A new family of $D=4$ $\mathcal{N}=8$ gauged supergravities is introduced, consisting in a mixture of Scherk-Schwarz and dyonic CSO gaugings that involves the trombone scaling symmetry. A specific theory in this class is shown to admit supersymmetric anti-de Sitter vacua, argued to be related to various wrapped M5-brane configurations. The mass spectrum of these vacua within our maximal theory is computed, extending previous results in smaller gauged supergravity models, and shown to exhibit some exotic features due to the trombone gauging. These vacua are obtained as solutions of certain subsectors of our $\mathcal{N}=8$ supergravity. Such subsectors in turn arise by consistent truncation of the parent maximal supergravity to the invariant fields under specific groups that are not contained in the $\mathcal{N}=8$ gauge group. Prompted by these and other similar subtruncations recently considered, we formalise sufficient conditions that allow any maximal supergravity to be truncated consistently to invariant subsectors thereof when the invariance group is not necessarily contained in the $\mathcal{N}=8$ gauge group.

Molybdenum disulfide (MoS$_2$), a layered van der Waals material, has attracted considerable attention as a promising alternative to graphene for applications in field-effect transistors and nanophotonic devices because of its sizable band gap, high carrier mobility, large on/off ratio, and strong photoluminescence efficiency. A particularly intriguing property of MoS$_2$ is the transition of its band gap character with layer thickness: while the monolayer exhibits a direct gap, the band gap becomes indirect in multilayer and bulk forms.In this study, we clarify the microscopic mechanism underlying this transition. Focusing on the roles of atomic orbitals and interlayer interactions, we perform an analysis combining first-principles calculations with a Wannier-based model. Although interlayer $p_z$--$p_z$ coupling between neighboring sulfur atoms has been recognized as a key factor in this transition, we find that a complete quantitative description additionally requires interlayer $p_z$--$p_x$ and $p_z$--$p_y$ couplings between neighboring sulfur atoms. These findings highlight the importance of both out-of-plane and in-plane orbital contributions in governing the electronic structure of layered MoS$_2$, providing deeper insight into its band gap engineering for future device applications.

Run-time integrity enforcement in real-time systems presents a fundamental conflict with availability. Existing approaches in real-time systems primarily focus on minimizing the execution-time overhead of monitoring. After a violation is detected, prior works face a trade-off: (1) prioritize availability and allow a compromised system to continue to ensure applications meet their deadlines, or (2) prioritize security by generating a fault to abort all execution. In this work, we propose PAIR, an approach that offers a middle ground between the stark extremes of this trade-off. PAIR monitors real-time tasks for run-time integrity violations and maintains an Availability Region (AR) of all tasks that are safe to continue. When a task causes a violation, PAIR triggers a non-maskable interrupt to kill the task and continue executing a non-violating task within AR. Thus, PAIR ensures only violating tasks are prevented from execution, while granting availability to remaining tasks. With its hardware approach, PAIR does not cause any run-time overhead to the executing tasks, integrates with real-time operating systems (RTOSs), and is affordable to low-end microcontroller units (MCUs) by incurring +2.3% overhead in memory and hardware usage.

We investigate a non-Hermitian quantum battery based on the Su-Schrieffer-Heeger (SSH) lattice, charged through a parity-time (PT)-symmetric protocol that alternates gain and loss between the two sublattices. The interplay between lattice topology and non-Hermiticity gives rise to both bulk and edge exceptional points (EPs), which govern the charging dynamics. In the topological regime, an edge-state EP appears at a smaller gain-loss strength than the bulk thresholds and gives rise to an additional edge-broken regime absent in the trivial configuration. This topology-specific spectral structure is reflected in the charging dynamics, where the topological phase exhibits more favorable transient and long-time performance in the representative non-Hermitian regimes considered here. We further examine the corresponding Lindblad dynamics, identifying the non-Hermitian model as the conditional no-jump description of the same gain-loss processes. The Lindblad results show that the topological advantage remains visible at the level of stored energy, extractable work, and extractable fraction under unconditional open-system evolution. These findings demonstrate that topology constitutes a genuine physical resource for enhancing the performance of quantum batteries.

The classical Banach spaces $L_\infty[0,1]$ and $\ell_\infty$ are isomorphic. We present here some lower and upper bounds for their Banach-Mazur distance.

Limit-cycle oscillators are the basic building blocks for synchronization; yet, the notion of a quantum limit cycle has remained unclear. Here, we study quantum limit cycles and synchronization in the presence of continuous heterodyne measurement. The resulting quantum trajectories, i.e., time evolutions of the quantum state conditioned on the measurement outcome, make the quantum limit cycles apparent. We focus on the paradigmatic model of the quantum van der Pol oscillator and on two-level systems. Our work provides insights into limit cycles in quantum systems, emphasizing their similarity to classical limit cycles subject to noise. Additionally, we connect theoretical measures of quantum synchronization to quantities experimentally accessible via heterodyne detection.

Accurate modeling of gravitational waveforms from compact binary coalescences remains central to gravitational-wave (GW) astronomy. Post-Newtonian (PN) approximations capture the early inspiral dynamics analytically but break down near merger, while numerical relativity (NR) provides the accurate yet computationally expensive waveforms over limited parameter ranges. We develop a physics-informed neural network (PINN) framework that learns corrections mapping PN dynamics and waveforms to their NR counterparts. As a demonstration of the approach, we use the TaylorT4 PN model as the baseline, and train the network on a remarkably small dataset of only eight hybridized NR surrogate waveforms (NRHybSur3dq8) to learn higher-order corrections to the orbital dynamics and waveform modes for nonspinning noneccentric systems. Physically motivated loss terms enforce known limits and symmetries, such as vanishing corrections in the Newtonian limit and suppression of odd-$m$ modes in equal-mass systems, promoting consistent and reliable extrapolation beyond the training region. We simultaneously incorporate corrections that account for the different meaning of mass parameters in PN and NR descriptions. The learned corrections significantly reduce the phase and amplitude error through the inspiral up to about $200M$ before the merger. This approach provides a differentiable and computationally efficient bridge between PN and NR, offering a path toward waveform models that generalize more robustly beyond existing NR datasets.

We present a data-driven control framework for partial differential equations (PDEs). Our approach integrates Time-Integrated Deep Operator Networks (TI-DeepONets) as differentiable PDE surrogate models within the Differentiable Predictive Control (DPC)-a self-supervised learning framework for constrained neural control policies. The TI-DeepONet architecture learns temporal derivatives and couples them with numerical integrators, while the DPC algorithm uses automatic differentiation to compute policy gradients by backpropagating the expectations of the optimal control loss through the learned TI-DeepONet. This approach enables efficient offline optimization of neural policies without the need for online optimization or supervisory controllers. We empirically demonstrate the proposed method across diverse PDE systems, including the heat, the nonlinear Burgers', and the reaction-diffusion equations. The learned policies achieve target tracking, constraint satisfaction, and curvature minimization objectives, while generalizing across distributions of initial conditions and parameters. Moreover, we demonstrate four orders of magnitude acceleration at inference time compared to nonlinear model predictive control benchmarks. These results highlight the promise of operator learning for scalable model-based control of PDEs.

We prove finite-field analogs of Bourgain's projection theorem in higher dimensions. In particular, for a certain range of parameters we improve on an exceptional set estimate by Chen in all dimensions and codimensions.

Standard model-based control design deteriorates when the system dynamics change during operation. To overcome this challenge, online and adaptive methods have been proposed in the literature. In this work, we consider the class of discrete-time linear systems with unknown time-varying parameters. We propose a simple, modular, and computationally tractable approach by combining two classical and well-known building blocks from estimation and control: the least mean square filter and the certainty-equivalent linear quadratic regulator. Despite both building blocks being simple and off-the-shelf, our analysis shows that they can be seamlessly combined to a powerful pipeline with stability guarantees. Namely, finite-gain $\ell^2$-stability of the closed-loop interconnection of the unknown system, the parameter estimator, and the controller is proven, despite the presence of unknown disturbances and time-varying parametric uncertainties. Real-world applicability of the proposed algorithm is showcased by simulations carried out on a nonlinear planar quadrotor.

We find a process satisfying the axioms of hyper-decoherence which produces standard quantum theory from the theory of quantum boxes (higher-order quantum theory with the non-signalling tensor product). This hyper-decoherence map evades the no-go theorem of Lee and Selby by relaxing constraints on signalling to the past and the uniqueness of purifications. We discuss some natural opposing conclusions: that the existence of this map might be evidence of a genuine hyper-decoherence process producing causal quantum theory from its causally-indefinite higher-order theory; or that it serves as an indication that the axioms of hyper-decoherence might need careful re-consideration, especially regarding the subtle albeit central role that purity plays.

For $g\geq 0$, a genus $g$ Kashiwara-Vergne associator, introduced by Alekseev-Kawazumi-Kuno-Naef as a solution to the generalised KV equations in relation to the formality problem of the Goldman-Turaev Lie bialgebra on an oriented surface with a framing, is directly constructed from a genus $g$ analogue of a Drinfeld associator formulated by Gonzalez, which we call a Gonzalez-Drinfeld associator. The proof is based on Massuyeau's work in genus $0$. The framing is determined from the choice of a Gonzalez-Drinfeld associator, and in the case of genus $1$, we show that only particular framings are realised by our construction.

Motivated by cosmological observations, we push the cosmological bootstrap program beyond the de Sitter invariance lamppost by considering correlators that explicitly break scale invariance, thereby exhibiting primordial features. For exchange processes involving heavy fields with time-dependent masses and sound speeds, we demonstrate that locality in the bulk implies a set of integro-differential equations for correlators on the boundary. These scale-breaking boundary equations come with a built-in memory kernel in momentum-kinematic space encapsulating the universe's evolution during inflation. Specialising to heavy fields with sinusoidal masses such as those found in axion monodromy scenarios, we show that a powerful synthesis of microcausality and analyticity allows an analytical solution of these equations at leading order in the amplitude of mass oscillations. Meanwhile, we also unveil non-perturbative information in the integro-differential equations by resumming apparent infrared divergences as parametric resonances. In addition, we provide a first-of-its-kind example of numerical bootstrap that directly maps out the solution space of such boundary equations. Finally, we compute the bispectrum and uncover, in the squeezed limit, a scale-breaking cosmological collider signal, whose amplitude can be exponentially enhanced (with respect to the Boltzmann suppression) due to particle production triggered by high-frequency mass oscillations.

We generalize the Hyperbolic Fracton Model from the $\{5,4\}$ tessellation to generic tessellations, and investigate its core properties: subsystem symmetries, fracton mobility, and holographic correspondence. While the model on the original tessellation has features reminiscent of the flat-space lattice cases, the generalized tessellations exhibit a far richer and more intricate structure. The ground-state degeneracy and subsystem symmetries are generated recursively layer-by-layer, through the inflation rule, but without a simple, uniform pattern. The fracton excitations follow exponential-in-distance and algebraic-in-lattice-size growing patterns when moving outward, and depend sensitively to the tessellation geometry, differing qualitatively from both type-I or type-II fracton model on flat lattices. Despite this increased complexity, the hallmark holographic features -- subregion duality via Rindler reconstruction, the Ryu-Takayanagi formula for mutual information, and effective black hole entropy scaling with horizon area -- remain valid. These results demonstrate that the holographic correspondence in fracton models persists in generic tessellations, and provide a natural platform to explore more intricate subsystem symmetries and fracton physics.

Non-axisymmetric, flow-driven instabilities in the incompressible Hall-MHD model are studied in a differentially rotating cylindrical plasma. It is found that in the Hall-MHD regime, both whistler waves and ion-cyclotron waves can extract energy from the flow shear, resulting in two distinct branches of global instability. The non-axisymmetric whistler modes grow significantly faster than non-axisymmetric, ideal MHD modes. A discussion of the global whistler instability mechanism is presented in the large-ion-skin-depth, `electron-MHD' limit. When the magnetic field is azimuthal, a subset of the whistler modes having zero axial wave number are uncovered to be destabilized by the `co-rotation amplifier' mechanism. It is observed that the effect of the Hall term on the non-axisymmetric modes can be appreciable when $d_i$ is on the order of a few \% of the width of the cylindrical annulus. Distinct global modes emerge in the strong Hall-MHD regime at significantly stronger magnetic fields than those required for unstable global MHD modes, as the Hall effect weakens the stabilizing `field-line bending' by decoupling ion motion from the magnetic field. These global non-axisymmetric modes may play an important role in weakly ionized accretion disks.

In this paper we analyze a generalized "single-trace $T\bar T$" deformation, defined by a TsT transformation, of the fibered $I$-brane solution from \cite{Nunez2023}. We use the Penrose limit to understand it, and we consider both the TsT followed by the Penrose limit, as well as the Penrose limit followed by TsT. We describe the spin chains obtained in field theory. In the first case we find that, indeed, the TsT transformation preserves solvability in a simple way, as in the standard $T\bar T$ case. In the second case, we have several options, but none is simple enough to be conclusive, however, one case gives us an asymptotically free and IR nontrivial field theory sector, and another a new parallelizable pp wave.

Empirical researchers often use slope-homogeneity tests to assess whether slopes can be treated as common across units. A key difficulty is that heterogeneity may be concentrated in a small number of units, so that a failure to reject homogeneity may reflect limited power rather than true homogeneity. We quantify this issue by analyzing the power of standard slope-homogeneity tests under doubly local alternatives - alternatives in which only small groups of units depart from the common slope and the magnitude of the deviations shrinks with sample size. We characterize detectability as a function of panel dimensions, the size of the departing groups, and the rate at which deviations shrink. The results tell the researcher clearly when homogeneity tests are informative and when they will miss small-group heterogeneity. A Monte Carlo study confirms the theory.

We derive several new quantum bit thread prescriptions for holographic entanglement entropy, equivalent for static states to the quantum extremal surface formula. Our new prescriptions come in many varieties: vector field-based or based on measures over bulk curves, dependent or independent of the bulk UV regulator, loose and strict versions of constraints, and more. We also explore how bit threads behave in the presence of entanglement islands and baby universes. Finally, our prescriptions inspire new measures of entanglement that we call entanglement distribution functions, which can be packaged into a convex polytope that we call the entropohedron.

Gravitational-wave (GW) events are generally believed to originate in galaxies and can thus serve, like galaxies, as tracers of the universe's large-scale structure. In GW observations, waveform analysis provides direct measurements of luminosity distances; however, without relying on a specific cosmological model, the redshifts of GW sources cannot be determined due to the mass-redshift degeneracy. By cross-correlating GW events with galaxies, one can establish a correspondence between luminosity distance and redshift shells, enabling cosmological inference. In this work, we explore the scientific potential of cross-correlating GW sources detected by third-generation (3G) ground-based GW detectors with the photometric redshift survey of the China Space Station Survey Telescope (CSST). We find that the constraint precisions of the Hubble constant and the matter density parameter can reach $1.04\%$ and $2.04\%$, respectively. Additionally, we have also constrained the precision of the GW clustering bias parameter. These results highlight the significant potential of the synergy between CSST and 3G ground-based GW detectors in constraining cosmological models and probing GW source formation channels using cross-correlation of dark sirens and galaxies.

Bayesian optimization (BO) is a powerful framework for estimating parameters of expensive simulation models, particularly in settings where the likelihood is intractable and evaluations are costly. In stochastic models every simulation is run with a specific parameter set and an implicit or explicit random seed, where each parameter set and random seed combination generates an individual realization, or trajectory, sampled from an underlying random process. Existing BO approaches typically rely on summary statistics over the realizations, such as means, medians, or quantiles, potentially limiting their effectiveness when trajectory-level information is desired. We propose a trajectory-oriented BO method that incorporates a Gaussian process surrogate using both input parameters and random seeds as inputs, enabling direct inference at the trajectory level. Using a common random number approach, we define a surrogate-based likelihood over trajectories and introduce an adaptive Thompson Sampling algorithm that refines a fixed-size input grid through likelihood-based filtering and Metropolis-Hastings-based densification. This approach concentrates computation on statistically promising regions of the input space while balancing exploration and exploitation. We apply the method to stochastic epidemic models, a simple compartmental and a more computationally demanding agent-based model, demonstrating improved sampling efficiency and faster identification of data-consistent trajectories relative to parameter-only inference.

We introduce and study the Hesse pencil variety $H_8$, obtained as the Zariski closure in the Grassmannian $G(1,9)$ of the set of pencils generated by a smooth plane cubic and its Hessian. We prove that $H_8$ has dimension $8$ and can be realized as the intersection of $G(1,9)$ with ten hyperplanes corresponding to the Schur module $\mathbb{S}_{(5,1)}\mathbb{C}^3$. Moreover, $H_8$ coincides with the closure of the $SL(3)$-orbit of the pencil $\langle x^3+y^3+z^3,\ xyz\rangle$ and contains eight additional orbits. The variety is singular, and its singular locus is precisely the union of two orbits, $O(\langle x^3,x^2y\rangle)$ and $O(\langle x^2y,x^2z\rangle)$. A key ingredient in our study is a cubic skew-invariant $R\in \bigwedge^3(\mathrm{Sym}^3\mathbb{C}^3)$ defined by $R(l^3,m^3,n^3)=(l\wedge m\wedge n)^3$, whose vanishing characterizes pencils generated by a cubic and its Hessian. This invariant allows us to write explicit equations defining $H_8$. A crucial geometric step in our argument is the fact that through four general points of $\mathbb{P}^2$ there pass exactly six Hesse configurations, which enables us to compute the multidegree of $H_8$ and conclude that it coincides with the variety defined by the invariant $R$.

We introduce the subdimensional entanglement entropy (SEE), defined on subdimensional entanglement subsystems (SESs) embedded in the bulk, as an entanglement-based probe of how geometry and topology jointly shape universal properties of quantum matter. By varying the dimension, geometry, and topology of the SES, we show that the subleading term of SEE exhibits sharply distinct responses in different phases, including cluster states, $\mathbb{Z}_q$ topological orders, and fracton orders. Treating the reduced density matrix of an SES as a many-body mixed state supported on the SES manifold, we further establish a general correspondence between bulk stabilizers and mixed-state symmetries on SESs, separating them into strong and weak classes, and use it to identify strong-to-weak spontaneous symmetry breaking within SESs. Finally, for SESs with nontrivial SEE, we show that weak symmetries act as transparent patch operators of the corresponding strong symmetries. This motivates the notion of transparent composite symmetry, which remains robust under finite-depth quantum circuits that preserve SEE, and implies that each $D$-dimensional SES holographically encodes a $(D+1)$-dimensional topological order. These results establish SEE not only as a sharp probe of geometric-topological response, but also as a route from bulk pure-state entanglement to mixed-state symmetry and holography on subdimensional manifolds.

With the advent of quantum cloud computing, the security of delegated quantum computation has become of utmost importance. While multiple statistically secure blind verification schemes in the prepare-and-send model have been proposed, none of them achieves full quantum fault-tolerance, a prerequisite for useful verification on scalable quantum computers. In this paper, we present the first fault-tolerant blind verification scheme for universal quantum computations able to handle secret-dependent noise on the verifier's quantum device. Composable security of the proposed protocol is proven in the Abstract Cryptography framework. Our main tools are two novel distillation protocols that turn secret-dependent noise into secret-independent noise. The first one is run by the verifier and acts on its noisy gates, while the second and more complex one is run entirely on the prover's device and acts on states provided by the verifier. Both are required to overcome the leakage induced by secret-dependent noise. We use these protocols to prepare states in the X-Y-plane whose noise is overwhelmingly secret-independent, which then allows us to verify with exponential confidence arbitrary fault-tolerant BQP computations.

We investigate the energy efficiency of a library designed for parallel computations with sparse matrices. The library leverages high-performance, energy-efficient Graphics Processing Unit (GPU) accelerators to enable large-scale scientific applications. Our primary development objective was to maximize parallel performance and scalability in solving sparse linear systems whose dimensions far exceed the memory capacity of a single node. To this end, we devised methods that expose a high degree of parallelism while optimizing algorithmic implementations for efficient multi-GPU usage. Previous work has already demonstrated the library's performance efficiency on large-scale systems comprising thousands of NVIDIA GPUs, achieving improvements over state-of-the-art solutions. In this paper, we extend those results by providing energy profiles that address the growing sustainability requirements of modern HPC platforms. We present our methodology and tools for accurate runtime energy measurements of the library's core components and discuss the findings. Our results confirm that optimizing GPU computations and minimizing data movement across memory and computing nodes reduces both time-to-solution and energy consumption. Moreover, we show that the library delivers substantial advantages over comparable software frameworks on standard benchmarks.

Coxeter pointed out that a number of polytopes can be projected orthogonally into two dimensions in such a way that their vertices lie on a number of concentric regular triacontagons (or 30-gons). Among them are the 600-cell and 120-cell in four dimensions and Gosset's polytope 4_21 in eight dimensions. We show how these projections can be modified into Kochen-Specker diagrams from which parity proofs of the Bell-Kochen-Specker theorem are easily extracted. Our construction trivially yields parity proofs of fifteen bases for all three polytopes and also allows many other proofs of the same type to be constructed for two of them. The defining feature of these proofs is that they have a fifteen-fold symmetry about the center of the Kochen-Specker diagram and thus involve both rays and bases that are multiples of fifteen. Any proof of this type can be written as a word made up of an odd number of distinct letters, each representing an orbit of fifteen bases. Knowing the word representing a proof makes it possible to infer all its characteristics without first having to recover its bases. A comparison is made with earlier approaches that have been used to obtain parity proofs in these polytopes, and some directions in which this work can be extended are discussed.

Axion inflation represents an intriguing source of gravitational waves (GWs) from the early Universe. In a companion paper, arXiv:2508.00798, we previously leveraged the gradient expansion formalism (GEF) to investigate pure axion inflation (PAI), i.e., axion inflaton coupled to a pure gauge sector. In this paper, we extend our analysis to fermionic axion inflation (FAI), i.e., we allow for the presence of fermions in the gauge sector. PAI predicts a strongly blue-tilted GW spectrum; in our GEF benchmark study, all parameter regions leading to observable GWs turned out to violate the upper limit on the number of extra relativistic degrees of freedom, $ΔN_{\rm eff}$. As we demonstrate in this paper, the situation is different for FAI: Schwinger pair creation of the charged fermions results in a damping of gauge-field production, which attenuates the GW signal. As a result, the GW signal from FAI can fall into the sensitivity reach of LISA and ET without violating the upper limit on $ΔN_{\rm eff}$. This result notably applies to the arguably most realistic variant of Abelian axion inflation, in which the axion couples to the hypercharge sector of the Standard Model. Besides, we discuss GW emission from the fermion gas, which may further enhance the total GW signal but which also requires a more quantitative investigation in future work. Additionally, we identify a new backreaction regime in which fermion production moderates the axion--vector dynamics. In this regime, the axion velocity and all energy-density components exhibit oscillations analogous to the strong backreaction in PAI, but here, the oscillations occur around the slow-roll trajectory and are damped by the presence of charged fermions. These observations define again an interesting GEF benchmark for future lattice studies.

Ecological interactions can dramatically alter evolutionary outcomes in complex communities. Yet, the framework of population genetics largely neglects interactions from a species-rich community. Here, we bridge this gap by using dynamical mean-field theory to integrate community ecology into classical population genetics models. We show that ecological interactions result in emergent frequency-dependent selection between parents and mutants, characterized by a single parameter measuring the strength of ecological feedbacks. This result generalizes classical population genetics models to highly diverse communities and enables predictions of mutation outcomes in these eco-evolutionary settings. We derive an analytic expression for fixation probability that extends Kimura's formula and reveals that ecological interactions strongly suppress the fixation of moderately beneficial mutations. This suppression arises because frequency-dependent selection leads to prolonged coexistence between parent and mutant lineages, which acts as a barrier to fixation. The strength of these effects increases with effective population size and the number of open niches in the ecosystem. Our study establishes a framework for integrating ecological interactions into population genetics, showing that evolutionary outcomes can be predicted using simple models even in the presence of complex community feedbacks.

We study self-energy effects induced by strong magnetic fluctuations in the paramagnetic phase of strongly-correlated itinerant magnets within the density functional theory combined with the dynamical mean field theory (DFT+DMFT approach) and its non-local extension. We show that both local and non-local magnetic correlations yield a splitting of the electronic spectrum in the paramagnetic phase, such that it closely resembles the DFT band structure in the ordered phase. We demonstrate these effects on $α$-iron, half-metal CrO$_2$, van der Waals material CrTe$_2$, and altermagnet CrSb. Although the obtained split bands do not possess a certain spin projection, their splitting suppresses spectral weight at the Fermi level. Even when originating from local magnetic correlations, the splitting is strongly momentum dependent as a consequence of the orbital selectivity of non-quasiparticle states. The relative importance of non-local vs. local correlations depends on the proximity to half filling of $d$ states: closer to half filling, the role of local correlations increases.

We propose a reduced-order modeling approach for nonlinear, parameter-dependent ordinary differential equations (ODE). Dimensionality reduction is achieved using nonlinear maps represented by autoencoders. The resulting low-dimensional ODE is then solved using standard integration in time schemes, and the high-dimensional solution is reconstructed from the low-dimensional one. We investigate the architecture of neural networks for constructing effective autoencoders that hold necessary properties to reconstruct the input manifold with exact representation capabilities. We study the convergence of the reduced-order model to the high-fidelity one. Numerical experiments show the robustness and accuracy of our approach in different scenarios, highlighting its effectiveness in highly complex and nonlinear settings without sacrificing accuracy. Moreover, we examine how the reduction influences the stability properties of the reconstructed high-dimensional solution.

As a consequence of their spin-orbit entangled ground state, many $5d^{5}$ iridate materials display a peculiar double peak structure in optical transport quantities, such as absorption and conductivity. Their common interpretation is based on the presence of Hubbard subbands in the half-filled $j_{\mathrm{eff}}=1/2$ manifold. Herein, we challenge this picture, proposing a scenario based on the presence of spin-polaron (SP) quasiparticles, and assigning a dominant SP character to the first peak. We illustrate it by taking the materials Ba$_2$IrO$_4$ and Sr$_2$IrO$_4$ as paradigmatic examples, which we investigate within the dynamical mean-field theory and the self-consistent Born approximation. Both theories reproduce nontrivial features revealed by angle-resolved photoemission spectroscopy and optical transport measurements, supporting our interpretation. In the case of Sr$_2$IrO$_4$, we show how the SP scenario survives in the low-doped regime. Similar optical transport fingerprints are expected to be found in the wider class of $5d^5$ iridates and more generally in strongly correlated antiferromagnetic regimes, such as those found in cuprates.