Peer Revenue Now accepting positions

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Friday 6 November

Analysis of PDEs

Nonlinear stability of planar contact waves for the 3-D compressible Navier–Stokes equations in the whole space

We prove the global nonlinear stability of small planar contact waves for the three-dimensional full compressible Navier–Stokes equations in $\R^3$ under general $H^4(\R^3)$ perturbations, without any $L^1$ integrability or zero-mass assumptions. Unlike the one-dimensional problem or the periodic channel, the problem in $\R^3$ admits neither a Poincaré inequality in the transverse directions nor a square-integrable anti-derivative. Our proof relies on three main ingredients: a perturbation around the one-dimensional Navier–Stokes solution starting from the smooth contact profile, a formulation of the low-order perturbation system in the velocity, pressure, and entropy variables with separate estimates for the velocity–pressure pair and the entropy, and the construction of characteristic and space-time Gaussian weight functions. Combining these weighted estimates with higher-order derivative bounds and homogeneous Gagliardo–Nirenberg interpolation yields global stability and algebraic $L^\infty$ decay rates in $\R^3$.
Read the full abstract

All authors Feimin Huang, Rui Li, Xiaochun Wu, Lingda Xu

Statistical Methodology

Scalable Estimation and Inference for Varying Coefficient Expectile Regression with Streaming Data

Nonparametric expectile regression methods face new challenges from streaming data, where retaining the full dataset for batch processing becomes increasingly impractical. In this paper, we develop a scalable nonparametric expectile regression framework for varying coefficient models. We estimate the coefficient functions online via functional stochastic gradient descent with Polyak–Ruppert averaging in a product reproducing kernel Hilbert space, requiring only gradient information and one pass over the data. For this online estimator, we establish non-asymptotic error bounds, prove its optimal nonparametric convergence rate under the proposed step-size schemes, and derive its Berry–Esseen bound for pointwise asymptotic normality. To perform online inference for the varying coefficient expectile regression models, we propose a multiplier bootstrap procedure that perturbs the stochastic gradients with independent random weights and recursively constructs bootstrap trajectories. We establish the bootstrap consistency of the online inference procedure, yielding asymptotically valid pointwise confidence intervals. Simulation studies and a real data application confirm the performance of the proposed methods.
Read the full abstract

High Energy Physics (Phenomenology)

Thermal Field Theory for General Equilibrium Density Matrices

Thermal field theory (TFT) is the combination of statistical mechanics and relativistic quantum field theory. This (book) chapter provides a pedagogical introduction to (and a review of) TFT for general equilibrium density matrices, which allow arbitrary values not only of temperature and chemical potentials, but also of average angular momentum. A general technique to compute ensemble averages is illustrated. Path-integral methods are developed to study the partition function and the thermal Green's functions of general operators, with an arbitrary number of points, in all field theories of physical relevance.
Read the full abstract

Analysis of PDEs

Well-posedness of weak solutions for the time-dependent thermal-fluid-solid multiphysical coupled model

This paper constructs a time-dependent thermal-fluid-solid multiphysical coupled model describing the interaction between a free fluid and a thermo-poroelastic medium and provides a rigorous mathematical analysis. The model consists of the incompressible Navier-Stokes equations and the heat equation in the free-fluid region, together with the thermo-poroelasticity equations in the porous medium. The two regions are coupled on a fixed interface through conditions for normal velocity, tangential velocity, stress balance, temperature continuity, and heat flux continuity. Under a small-data assumption, we prove the existence and uniqueness of a weak solution for this fully dynamic nonlinear coupled system. The proof employs the Galerkin method to construct semi-discrete approximations, establishes uniform a priori estimates that enable the extension of local solutions to global ones, and invokes compactness arguments to pass to the limit, yielding a solution to the original coupled system.
Read the full abstract

All authors Yi Qin, Siyu Yang, Luling Cao, Jian Li

Optics

A hair-thin implant opens the entire depth of the mouse brain to revisiting the same structures at subcellular resolution

Ultrathin multimode-fibre endoscopes provide minimally invasive access to subcellular structures throughout the depth of the mouse brain. Here we introduce a chronically implantable interface with a diameter of 144 {\mu}m, about the thickness of a human hair, that extends this capability to longitudinal imaging, enabling repeated access to the same fine structures over extended periods. The system supports weeks-long tracking of individual dendritic spines, allowing quantitative comparison of their dynamics and survival across dendritic compartments. We further demonstrate longitudinal intracellular calcium imaging during spontaneous and stimulus-evoked activity in the same cells across days, as well as repeated measurements of blood flow in the same vessels across multiple brain regions. This chronic endomicroscopic platform provides stable, minimally invasive optical access for longitudinal subcellular-resolution imaging of brain structure, activity and blood flow across depth.
Read the full abstract

All authors Miroslav Stibůrek, Tomáš Pikálek, Petra Ondráčková, Kristina Herzánová, Adrian Lorenz, André Gomes, Julia Henschke, Endre Levente Marosi, Bhargavi Murthy, Petra Kolbábková, Hongbo Jia, Sanja Bauer Mikulovic, Janelle Pakan, Alessio Attardo, Hana Uhlířová, Tomáš Čižmár

General Topology

Rigid continua containing an arbitrary Peano continuum

Bani\v c and Kac, and later \v Crepnjak and Kac, constructed an uncountable family of continua with different types of rigidity. For every continuum $X$ with the mentioned property, the inverse limit $\varprojlim\{X,f_n\}$ is homeomorphic to $X$ for an arbitrary sequence of continuous surjective functions $f_n: X \to X$. We extend these results by proving that for any Peano continuum $X$ there exists an uncountable family ${\cal F}$ of rigid, for each positive integer $m$, $\frac{1}{m}$-rigid, and $0$-rigid continua such that each $Y \in {\cal F}$ contains a homeomorphic copy of $X$ and for each $Y \in {\cal F}$ it holds that the inverse limit $\varprojlim\{Y,f_n\}$ is homeomorphic to $Y$ for arbitrarily sequence of continuous surjective functions $f_n: Y \to Y$.
Read the full abstract

Populations and Evolution

Resolving non-neutrality across scales in a hyperdiverse tropical forest

Different ecological processes operate at different spatial and temporal scales. Macroscopic regularities can therefore be captured by descriptions that ignore microscopic details, even when those details are mechanistically relevant to ecological processes. Here, we develop a framework inspired by statistical mechanics to track interspecific spatial correlations under spatial coarse-graining. To do so we combine random-matrix filtering with a spatially explicit neutral model. Applying this approach to eight censuses of the Barro Colorado Island tropical forest, we observe a continuous crossover from an interacting to an effectively neutral organization. Below a few tens of metres (20–35~m), the empirical correlation spectrum differs from neutral expectations, revealing scale-dependent species-specific structure. Part of this structure can be attributed to environmental heterogeneity. However, residual spatial associations provide candidate signatures of biotic interactions or other unmeasured ecological, non-neutral processes. With increasing spatial coarse-graining, empirical and neutral correlation spectra progressively converge, and above the crossover scale the neutral model provides an accurate description of the measured collective organization. Neutrality thus emerges as an effective description of the collective organization of the forest above a characteristic scale, without implying that species are equivalent at the individual level. Our framework provides an operational way to locate that scale.
Read the full abstract

All authors Pietro Zgaga, Pablo Villegas, Tancredi Caruso, Tommaso Gili

High Energy Physics (Phenomenology)

An interrelation of three neutrino mixing angles

We conjecture a novel interrelation of three neutrino mixing angles: $\theta^{}_{23} = \left(\theta^{}_{12} + \theta^{}_{13}\right)/2 + \pi/8$, which predicts $\theta^{}_{23} = 43.46^\circ \pm 0.26^\circ$ with inputs of the latest JUNO and Daya Bay precision data. A simple ansatz is proposed to derive this phenomenologically favored and experimentally testable angle relation, and a new expansion of the lepton flavor mixing matrix in powers of a single small parameter is presented as a by-product.
Read the full abstract

Probability

Sharp universality thresholds for random profinite groups

We determine sharp universality thresholds for quotients of the free profinite group $\widehat F_n$ on $n$ generators by $n+u$ independent random relations, where $u$ is a fixed nonnegative integer. For products of mutually independent uniformly coset-balanced factors, convergence to the Liu–Wood measure $\mu_u$ holds when the minimum number of factors grows faster than $\log n$, which is a uniformly sharp condition. For random words of fixed length $\ell_n$, convergence holds exactly when $\ell_n$ is eventually odd and $\ell_n-\log n\to\infty$. We also prove universality for randomly ordered products of independent integer powers under the balance condition $n\alpha_{n,p}/\log n\to\infty$ at each fixed prime, recovering finite-prime cokernel laws by abelianization.
Read the full abstract

Artificial Intelligence

LEMMO: A Large Electromagnetic Model leveraging Pretrained LLMs for Million-Sample Signal Understanding

Large language models have expanded beyond text to vision, audio, and other modalities, but understanding raw electromagnetic (EM) signals remains challenging. Wideband recordings can contain millions of samples, whereas existing waveform-based EM models are primarily evaluated on short in-phase and quadrature (IQ) segments, leaving long-recording understanding and detailed description insufficiently explored. We introduce LEMMO, a large EM model for instruction-driven recognition and description of long raw IQ recordings. LEMMO combines a linear-attention EM encoder with anti-aliased temporal resampling to process up to eight million complex samples. Its EM-text interaction module conditions learnable queries on instructions before retrieving signal evidence. Compact queries support recognition, while structured queries with auxiliary factual supervision organize scene-level properties, signal attributes, and cross-signal relationships for detailed description. We also construct and release a unified 1.38-TB corpus of real-world and synthetic EM signals, including LEMMO-ShortQA and LEMMO-LongDesc. Compared with MERLIN-LoRA, LEMMO improves mean EM-Bench perception accuracy by 15.77 %, mean strategy-generation ROUGE-L F1 by 4.38 points, and mean LEMMO-ShortQA accuracy by 14.31 %. It also improves categorical and numerical fact F1 on both LEMMO-LongDesc tasks while preserving its language backbone's MMLU performance. Our code and datasets are available at https://github.com/N1rv4n4/LEMMO and https://huggingface.co/datasets/DaftJun/RFLAS, respectively.
Read the full abstract

All authors Mingkai Li, Jiajun Wu, Yihan Mao, Zilin Sang, Taosong Yu, Yuan Ma, Minghe Liu, Qianqian Yang, Zhiguo Shi

Earth and Planetary Astrophysics

Validation of Habitable Zone Limits with a New Radiative-Convective Equilibrium Climate Model

The habitable zone (HZ) boundaries calculated with the Clima one-dimensional radiative-convective equilibrium (RCE) model have served as a standard reference for exoplanet target selection for three decades, but they have never been independently validated with a comparable model. We present ExoColumn, a new one-dimensional RCE model that couples an independently developed atmospheric column physics package to the ExoRT correlated-k radiative transfer core that is also used by the three-dimensional ExoCAM model. We calibrate ExoColumn to present-day conditions, which matches other RCE models to within a few kelvin through the troposphere. Line-by-line benchmarks show that the ExoRT and Clima radiation cores each reproduce the outgoing longwave flux to within a few percent, with the differences due to the spectral line databases used by each model. We follow the same inverse climate modeling method as previous studies have done to find inner edge moist- and runaway-greenhouse limits of 0.970 au and 0.966 au, and an outer-edge maximum-greenhouse limit of 1.611 au at a CO$_2$ partial pressure of 8.7 bar. We extend these calculations to 12 host stars spanning 2000-7200 K and to planets of 0.1, 1, and 5 Earth mass, and we provide updated HZ boundary coefficients for community use. The ExoColumn limits systematically fall closer to the host star than those of Clima but agree to within a few percent in orbital distance, which validates the robustness of the published Clima HZ boundaries.
Read the full abstract

Distributed Computing

Nebula: Distributed Position-Independent Caching for LLM Serving

Modern LLM workflows, such as retrieval-augmented generation (RAG) and LLM agents, build long prompts from multiple independent text segments, making prefill a major source of computation. Reusing KV caches is therefore important for reducing serving costs and latency: distributed caching systems spread KV cache across a cluster to increase aggregate capacity, while position-independent caching (PIC) enables reuse of recurring segments under different prefixes. However, existing systems distribute KV caches but still rely on prefix caching (PC), while existing PIC methods enable position-independent reuse without cluster-wide resource management. We present Nebula, the first distributed LLM serving framework to treat PIC segments as first-class citizens, enabling remote PIC reuse and parallel prefill. To track PIC's fine-grained, asynchronous state changes, Nebula combines optimistic accounting with event-driven updates to provide a timely and truthful cluster view for cache management and prefill scheduling. To accommodate diverse reuse patterns across workloads, Nebula leverages cluster-wide ghost-hit feedback to automatically tune the allocation of shared cache capacity between PC and PIC. Finally, Nebula incorporates queueing, computation, and KV transfer costs into a unified prefill completion time model to guide elastic parallelism adjustment and locality-aware compute placement. Across four models and four real-world RAG and agentic workloads, Nebula reduces median TTFT by an average of 61.4% compared with the strongest existing baseline and recovers an average of 76.1% of the tokens missed by prefix caching.
Read the full abstract

All authors Yifei Liu, Junbin Pan, Junhao Hu, Yang Liu, Junyi Chen, Juntong Wu, Chen Chen, Haiyuan Zhao, Xiaoxu Chen, Weihang Chen

Materials Science

Stacking, symmetry and exciton fine structure in boron nitride bilayers

Engineering of two-dimensional material heterostructures offers a broad range of degrees of freedom through the selection of type and number of layers and their relative stacking configurations. In these compound materials, the stacking sequence plays a central role in determining electronic, mechanical, and optical properties. Here, we present a theoretical study of its signature on the excitons of hexagonal boron nitride bilayers based on first-principle calculations within the GW approximation and the Bethe-Salpeter equation. We explain how stacking-dependent symmetry variations lead to distinct electronic structures, particularly near the band edge around the K high-symmetry point, and consequently to markedly different excitonic landscapes. Indeed, we analyze the symmetry-governed character of the low-lying excitonic states, identifying their bright or dark nature for each stacking configuration and performing a systematic comparison. Our results clarify how van der Waals interlayer coupling reshapes excitons in hexagonal boron nitride bilayers and outline prospects for twist-controlled exciton engineering in this material.
Read the full abstract

Artificial Intelligence

ENGRAM: From Agent Experience to Weights through Verified Consolidation

Language-model agents use experience through external retrieval and parameter learning. Consolidating stored experience into model weights requires deciding both which experience to train on and when its external records can leave retrieval. Admission based solely on student mastery can exclude useful teacher demonstrations, while removal immediately after training can discard guidance that the model still needs. We introduce ENGRAM, a framework that separates training admission from memory release. A demonstration-aware gate admits skills based on student mastery or sufficient successful teacher demonstrations. The student learns from selected action traces with context from an Engram Trace Graph (ETG). A verification gate then compares the updated model without episodic retrieval against the base model with retrieval on reserved probes. Release requires valid training participation and scores meeting both relative and absolute criteria; records for skills that fail verification remain retrievable. Experiments across interactive and reasoning benchmarks show consistent gains over frozen-weight ETG controls, with ablations supporting demonstration-aware admission and selective adapter activation. A controlled held-out comparison shows that verified release substantially reduces the retrievable footprint while remaining close to retain-all performance and outperforming immediate and budget-matched random release. ENGRAM also improves task performance while reducing input-token usage and model calls. Skill-level verification further reduces retrievable experience through selective archival, retaining external guidance for skills that do not meet the release criteria. Code is available in anonymous.4open.science/r/engram-anonymous.
Read the full abstract

All authors Ruiyi Yang, Zechen Li, Lihuan Li, Du Yin, Hao Xue, Flora D. Salim

Optics

Interface-driven second harmonic generation in indium tin oxide/gold nanolayers: density and mass discontinuities matter

Second harmonic generation from transparent conductive oxides and metals is traditionally attributed to Coulomb, magnetic Lorentz, and convective sources. We show that this picture is incomplete in deeply subwavelength indium tin oxide/gold bilayers. Measurements reveal two robust spectral features, a broad maximum near 1000 nm and an epsilon-near-zero peak near 1350 nm, that persist across wide variations in material and excitation parameters. Hydrodynamic-Maxwell simulations identify an interface-specific convective mechanism activated by freeelectron density and effective-mass discontinuities at the oxide-metal boundary. As electronic homogenization develops across the bilayer, the response evolves toward one described by conventional Coulomb, Lorentz, and leading convective terms. These results establish second harmonic generation as a sensitive probe of interface-driven nonlinearities and suggest a broader role for electronic discontinuities in shaping nonlinear optical processes at surface and buried material interfaces.
Read the full abstract

All authors J. Arya, D. Ramonell, J. Trull, N. Akozbek, M. Scalora, C. Cojocaru

Machine Learning

Consensus-Constrained Dual-Federated Optimization with Client-Side Dynamic Weighting for Distributed Android Device Threat Defense

Smartphones, smart-home hubs, and wearables are increasingly exposed to sophisticated Android malware, while centralized threat-detection pipelines raise privacy and scalability concerns. Federated learning (FL) enables collaborative model training without collecting raw client data, but partial participation and variation in local model quality can destabilize global optimization. In this article, we propose DW-FedADMM, a federated optimization framework that combines primal-dual coordination with performance-dependent client-side weighting. The dynamic coefficient modulates each participating client's local state before the augmented-Lagrangian update, while the dual variables regulate disagreement between local and global model states. We evaluate DW-FedADMM on three Android malware benchmarks, namely Drebin, Malgenome, and TUANDROMD, with 5, 10, and 15 clients holding IID data shards, partial client participation, and communication budgets of 5 and 10 rounds. Each configuration is repeated with ten random seeds. DW-FedADMM obtains the highest seed-averaged accuracy on Drebin (97.64% +/- 0.20%) and the best mean accuracy in 9 of the 18 dataset-client-round configurations, including 7 of the 9 configurations with a 10-round budget. Pooled over all configurations, it reaches 95.42% accuracy, 94.62% F1-score, 96.06% precision, and 99.45% AUC, compared with 95.32%, 94.51%, 95.56%, and 99.41% for FedADMM, while also giving the lowest mean false-positive rate and the smallest client update norm among the six compared methods. Both primal-dual methods are well ahead of the averaging- and proximal-based baselines on Malgenome and TUANDROMD, where the latter often remain at the majority-class solution when clients are fragmented. The gain of dynamic weighting over FedADMM is most pronounced on Drebin and with the longer communication budget, where it also reduces the variation across seeds.
Read the full abstract

Strongly Correlated Electrons

Coexistence of antiferromagnetism and topological $s+d+id$ superconductivity in twisted cuprate bilayers

We investigate the interplay between antiferromagnetism and topological superconductivity in twisted cuprate bilayers. We demonstrate how topological $s+d+id$ superconductivity is induced by a twist of $\theta \sim 45^\circ$ in a microscopic $t$-$J$ type model of cuprate bilayers, where it robustly coexists with antiferromagnetic order at low hole doping. This novel pairing state becomes physically accessible at realistic interlayer couplings in the overdoped regime. Crucially, it displays a characteristic $2$-fold rotational symmetry in angle-resolved photoemission spectroscopy (ARPES) not observed in the competing time-reversal symmetry breaking $d+id$ or $s+id$ states previously found. A finite Josephson effect at $45^\circ$ twist together with signatures in scanning tunneling spectroscopy of its two Majorana edge modes identifies this topological $s+d+id$ state unambiguously.
Read the full abstract

All authors Javier Castro Luaces, Manuel Fernández López, Jorge Bravo-Abad, Jaime Merino

Robotics

Mechanical Interaction Sensing for Robotic Kiwifruit Handling with an Optoelectronic Soft Suction Cup

Robotic handling of delicate fruit benefits from compliant contact and quantitative sensing of the mechanical interaction at the end-effector. This study investigates an optoelectronic soft suction cup with a kiwifruit-specific bicomponent design for estimating interaction forces during contact with real fruit. During known-contact intervals, the optical signals accurately estimate interaction force on fruits excluded from model training: the best shallow network achieves grouped out-of-fold pooled R-squared = 0.968 +/- 0.005 (MAE = 0.138 N), with a median per-fruit held-out R-squared = 0.973 (MAE = 0.134 N). A simple monotonic mapping from the mean optical response alone achieves R-squared = 0.925 +/- 0.005, showing that substantial force-related information is already present in the optical response. Leave-one-collection-day-out validation remains strong (R-squared = 0.955 +/- 0.031), supporting robustness across acquisition sessions. As an exploratory secondary analysis, binary Soft/Firm classification from compression-test reference labels achieves moderate balanced accuracy (approximately 0.705) under grouped held-out-fruit validation, while finer three-class grading shows limited performance.
Read the full abstract

All authors Romina Adhami, Ebrahim Shahabi, Alessio Mondini, Samuel M. Youssef, Cosimo Della Santina, Gualtiero Volpe, Barbara Mazzolai

Complex Variables

Ahlfors $1$-Regularity and Conformal Criteria for Carleson Measures

We prove that a bounded Jordan domain $\Omega$ has Ahlfors $1$-regular boundary if and only if the following criterion holds for every positive Borel measure $\mu$ on $\Omega$: $\mu$ is Carleson exactly when $\int_\Omega |\phi'(w)|\,d\mu(w)<\infty$ for every conformal bijection $\phi$ from $\Omega$ onto a bounded domain with rectifiable boundary. The same characterization holds using only arclength measures on simple locally rectifiable curves, with the integral condition replaced by finite length of every such conformal image. These results extend Bishop's theorem on the unit disk. We also study analogous criteria for vanishing Carleson measures.
Read the full abstract

Information Theory

Symmetry Breaking over Multiway Channels

Identical users running the same algorithm must often break their symmetry and agree on distinct roles, as in leader election, naming, and resource allocation. We study anonymous symmetry breaking over a discrete memoryless multiway channel invariant under permutations of its $J$ users: each user observes only a private output and must choose a number in $\{1,\dots,J\}$, and the symmetry is broken if the $J$ numbers are distinct. Using a Rényi change of measure, we derive a finite-blocklength converse on the error probability of every symmetric causal protocol, including adaptive and randomized ones. For achievability, block coding with random names reduces the problem to all-source multicast, for which we derive a Gallager random-coding exponent for every fixed $J$, and a cut-set bound that limits the exponent of the probability that two names coincide or are decoded wrongly. Interactive role controllers instead let the users contend for identifiers directly: we give their exact error probability, an achievable exponent given by a spectral radius when some outputs certify a role, and a floor for finite-state controllers when every output has positive probability, and we introduce generalized musical chairs. On a noisy bandit channel motivated by multiplayer multi-armed bandits, with noiseless collisions and arms of a common mean below one, interactive schemes provably exceed the limit of name decoding, and numerical estimates suggest the same with noisy collisions. For two users, noiseless collisions, one informative arm and a silent arm, the optimal error probability is determined up to a factor of two.
Read the full abstract

Category Theory

Two-step nilpotent varieties as categories of modules over a square ring

We characterise the two-step nilpotent varieties of algebras: the semi-abelian varieties in which all ternary cosmash products vanish. Just as abelian varieties are the categories of modules over a ring, two-step nilpotent varieties are exactly the categories of modules over a square ring in the sense of Baues, Hartl and Pirashvili. We construct the square ring explicitly from the full subcategory on the free algebras on one and two generators, and give complete proofs. As consequences we obtain a description of any two-step nilpotent variety by generators and relations, an intrinsic characterisation in terms of quadratic term operations, and an account of its abelian core and of the lower central series of its objects in terms of the square ring. Along the way, we associate with any commutative ring $R$ a square ring whose modules, the two-step nilpotent $R$-groups, extend Hall's nilpotent $R$-groups of class two from $2$-binomial rings to arbitrary rings of coefficients.
Read the full abstract

Quantum Physics

Dynamics and pseudospectra around quantum exceptional points

Exceptional points (EPs) are a hallmark of non-Hermitian physics, giving rise to extreme sensitivity and enhanced dynamic response. Such unique spectral singularities have been extensively studied in classical photonics, both at theoretical and experimental levels, where the relevant observables are evident. That is not necessarily the case in the quantum regime, where this kind of behavior around EPs, to a large extent, remains elusive. In fact, the interplay between dissipation and quantum fluctuations makes the complete Lindbladian evolution quite different from effective non-Hermitian Hamiltonian descriptions. In this context, we investigate the signatures of the Liouvillian EPs on spectral and dynamic response by considering properly normalized long-time averages and imbalances. Even though our approach is generic, we focus on two paradigmatic models-a lossy Bose-Hubbard lattice and a dissipative two-spin system, where we investigate the role of quantum jumps and interactions in the Liouvillian spectrum, in terms of complex and structured pseudospectra. Our results demonstrate how non-Hermitian phenomena emerge in realistic quantum lattices, providing a bridge between classical EP physics and experimentally accessible open quantum systems.
Read the full abstract

All authors Georgios E. Pavlou, Ioannis Kiorpelidis, Georgios K. Simopoulos, Manas Kulkarni, Konstantinos G. Makris

Functional Analysis

Step-isometries on Banach spaces

Step-isometries are maps preserving the integer part of pairwise distances on metric spaces. They arise naturally in the study of infinite random geometric graphs, where graph isomorphisms between LARG-type random graphs reduce to step-isometries on the underlying vertex sets. Several fundamental open problems concerning the rigidity, extendability, and continuity of step-isometries were raised in recent literature, notably by Šemrl, and by Balogh, Walters, and Zsák. In the present paper we resolve a number of these long-standing open questions. We establish rigidity results for step-isometries defined on norm-dense subsets of both real and complex Banach spaces. For complex Banach spaces, we prove that every almost-surjective step-isometry on a norm-dense subset admits a unique extension to a surjective real-affine isometry of the whole space. In the real setting, we investigate three key properties $A$, $B$, $C$ formulated by Šemrl, which ask whether certain classes of step-isometries extend to surjective affine isometries. We verify these properties for uniformly non-square Banach spaces, for spaces with a dense set of Fréchet-smooth points, and for strictly convex Banach spaces with separable dual. In particular, we confirm Conjecture 2 due to Balogh, Walters, and Zsák without countability assumptions on the dense domain set. We further show that every bijective step-isometry defined on the entire Banach space is a homeomorphism, answering their Question 3 affirmatively. Specializing to $\ell_1^n$, we obtain that $\ell_1^n$ satisfies Property $A$ for all integers $n\ge 3$, settling an open problem posed by Šemrl.
Read the full abstract

Mesoscale and Nanoscale Physics

Helical magnetism from sliding Luttinger liquids in hole-doped twisted bilayer WTe$_2$

We investigate carrier-mediated helical magnetism in hole-doped twisted bilayer WTe$_2$ coupled to localized moments. The itinerant carriers are described by a valleyless two-flavor sliding Luttinger liquid. Via the feedback mechanism of the helical order formed within individual channels, we show that the interchannel spin susceptibility mediates a Ruderman-Kittel-Kasuya-Yosida interaction between localized moments that can stabilize a coherent helical magnetic texture across the coupled-wire array. A spin-wave analysis of the ordered state allows us to estimate the magnon energies and helix-ordering scale. The magnon energies depend on the carrier interactions with representative values falling in the microwave regime, suggesting possible detection by magnetic-resonance techniques. The helical order partially gaps the carrier spectrum and renormalizes the spin-correlation and relaxation exponents. These results identify twisted bilayer WTe$_2$ as a tunable low-dimensional platform for realizing and probing carrier-mediated helical magnetism.
Read the full abstract

Signal Processing

A Hybrid Digital/Analog Polyphase Filtering Approach for High-Speed RF-DACs

This paper proposes a hybrid polyphase interpolation for RF-DACs that removes the need for full-rate digital commutation by using passive combining on the RF-DAC itself. A P-branch polyphase FIR filter is factored into M streams, each formed by a (P/M)-way digital commutator at fRF/M and driving a dedicated RF-DAC slice, giving either partial digital (1<M<P) or fully analog (M=P) recombination. We show that the slice array acts as a single full-rate DAC with an M-times longer zero-order hold, derive yield laws for slice gain, timing and unit-capacitor mismatch, and find that with retiming the binding requirement is sub-picosecond matching of the fRF clock arrival, so the best choice is the smallest M that closes digital timing. Two 28-nm digital macros, designed for a 12.8GSps output rate and verified by post-layout simulation, illustrate a 4-phase variant that closes timing up to 14.4GSps with its digital logic at fRF/4 and an 8-phase variant that closes timing up to 16GSps with its digital logic at fRF/8. The analog recombination is evaluated with a behavioral model of the slice array. The architectures target wideband software defined radio (SDR) platforms, including cognitive radio, electronic warfare, and satellite communications.
Read the full abstract

All authors Behdad Jamadi, Jungmin Lee, Tarun Cousik, Rohit Rangaraj, Jeffrey S. Walling

Rings and Algebras

Weighted \texorpdfstring{$\mathcal{O}$}-operators on pre-Leibniz algebras and post-pre-Leibniz structures

In this paper, we introduce weighted $\mathcal{O}$-operators (also called weighted relative Rota-Baxter operators) on pre-Leibniz algebras. We construct a differential graded Lie algebra whose Maurer-Cartan elements are exactly these operators. This allows us to define their cohomology as that of the induced pre-Leibniz algebra with coefficients in a suitable representation. As an application, we study formal and finite order deformations of these operators. We show that this theory splits the corresponding theory for weighted $\mathcal{O}$-operators on Leibniz algebras. Finally, we introduce post-pre-Leibniz algebras as the algebraic structures underlying weighted $\mathcal{O}$-operators on pre-Leibniz algebras.
Read the full abstract

Probability

Non-asymptotic First- and Second-Order Tail Brackets for the Kostlan-Shub-Smale Field and Spherical k-Spin Complexity

For the Kostlan-Shub-Smale field $X$ of degree $k\ge3$ on the sphere $\mathbb{S}^{d-1}$, we prove the two-sided bound $(1-\varepsilon(u))\,\delta_{*}(u)\le\mathbb{P}\{\sup X>u\}\le\delta_{*}(u)$ at every finite $(k,d)$, where $\delta_{*}(u)$ is the expected number of critical points above $u$, halved when $k$ is even, and $\varepsilon(u)$ is explicit for $\rho u\ge\min\{d-1,\frac32\sqrt{2(d-1)}\}$. The gap is therefore at most of order $e^{-(1+\rho^{2})u^{2}/2}$ with $\rho^{2}=k/(2(k-1))$. We also give $\delta_{*}$ itself in closed form, as a finite sum of Hermite polynomials, together with a two-term envelope that is asymptotically sharp and a single-monomial relaxation of the form $Cu^{d-2}e^{-u^{2}/2}$. As an application we bracket the annealed complexity of the spherical $k$-spin Hamiltonian at every finite $(k,d)$: the upper bound holds at every energy, a lower bound of the same form at every energy above $9/4$ times the spectral edge $\sqrt{2(d-1)}/\rho$, above twice the edge when $d\ge10$, and is consistent with the Auffinger-Ben Arous-Černý complexity function as $d\to\infty$.
Read the full abstract

Computer Vision

Left-Right-Consistent Iterative Stereo Matching via Cross-Injection

State-of-the-art stereo matching models predominantly predict the disparity of only the left (reference) view. However, applications such as per-eye AR/VR rendering, novel view synthesis, or two-view depth fusion require disparity for both views, and need those two maps to agree. The usual workaround of running the network a second time on the horizontally flipped and swapped pair gives two independent predictions that can be inconsistent, especially on transparent surfaces, in textureless regions, and at occlusion boundaries. We introduce cross-injection (CI), a lightweight extension for iterative (refinement-based) models that can be added to any pretrained left-only iterative matcher. It couples the two passes by injecting the explicit, signed left-right inconsistency as an additional input cue at every step of the refinement stage. We empirically show that CI substantially reduces left-right disagreement on every benchmark (more than halving it on Booster for FoundationStereo), while largely preserving sharp object boundaries. Furthermore, using two state-of-the-art stereo matching models, we show that CI also reduces the single-view disparity error, making it a worthwhile extension even if only left disparity is needed.
Read the full abstract

All authors Manuel Knott, Diogo Luvizon, Daniel Scharstein, Elias Vansteenkiste

Artificial Intelligence

Definitions, Label Names, or Demonstrations? Auditing What Governs LLM Annotation

Large language models (LLMs) are increasingly used to annotate social-science texts, but high accuracy does not establish that they apply researcher-supplied category definitions: predictions may instead reflect prior associations with category labels or labeled demonstrations. We introduce a counterfactual audit that reassigns definitions among output labels, distinguishing adherence to supplied definitions from preservation of a dataset's original mapping. Across three political-text datasets, models rarely follow reassigned definitions, and standard fine-tuning strengthens the original mapping without reliably increasing definition adherence. We then study retrieval-augmented annotation. Retrieving the nearest training examples generally outperforms retrieving examples separately within each category despite using fewer demonstrations. On the two tasks with substantial gains, direct \(k\)-nearest-neighbor classification recovers 74–93% of the improvement, indicating that neighborhood labels provide much of the predictive signal. Relabeling fixed demonstrations redirects predictions, whereas aligning demonstrations with reassigned definitions substantially increases adherence to the new mapping. Thus, accuracy-improving interventions need not increase codebook adherence; definitions, category labels, demonstrations, retrieval, and training data should be aligned and validated jointly.
Read the full abstract

All authors Viraj Mishra, Shubham Gupta, Dhruv Gupta, Yuehong Cassandra Tai, Qunhua Li

Chemical Physics

Increased Crystal Nucleation in the Glassy State of a Polymer/Water Mixture

Within classical nucleation theory, crystal nucleation requires sufficient supersaturation, but is suppressed at low temperatures due to reduced molecular mobility, leading to a maximum nucleation rate typically above the glass transition temperature $T_g$. Here, we experimentally show that this conventional picture breaks down in a biocompatible polymer poly(2-methoxyethyl acrylate)(PMEA)/water mixture. We find that the number of nuclei increases dramatically below $T_g$, with the effective nucleation rate exhibiting an unusual maximum at approximately $T_g-10$ K. Calorimetric experiments reveal a strong temperature-history dependence of cold crystallization of water in hydrated PMEA, which is explained by the observed nucleation behavior. We propose that the unusual enhancement of nucleation below $T_g$ provides a key mechanism linking high glass-forming ability and cold crystallization upon heating, thereby explaining the striking asymmetry in crystallization behavior between cooling and heating. Our findings suggest that a hidden nucleation-rate maximum below $T_g$ may also underlie cold crystallization in other glass-forming liquids.
Read the full abstract

All authors Mika Kobayashi, Shingo Kobayashi, Kazuyoshi Shiga, Hiroaki Yamada, Masaru Tanaka