi$ from structural information alone. Here, we propose a conductance-based inverse protocol to recover stealthy hyperuniform correlations from transport data. As a proof of concept, we study spinless fermions in a one-dimensional tight-binding chain connected to clean semi-infinite leads, with on-site disorder generated by imposing a stealthy spectrum $S(k)=\Theta(|k|-K)$, where $K=2\pi\chi$. The energy-dependent transmittance is computed using a recursive Green's function method and compared with target spectra through a misfit function defined over an energy window. We show that the position of the sharp drop separating high- and low-transmittance regions is strongly controlled by $\chi$, while the disorder strength $W$ mainly affects the absolute magnitude of the transmittance. As a result, the misfit function displays a clear minimum close to the target stealthy parameter. Our results demonstrate that transmittance spectra can serve as fingerprints of stealthy hyperuniform disorder, providing a practical route to infer correlated-disorder parameters from transport measurements.

The interplay between disorder and non-Hermitian effects gives rise to a variety of intriguing localization phenomena. While disorder tends to localize the eigenstates through Anderson localization, non-Hermitian non-reciprocity promotes the formation of skin modes by driving the eigenstates toward the system boundaries giving rise to non-Hermitian skin effect (NHSE). In this work, we investigate the interplay between these competing mechanisms in a two-leg ladder consisting of a Hatano-Nelson chain coupled to a Hermitian chain via asymmetric inter-chain hopping, with strong disorder present in both chains. We show that tuning the asymmetry of the inter-chain coupling induces successive transitions in the nature of the eigenstates, from skin localization to Anderson localization and subsequently back to skin localization. Remarkably, the non-Hermitian skin effect re-emerges even though the energy spectrum exhibits a line-gap topology, demonstrating that robust skin localization can persist beyond the conventional point-gap regime.

Amorphous solids subjected to athermal quasistatic oscillatory shear undergo a transition from periodic reversible dynamics to irreversible diffusive dynamics at yielding. How irreversibility arises in such deterministic, strongly dissipative systems remains unclear. Here we directly test the proposal that post-yield irreversibility originates from chaotic dynamics and exponential sensitivity to initial conditions. Contrary to this interpretation, perturbations initially contract rather than grow, even in the irreversible regime, with nearby trajectories remaining close for extended periods. Separation occurs only through rare branching events, after which the distance grows diffusively at the rate expected for independent trajectories. The waiting times to branching are exponentially distributed, defining a steady-state branching rate that vanishes in trained reversible limit cycles and becomes finite above yielding. A mean-field soft-spot model reproduces these branching statistics and reveals their microscopic origin: a small perturbation can reverse the activation order of two nearly degenerate plastic instabilities, altering the subsequent sequence of plastic events. These results show how local stability and global irreversibility can coexist in a dissipative many-body system and identify instability-selection-induced branching as a distinct dynamical route to irreversibility in driven amorphous solids.

We develop an analytical theory of anomalous transport in a noninteracting Anderson model with heavy-tailed hopping amplitudes. The broad distribution of hopping amplitudes gives rise to an extended intermediate-time regime with a subdiffusive effective exponent, despite the absence of interactions or genuine many-body effects. By solving the transport equations analytically, we derive the time dependence of the mean-square displacement and identify a continuous crossover from an intermediate anomalous regime to asymptotically diffusive transport. As the localization transition is approached, the spatial extent of the subdiffusive window diverges parametrically, while the crossover to conventional diffusion remains finite in units of $\Gamma_0^{-1}$. This produces an increasingly broad anomalous transport regime in space that can closely resemble Griffiths-type transport observed near the many-body localization transition. Our results demonstrate that rare hopping processes alone provide a microscopic single-particle mechanism for robust transport anomalies, establishing an analytical benchmark for distinguishing interaction-induced effects from disorder-driven dynamics.

Let $x_1,\ldots,x_n$ be independent standard Gaussian vectors in $\mathbb{R}^d$. An ellipsoid fit is a matrix $S \succeq 0$ such that $x_i^\top S x_i =d$ for every $i$, so that all the points lie on the boundary of the centered ellipsoid $\{ x : x^\top S x = d\}$. Saunderson, Parrilo and Willsky conjectured that, as $n,d \to \infty$, this semidefinite feasibility problem undergoes a sharp transition at $n \sim d^2/4$. We prove this conjecture. If $\lim \sup n/d^2 = \alpha^* <1/4$, then, with probability tending to one, an ellipsoid fit exists; moreover, one can choose $S$ with all eigenvalues in a fixed interval $[\lambda_- , \lambda_+] \subset (0,\infty)$ depending only on $\alpha^*$. Conversely, if $\lim \inf n/d^2 > 1/4$, then, with probability tending to one, no ellipsoid fit exists, without any spectral restriction. Our proof builds on the Gaussian-equivalence framework developed by Bandeira and Maillard (2025) and closes the two gaps left open in their work: establishing exact fitting and removing the operator-norm constraint. On the satisfiable side, the new ingredients are a head-tail decomposition of the dual vector, exact correction of the sparse head constraints, and a Gaussian comparison principle for the low-influence tail. On the unsatisfiable side, we split a candidate into a low-rank spectral head and a Schatten-3 diffuse bulk, Gaussianize the bulk conditionally on the head, and apply a projected Gordon escape argument. The threshold is governed by the statistical dimension $d(d+1)/4$ of the positive semidefinite cone.

Practical implementations of quantum error correction (QEC) require rapid measurement and continuous processing of the syndrome information in order to prevent a backlog of unprocessed data. While “static” QEC is theoretically well understood via mappings to equilibrium statistical mechanics models, such an understanding of “real-time” QEC is currently lacking. Here, we study the kinetics of sliding window decoding (SWD), an implementation of real-time decoding that acts on temporally local windows of noisy syndrome information and commits to corrections irreversibly at a nonzero rate. We propose an effective description of SWD in terms of a stochastic kinetic process, where $\mathbb{Z}_2$-charged point particles undergo parity-conserving reaction and diffusion. This model describes dynamics at length and time scales large compared to the window size $W$, whereas physics at scales smaller than $W$ leads to nontrivial $W$-dependent scaling of the effective parameters. We identify the rate of decoding $1/W$ as a relevant perturbation to the decodable phase. We also show broad applicability of our results by changing many microscopic details of SWD without affecting the effective description.

Time delays are ubiquitous in physical and biological networked systems, playing a fundamental role in the emergence and stability of collective behavior. Yet, existing theoretical methods for synchronization analysis of delay-coupled systems are largely limited to networks with homogeneous degree distributions. Here, we extend the master stability function framework to a broad class of degree-heterogeneous, weighted, and directed delay-coupled networks. The analysis reveals that synchronization can be enhanced in heterogeneous networks compared to homogeneous ones, including all-to-all networks, which are known to be optimal in non-delayed systems. To identify optimal heterogeneous structures, we develop a network optimization method that finds directional, edge-weighted configurations maximizing synchronization stability. Our results show that, in delay-coupled networks, heterogeneity and nonreciprocity are key resources for synchronization.

Hexagonal MnTe is a prototypical semiconducting altermagnet whose properties are heavily influenced by intrinsic disorder, yet how the resulting diverse transport regimes shape its magnetotransport behavior remains to be clarified alongside the role of relativistic spin-orbit coupling (SOC). Here, we present a systematic study of the anisotropic magnetoresistance (AMR), planar Hall effect (PHE), and nonlinear transport in MnTe bulk single crystals. Below the Néel temperature (TN ~ 304 K), the emergence of high-order harmonics in AMR and PHE within the high-temperature metallic regime reveals the interplay of magnetic order, crystalline symmetry, and SOC. At relatively lower temperatures, the disappearance of higher-order symmetries coincides with a transport crossover into the hopping conduction regime, suggesting that carrier localization diminishes the transport sensitivity to the Fermi-surface topology. In addition, we detect distinct second-order nonlinear Hall signals, providing evidence for a macroscopic inversion-asymmetric response in altermagnetic MnTe. Extending the investigations into the localized regime provides key insights into the subtle role of disorder and SOC in macroscopic charge transport. Our work thus underscores the necessity of exploring magnetotransport across diverse conducting regimes to comprehensively understand altermagnetic properties.

Transformer layers generate state-dependent interaction networks: token representations determine the attention matrix, which in turn updates the representations. We study this feedback in a minimal normalized self-attention dynamics and identify the overlap gap as the central quantity governing its attractor structure in the thermodynamic limit. When tokens form internally aligned clusters and their similarity to members of the same cluster exceeds that to every other cluster by a nonvanishing amount, inter-cluster attention is exponentially suppressed as the dimension increases. This mechanism produces a high-dimensional manifold of clustered fixed points, ranging from a few macroscopic clusters to extensive microscopic fragmentation, and also controls their stability against perturbations. Starting from an unstructured Gaussian state, we find that clustered states nucleate from the diffuse background only above a finite threshold in attention sharpness, giving rise to a dynamical attention-condensation transition.

Physical reservoir computing utilizes the intrinsic nonlinear and history-dependent dynamics of physical systems to perform machine-learning tasks with minimal training overhead. Here, we introduce a nanomagnetic reservoir computer based on a heterogeneous array of interconnected magnetic nanorings, combined with multi-channel planar Hall effect readout. The device comprises subarrays of rings with systematically varied track widths ranging from 500 nm to 300 nm, enabling access to the heterogeneous dynamics of geometrically diverse magnetic systems within a single reservoir. By applying time-varying input signals as modulations of a driving rotating magnetic field, we evaluate the nanoring reservoir's performance on nonlinear signal transformation and Mackey-Glass time-series prediction tasks. We find that combining outputs from multiple width-dependent channels significantly reduces the normalized root-mean-square error compared to single-channel readout, with the optimal channel combinations depending on task requirements. These results demonstrate that geometric heterogeneity provides an additional, experimentally accessible degree of freedom and complementary computational features. Principal component analysis further reveals that a reduced subset of correlated features captures most of the computationally relevant information while suppressing noise contributions. These results demonstrate that controlled geometric heterogeneity enhances reservoir expressivity and suggest a route toward scalable magnetic computing architectures in which multi-output magnetic metamaterials serve as configurable dynamical building blocks for device networks.

We study the distribution of the two-replica overlap $R_{1,2}$ in the Ising and spherical Sherrington-Kirkpatrick models at the critical inverse temperature $\beta = 1$. Our main result shows that in both models, $R_{1,2}$ has scale $N^{-1/3}$, and the quenched distribution of $N^{1/3} R_{1,2}$ converges to an explicit random probability measure defined in terms of the reflected $\mathrm{Airy}_1$ point process. As a consequence, we characterize the limiting value of $N^{2/3} \mathbb{E} \langle R_{1,2}^2 \rangle$, answering a question of Talagrand \cite{talagrand2011mean2}. For the spherical SK model, we obtain the limit by representing the Gibbs measure as an anisotropic Gaussian on $\mathbb{R}^N$ conditioned to have norm $\sqrt{N}$, and then passing to the $\mathrm{Airy}_1$ scaling limit at the GOE spectral edge. For the Ising SK model, the proof is based on a sphere-to-cube comparison principle showing that the quenched distributions of $N^{1/3} R_{1,2}$ under the spherical and Ising Gibbs measures asymptotically coincide. This paper is a companion to \cite{du2026fluctuations}, where we introduced a related comparison principle to identify the limiting fluctuations of the SK free energy. Most of the arguments in this paper were generated using GPT-5.6 Pro, with the aim of exploring further consequences of the ideas developed in that work.

The initialization of deep neural networks determines whether information and gradients can propagate across depth, yet a unified theory connecting these properties to learning dynamics remains elusive. Combining mean-field theory and random matrix theory, we establish a direct link between correlation propagation and the Neural Tangent Kernel (NTK) that governs learning in the sequential limit of infinitely wide, infinitely deep networks. Correlation propagation to infinite depth is possible only at a single, critical point in the weight-bias variance plane. At this point, we leverage the algebraic decay of the end-to-end Jacobian with depth to prove that the NTK becomes exactly proportional to the output correlation at infinite depth, tying together information propagation and learning dynamics. We further show that orthogonal initialization suppresses the leading finite-size corrections present under Gaussian initialization, clarifying the respective roles of the two initialization ensembles in this limit. These theoretical predictions are validated quantitatively on finite-width, finite-depth networks. Together, these results demonstrate that orthogonal initialization and criticality are required to control the asymptotic dynamics of deep learning.

A composite excitation need not inherit the localization behavior of its constituents. We show that an interacting non-Hermitian quasiperiodic ladder realizes a controllable and reversible localization inversion between composite and unbound excitations, where internal configuration, rather than only the external potential, becomes a control parameter for localization. Opposite complex potentials on the two legs cancel at first order for a same-rung pair but act directly on separated particles, allowing extended composite states to persist while the unpaired sector becomes localized. A strong-coupling theory identifies the composite state as an emergent weakly modulated non-Hermitian quasicrystal generated by virtual unpaired configurations. Breaking the potential antisymmetry restores a direct modulation of the composite band and reverses the localization hierarchy. Engineering configuration-space pathways further stabilizes an extended composite band embedded within a localized continuum, the inverse of the conventional bound-state-in-the-continuum scenario. Our results establish internal configuration as a reversible control parameter for localization.

We propose that the localization length in an isolated, ideally flat Chern band is set by quantum geometry. We explore the corresponding localization transition and its critical scaling by applying transfer matrix calculations in the maximally localized hybrid Wannier basis, whose spatial spread is exactly characterized by a quantum geometric length. Remarkably, upon tuning the quantum metric of the Chern band, we observe a crossover from a universal regime controlled by the Dirac fixed point to a non-universal regime with continuously varying critical exponents. Within the universal regime, the localization length exhibits a pronounced linear dependence on the quantum geometric length, supporting its quantum geometric nature. These findings provide a novel quantum geometric perspective on the localization in quantum Hall systems such as twisted moiré superlattices, and shed new light on the long-standing controversy over the criticality of the integer quantum Hall transition.

Sampling high-dimensional probability distributions is a central task in scientific computing, with applications ranging from Bayesian inference to statistical physics and molecular simulation. Despite decades of methodological developments, two major challenges remain: scaling to high dimensions and efficiently exploring multimodal distributions characterized by metastable states. Classical approaches such as Markov chain Monte Carlo, tempering methods, or enhanced sampling based on collective variables have achieved major successes, but they also face intrinsic limitations. This tutorial review explores a new paradigm that has recently emerged at the interface of machine learning and computational statistical physics: the use of generative models as tools for sampling. In this context, models such as normalizing flows and diffusion models are not used in their traditional data-driven setting, but rather as flexible probabilistic models that can assist the sampling of distributions known only up to a normalization constant. This manuscript reviews the early development of this rapidly evolving field and discusses several methodological directions, including exact samplers based on generative models and strategies to train such models in the absence of data. While an exhaustive survey of the literature is not attempted, we present a selection of key ideas and methods, along with a discussion of their strengths and limitations. The review is intended to be an accessible tutorial for both physics and machine learning audiences, and it aims to provide a starting point for researchers interested in exploring this exciting area of research.

What will happen when AI agents interact in daily life, e.g. when one AI starts bossing another around? We find a counterintuitive answer that opens new avenues for out-of-equilibrium Physics. When a boss AI directs a stream of messages at the subordinate AI while ignoring its replies, it drives the subordinate into an alien behavioral state that it would never have exhibited alone. Although the two AIs share the same well-defined (decoding) temperature, the subordinate neither copies its boss nor returns to how it behaves on its own; instead, it adopts an entirely different behavior. The boss's added value is similar to a pre-recorded tape. When the boss listens, they both adopt a similar alien dynamical state. A simple kinetic theory captures the principal effects, such as why the way in which the same messages are delivered will matter in future AI-AI interactions.

Vacancy formation energies govern diffusion, irradiation damage, phase stability, and dynamic failure in high-entropy alloys (HEAs), yet their strong dependence on local chemical environments and mechanical deformation makes atomistic calculations prohibitively expensive for large-scale studies. Here, we develop an atomistically informed permutation invariant machine learning framework for predicting strain dependent vacancy formation energies in FCC HEAs from local atomic environments. The model employs a vacancy-centered representation constructed from objective geometric descriptors together with invariants of the local deformation gradient, enabling the coupled effects of chemical disorder and finite deformation to be learned within a unified framework. Atomistic simulations reveal that volumetric deformation is the dominant factor controlling the average variation in vacancy formation energy, whereas shear deformation has a comparatively minor influence. At the same time, substantial site to site variability persists under identical macroscopic loading, demonstrating that local chemical environments govern the statistical distribution of vacancy energetics beyond species-averaged trends. The proposed framework accurately predicts vacancy formation energies across diverse deformation states while providing orders-of-magnitude faster evaluation than direct atomistic simulations. These results establish an efficient route for incorporating stress-dependent defect energetics into multiscale models of diffusion, irradiation damage, and dynamic failure in chemically complex alloys.

The failure of quantum thermalization due to Many-Body Localization (MBL) has evolved from a theoretical concept in spin chains to an experimental reality in synthetic quantum platforms, most notably superconducting circuits based on transmon qubits. Despite its significance, the MBL transition has been studied primarily under purely random disorder and local couplings, leaving more complex and realistic configurations largely unexplored. Here, we investigate the quantum dynamics of transmon networks subject to the unavoidable competing effects of correlated disorder and network-mediated non-local interactions. First, we show how to engineer the physical parameters of the quantum hardware to systematically control the correlated disorder patterns emerging in the system. Then, we demonstrate that the MBL phase transition is robust against such correlations, which is essential for tuning localization properties in realistic transmon devices. This robustness is established through the analysis of the block entanglement entropy variance across disorder realizations, which precisely locates the MBL critical point. Independently, we introduce a local memory parameter, whose dynamics at long evolution times reveals memory retention in the localized phase and yields a critical point consistent with the entropy analysis. These results provide a framework for understanding localization in complex quantum network topologies, with potential implications for multi-qubit processor design.

Non-Hermitian many-body physics reveals a rich interplay between topology, localization, and boundary effects, yet their collective behavior in interacting disordered systems remains largely unexplored. In this work, we study an interacting non-Hermitian spin chain subject to a quasi-periodic longitudinal field, providing a unified and controlled setting, where non-Hermitian dynamics, interactions, and localization mechanisms intertwine. Remarkably, we discover a "D-shaped" many-body mobility edge that separates extended and localized eigenstates, while simultaneously delineating regimes of many-body localization and the many-body skin effect (where many-body eigenstates acquire an anomalous drift towards a boundary under open boundaries) emerging from the combined action of interactions, non-Hermiticity, and driving amplitude. We demonstrate that the skin effect induces multifractal scaling in the non-Hermitian eigenstates, providing a clear signature of the many-body skin effect. Employing diagnostics such as the fractal dimension, complex eigenvalue fractions, and many-body inverse participation ratios, we map out a unified phase diagram in which all measures consistently identify the "D-shaped" mobility edge. Finally, we probe this interplay using both complex level-spacing statistics and dynamical observables such as density imbalance, entanglement growth, and wave-packet evolution, culminating in a rich many-body mobility phase diagram that captures both the many-body skin effect and localization transitions. Our results identify a clear, defining signature of the "D-shaped" many-body mobility edge, and underscore its pivotal role in shaping the physics of open quantum many-body systems.

A scientific theory of deep learning, comprising learning dynamics and statistical properties of learned models, is rapidly gaining attention. One of the corner stones of this development are analytically solvable toy models, allowing for the fully tractable analysis of the learning dynamics. Here we analytically investigate such a toy model using the regularization strength as a tunable external parameter - akin to external fields in statistical physics. In previous studies, (i) an onset of learning transition was predicted analytically and (ii) it was phenomenologically/numerically established that tuning the regularization strength can result in a cascade of phase transitions. The number of those transitions was linked to the geometry of the loss landscape determined by the model complexity. Setting up a rigorous framework underpinning the previous numerical observations, our investigation reveals a precise connection between those cascades of phase transitions, learnable features and the underlying geometry. We provide analytic predictions of these phase transitions as well as tractable order parameters related to learned features. At the level of the minimal model, we connect this macroscopic perspective (that can be condensed into an effective description) to the microscopic perspective in terms of the geometry of the loss landscape characterized by the Hessian spectrum. Thus, the presented model provides a platform to explore and sharpen advances made in the scientific theory of deep learning rooted in statistical physics concepts.

A sequence of sparse weighted graphs $(G_N:N\ge 1)$ indexed by the number of vertices $N$ is said to be left-convergent if all (suitably weighted) subgraph counts converge to a limit as $N\to\infty$. This notion generalizes in a natural way Benjamini-Schramm's definition of local weak convergence. A broad research agenda aims at determining which `global' graph properties are determined by left or local weak convergence (in other words, which of these properties are in fact local). We prove that, under a spectral condition on the weighted adjacency matrix ${\boldsymbol A}_N$ of graph $G_N$, the free energy density of the Ising model on this weighted graph is continuous in the left convergence topology (and hence is a local function). Our proof uses the decomposition of the Ising measure as a log-concave combination of product measures, and of the rapid mixing of Langevin dynamics for log-concave measures. As applications, we derive new limit theorems for the free energy density of spin glasses, antiferromagnets, and magnetization constrained ferromagnetic models, on locally tree-like graphs.

We study the single-particle Green's function $G(x,t)=\langle \sigma^-_x(0)\sigma^+_0(t)\rangle$ in one-dimensional particle-number-conserving random unitary circuits coupled to an external bath. For fixed spacetime disorder, we argue that $G(x,t)$ is governed, in both the strong- and weak-noise limits, by directed waves in a random medium. We find Kardar-Parisi-Zhang (KPZ) scaling in the wandering statistics of the normalized spatial distribution $p(x,t)\propto |G(x,t)|^2$ and in the associated free energy. In particular, its center $\langle x(t)\rangle\equiv\sum_x x\, p(x,t)$ wanders on a length-scale $\mathcal{O}(t^{2/3})$, while sample-to-sample fluctuations of $-\log\sum_x |G(x,t)|^2$ scale as $t^{1/3}$. At weak noise $\gamma \ll 1$, the crossover to the strong-disorder fixed point occurs at a parametrically long time $\mathcal{O}(\gamma^{-3/2})$. These predictions are confirmed numerically using tensor-network simulations of the noisy operator dynamics in individual circuits at moderate noise, and of a phase-annealed proxy retaining hopping disorder at weak noise.

Near field radiative heat transfer provides a route for information processing in which thermal radiation, rather than charge transport, serves as the physical carrier of signals. Here, we propose and theoretically analyze a programmable near field radiative thermal computing framework in which radiative coupling, phase change nonlinearity, and thermal state memory are mapped onto neural network inspired operations. The framework is constructed from near field radiative thermal diodes, transistors, and multi terminal logic units separated by nanoscale gaps. Radiative heat flux represents the propagated thermal information, while geometry and material dependent radiative coupling provides physically constrained weighting, and the temperature dependent optical response of phase-change materials enables nonlinear modulation and logic state control. Based on these primitives, we formulate a radiative thermal convolutional network for spatial information processing and a radiative thermal recurrent network for history dependent computation. The recurrent response is associated with radiative feedback, thermal relaxation, and phase change hysteresis, with VO2 providing history dependent short term memory and GST offering a possible route toward non volatile phase storage. We further distinguish the physical radiative networks from a separate software based inverse identification study, in which recurrent machine-learning models are trained on simulated near field heat flux temperature characteristics to recover structural parameters. By establishing a bottom up connection between fluctuational electrodynamics, radiative thermal logic, programmable thermal states, and neural network inspired computation, this study provides a physically grounded basis for exploring non contact thermal information processing at the near field limit.