Peer Revenue Publish or profit

Papers

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

All of arXiv

Friday 4 September

Networking

WIP: Energy-Efficient LLM-Based Serving Cluster Formulation in Cell-Free Massive MIMO

One way to increase the Energy Efficiency (EE) of 6G wireless networks is to utilize existing network infrastructure more efficiently. This can be achieved by introducing User-Centric Cell-Free Massive Multiple-Input-Multiple-Output (UCCF MMMIMO), which allows for simultaneously serving a single user by multiple Base Stations (BSs). From this perspective, the key challenge is to decide which BSs should serve a given user, known as the Serving Cluster Formulation (SCF). In this paper, we propose to deal with this problem by using an Artificial Intelligence (AI) agent based on a Large Language Model (LLM), targeting improvement of EE. We evaluated the proposed AI agent using a complex, 3D Ray Tracer-based, cellular network simulator, comparing a few GPT models and state-of-the-art algorithms. The results show up to 32% gain in EE of the proposed AI agent compared to the baseline.
Read the full abstract

Strongly Correlated Electrons

Deviation from Fermi-liquid $T^2$ resistivity caused by collective transport

`Fermi-liquid behavior' is nowadays used as a shorthand for quadratic temperature dependence of the electrical resistivity. In a metal, for this to occur, electrons must be cooled well below the Fermi degeneracy temperature, $T_{\mathrm{F}}$. A departure from this behavior is observed at finite temperature. Here, we note that this departure has opposite signs in two distinct types of Fermi liquids. In weakly correlated ones, the upward departure implies a higher exponent of the inelastic resistivity, attributable to additional scattering by phonons. In contrast, there is a downward deviation in strongly correlated metals, which implies either reduced scattering or additional conduction, as in the case of normal liquid $^3$He, in which a distinct contribution to heat transport by a sound mode has been recently identified. Such a channel of conduction smoothly transforms across $T_{\mathrm{F}}$ to the one captured by the Bridgman formula for thermal conductivity of classical liquids. In three metals (UPt$_3$, Sr$_2$RuO$_4$, and heavily overdoped LSCO), according to the available experimental data, the normalized amplitude of excess conductivity at $T/T_{\mathrm{F}}\simeq0.02$ is comparable to what has been seen in $^3$He, suggesting, without establishing, a common origin.
Read the full abstract

Probability

Planar Obliquely Reflected BSVIs on General Filtered Spaces: Non-Symmetric Rotation Fields and Associated Control Problems

We prove existence and uniqueness of a càdlàg solution to a planar backward stochastic variational inequality on a general complete filtered probability space, driven by a square integrable martingale, which may have jumps. The multivalued term is the exterior normal cone operator of a bounded uniformly convex planar domain, and the reflection direction is generated by a time-dependent non-symmetric rotation field. The non-symmetry creates a first-order tangential boundary term that destroys the standard monotonicity and quadratic contraction estimates, used in the symmetric oblique-reflection theory. We overcome this obstruction by constructing an explicit symmetric two-point kernel with state dependent coefficients, whose boundary derivative cancels the leading tangential contribution. A weighted martingale-exponential estimate then controls the second order defect and the jump terms. Under suitable geometric and quantitative compatibility constraints, we obtain the unique strong càdlàg solution. We also formulate associated control problems for the rotation angle and prove existence of an optimal control in a compact class of bounded-rate angle paths.
Read the full abstract

Solar and Stellar Astrophysics

Unresolved triple systems in the aged open cluster Trumpler 19: Discovery, confirmation, and implications

It is difficult to identify unresolved triple and higher-order systems beyond the solar vicinity. Star clusters yield controlled environments with known physical parameters, providing a way to explore homogeneous stellar samples. Combining modern optical and near-IR surveys, our main goal is to explore the region of the color-magnitude diagrams located above the sequence of equal-mass binaries, which is inhabited by unresolved triple and quadruple systems. Specifically, here we aim to discover and characterize the presence of multiple systems in the aged open star cluster Trumpler 19. We selected cluster members using Gaia astrometry and photometry from the VVVX and DECAPS2 surveys. We made optical and near-IR color-magnitude diagrams to select stars located >0.75 mag above the cluster main sequence, including suitable unresolved triple and quadruple systems. We confirm previous results that there is a sizable sequence of unresolved binaries in Trumpler 19. We also report the discovery of a sequence of unresolved multiple systems in this cluster, and confirm some of these multiple systems via radial velocities measured by the ESO KMOS VVVX-GalCen spectroscopic survey. The fraction of multiple stars in this cluster is comparable to that of the solar neighborhood, with the caveat that the local sample probes deeper down the main sequence. The fraction of unresolved triples and quadruples is also three times higher than that of the open cluster M 67, which is of a similar age and chemical composition. The discovery of unresolved triple systems in Trumpler 19 leads to our main conclusion that these systems were able to survive within the cluster for nearly 4 Gyr. In addition, this cluster contains more multiple systems than M 67, which has a similar total present mass. Further investigations are needed to unveil if this is a primordial feature or the result of dynamical evolution.
Read the full abstract

All authors D. Minniti, R. K. Saito, V. D. Ivanov, C. O. Obasi, M. Gómez, P. W. Lucas, J. G. Fernández-Trincado, J. Alonso-García, E. R. Garro, J. Corral-Santana, A. Luna, Z. Guo, R. Kurtev, J. Borissova, V. Fermiano, J. Osses, C. Morris, B. Dias, C. Cáceres, S. Saroon, J. B. Pullen, P. Rodriguez, M. Lad, S. Federle, I. Petralia, D. Bhadrakumar, P. Esteves, M. G. Navarro, T. Palma, L. Baravalle, D. Galdeano, M. V. Alonso, J. L. Nilo Castellon, A. N. Chené

Artificial Intelligence

Recovering Temporal and Geographic Signals from Language Model Embeddings

Understanding whether language-model embeddings encode structured real-world information is important for both representation analysis and information retrieval. We study this question for temporal and geographic signals using a simple projection-based method that operates directly on output embeddings. Given a small set of seed examples, the method defines an axis in embedding space and ranks texts or entities by their projection onto that axis. Our approach is fully black-box and model-agnostic: it requires only embeddings, without access to model weights, internal activations, auxiliary probes, or additional training. This makes it applicable to modern embedding models available only through APIs and provides a lightweight way to analyze whether temporal and spatial dimensions are present in their representation spaces. We apply the method to temporal and geographic datasets and find that embedding projections recover meaningful chronological and spatial structure. These results provide evidence that output embeddings encode signals relevant to time and space, while also offering a practical tool for interpretability and for downstream temporal and geographic information retrieval tasks, such as temporal ordering, geographic ranking, and tagging.
Read the full abstract

All authors Esteban Feuerstein, Victoria Klimkowski, Juan Manuel Ortiz de Zarate, Federico Hernán Suaiter

Analysis of PDEs

On Decay of the Local Mean Oscillations of the Vorticity Direction in Critical Navier-Stokes Flows

We isolate and analyze the geometric PDE governing the evolution of the vorticity direction in the 3D incompressible (unforced) Navier-Stokes equations (NSE), restricted to the case of a critical spatial point singularity where the vorticity magnitude concentrates as $O(|x|^{-2})$, inhabiting the critical Lorentz space $L^{3/2, \infty}$. The PDE consists of the Harmonic Map Heat Flow (HMHF) into the sphere supplemented with the fluid transport, cross-diffusion and tangential strain. The question is whether the NSE mechanics can propagate logarithmic decay of the local mean oscillations of the direction – the condition $\xi \in \mathrm{bmo}_{1/|\log r|}$ – which (in this setting) was shown in the companion paper to prevent finite time blow-up. A key observation is that the $O(|x|^{-2})$ concentration, factored out of the viscous cross-diffusion, generates an outward radial drift $4\nu\, x/|x|^2$ at the core. Under a suitable condition we prove a barrier lemma for the total drift-diffusion operator and show that the HMHF nonlinearity is harmless for the quantity $\frac12|\xi - e|^2$ which is a subsolution on any hemisphere. This yields a transfer theorem: the $\mathrm{bmo}_{1/|\log r|}$ regularity of the direction at the core, uniformly up to the singular time, is controlled by a logarithmic modulus in time of the direction at the inner scale, together with the physical strain. Crucially, the strain enters the direction equation only through its tangential component which vanishes precisely when the direction is an eigenvector of the strain tensor. Since this includes the eigenvector carrying the maximal stretching, the result is in contrast to the classical geometric regularity theory which is built on depleting the full vortex-stretching term and thus confined to configurations of weak stretching.
Read the full abstract

Physics Education

AI in Particle Physics Education: Research Problems and Foundational Skills

Generative AI can produce perfect solutions to standard physics homework, weakening the connection between submitted work and knowledge a student can use independently. We report a redesign of the introductory nuclear and particle physics course at Ruhr University Bochum, in which AI was permitted and encouraged alongside traditional tools on unusually difficult, research-shaped assignments. We found that most students remained engaged with the assignments throughout the course and produced ambitious work, while the course did not reliably secure its foundational objective. We offer this mixed experience as a reference point for discussion.
Read the full abstract

All authors M. Mikhasenko, M. Stahl, I. Segal, D. Parmar, A. Zimmer, A. Kazatsky

Quantum Physics

Tight Lower Bounds for State Tomography with Limited Entanglement

We study state tomography when each measurement acts on at most $k$ fresh copies and no quantum memory is retained between blocks. We prove a lower bound matching the upper bound in [arXiv:2510.07788]. Thus the copy complexity of estimating an arbitrary $d$-dimensional state to trace distance $\epsilon$ is, up to absolute constant factors, $\max\{d^3/(\sqrt{k}\epsilon^2),d^2/\epsilon^2\}$ for every $k$ and all sufficiently small $\epsilon$. This removes the earlier restriction that $k$ be small as a function of the accuracy. The lower bound applies to arbitrary measurements within each block and adaptive choices between blocks. The lower bound already applies in a small neighborhood of any state whose smallest eigenvalue is of order $1/d$, even when the center is known. The main ingredient is a uniform Fisher information bound for one measurement block that depends only on the smallest eigenvalue of the state. The proof avoids the perturbative expansion responsible for the restriction in [arXiv:2402.16353]. Fano's inequality for metric balls and a log-Sobolev comparison between mutual and Fisher information then reduce the adaptive protocol to this block bound [arXiv:1607.00550, arXiv:1902.08582].
Read the full abstract

All authors Ufuk Keskin, Jason Luo, Mahbod Majid, Matthew Radzihovsky

Materials Science

Plasmon drag and photoinduced magnetic effects in plasmonic and magnetic metals

Photoinduced electric effects in plasmonic and magnetic materials under pulsed laser illumination exceed the prediction of the electromagnetic momentum-transfer mechanism by orders of magnitude. In order to get more information on the nature of the effect we study the kinetics of photovoltages in permalloy thin films at different photoexcitation configurations. The photoinduced electric signals consist of magnetically dependent and magnetically independent components which are of significant magnitude and mainly follow the temporal profile of the laser pulse. In contrast, the predicted contribution from the Anomalous Nernst Effect is much weaker and exhibits substantially slower kinetics. These observations suggest that ultrafast photoexcited hot electrons, rather than thermal mechanisms, play the dominant role in generating the observed photovoltage.
Read the full abstract

High Energy Physics (Theory)

Landau and cluster structures of one-loop amplitudes in ${\mathcal N}=4$ SYM in dimensional regularization

Generalized unitarity, Landau analysis, and cluster adjacency encode complementary aspects of scattering amplitudes. We use one-loop planar $\mathcal N=4$ SYM amplitudes in dimensional regularization, at arbitrary multiplicity and helicity, to make their interface explicit. The weight-two symbol decomposes into an LS part, in which maximal-cut leading singularities furnish the coefficients and Landau loci associated with nested cuts organize the ordered symbol entries, an algebraic four-mass sector, and residual terms. Cancellations of certain letters contributed by individual box integrals, as well as further simplifications, are explained by the two-mass triangle relations among box coefficients. We prove these relations using a BCFW-like application of the global residue theorem and show that they can be understood geometrically as different dissections of the same region in the tree amplituhedron obtained by projecting the loop geometry of a triple cut. We then prove that the full rational symbol, including its infrared-divergent part, obeys cluster adjacency in the flag cluster algebra $\mathrm{Fl}_{2,4;n}$ for all multiplicities and helicities. Within the sector depending only on momentum-twistor four-brackets, we conjecture a stronger cluster property in $\operatorname{Gr}(4,n)$ for all helicities and prove it for NMHV amplitudes: the amplitude admits a representation in which every pole of each coefficient is compatible with both symbol entries. Finally, we observe that the algebraic four-mass letters, although non-rational, exhibit a suggestive Sklyanin-bracket pattern.
Read the full abstract

Differential Geometry

Interface foliation near a minimal isoparametric hypersurface for the Allen-Cahn equation

Let $(M,g)$ be a closed Riemannian manifold of positive Ricci curvature. Let $f$ be a proper isoparametric function on $M$, and $\Gamma$ the unique level set of $f$ which is a minimal hypersurface. For any positive integer $k$, and $\lambda >0$ large enough, we construct a solution of the Allen-Cahn equation $ \Delta u + \lambda (u-u^3 ) =0$ which is constant along the level sets of $f$ and has exactly $k$ nodal components. We prove that the nodal components approach the minimal isoparametric hypersurface $\Gamma$ as $\lambda \rightarrow \infty$ and the energy is uniformly bounded for all such solutions.
Read the full abstract

Materials Science

FP-DeErr: Application-Oriented Error Decomposition for Foundation Potentials

Foundation potentials (FPs) have emerged as a new basis for atomistic modeling. While their evaluation using average energy and force errors often indicates near-DFT accuracy, their performance in practical computational studies remains inconsistent. While recent benchmarks evaluate FPs on downstream computational tasks, the underlying errors that determine task success or failure are not always clear. Here, we develop FP-DeErr, an application-oriented error-decomposition framework that resolves FP errors according to the physically meaningful quantities, configurations, and computational stages governing specific tasks. Rather than averaging errors over an entire dataset, FP-DeErr uses decomposed error metrics devised for physically meaningful quantities, focusing on the configurations where errors arise and matter most. We demonstrate FP-DeErr by evaluating multiple state-of-the-art FPs on three fundamental tasks: force prediction for atomistic simulations, energy ranking for substitutional and vacancy orderings, and ion/vacancy migration. These error-decomposition metrics resolve force errors among highly accurate, large-error, and far-from-equilibrium atoms; relative-energy errors among competing orderings, phases, and compositions; and ion migration errors among endpoints and along-path errors. By identifying where FP errors arise, this error decomposition provides targeted guidance for FP development. FP-DeErr also provides an open benchmark, evaluation code, and a public leaderboard for rigorous FP assessment.
Read the full abstract

All authors Kiyan Amirian, Ramanuja Srinivasan Saravanan, Felix Adams, Charles E Schwarz, Yifei Mo

Statistical Methodology

Approximating Bayesian leave-one-group-out cross-validation

When data are grouped, hierarchical or multilevel models are commonly used to account for group-level variation with group-specific parameters. Leave-one-group-out cross-validation (LOGO-CV) is a suitable tool for evaluating predictive performance for new groups, providing an estimator of the expected log predictive density (elpd). Brute-force LOGO-CV requires one model refit per held-out group, often using computationally expensive inference algorithms such as MCMC. This is costly, particularly for large numbers of groups or complex model structures. Commonly used importance sampling approximations, intended to reduce this cost, tend to fail because the group-specific parameters of the held-out group must be integrated out. We identify two key challenges in LOGO-CV elpd estimation: approximating the LOGO posterior and computing the grouped marginal likelihood. We compare 11 strategies, including 5 newly proposed, to address them. Among others, we combine Pareto-smoothed importance sampling or adaptive importance sampling with integration techniques such as Laplace approximation, adaptive Gauss-Hermite quadrature, and bridge sampling. We evaluate these strategies in both simulation experiments and real-world case studies, which show that marginalising over the group-specific parameters substantially improves the reliability of the importance sampling approaches.
Read the full abstract

All authors Anna Elisabeth Riha, Svenja Jedhoff, Paul-Christian Bürkner, Aki Vehtari

Plasma Physics

Novel approach to general curvilinear coordinates for plasma fluid applications

In general geometry, plasma fluid equations include nonlinear geometric sources associated with fictitious forces, which pose significant challenges to computer simulations. We reformulate the plasma fluid hierarchy to rigorously preserve geometry and conservation properties critical to numerical simulations, while concealing the geometric sources. In their discrete form, the reformulated models conserve mass, angular momentum, and energy naturally, by simple analogy with the continuum equations. These conservation properties have minimal requirements in discrete space, namely, the anti-symmetry of the first derivative and the orthogonality of the scalar and cross products. By decoupling magnetic geometry, coordinate systems, and numerical discretization, this enables maximum flexibility while preserving physics fidelity. As a testbed, we apply the novel representation to the resistive magnetohydrodynamic system, which involves a complete set of curvilinear operations. We verify the correctness of the approach using steady state liquid metal flows and the classic Orszag-Tang vortex.
Read the full abstract

Logic

A counterexample to Petrykowski's conjecture

We construct a simple expansion of the theory of nonabelian free groups in which the group admits a global type with bounded left-translation orbit but is not definably amenable. This extends the construction of Chernikov, Hrushovski, Kruckman, Krupiński, Moconja, Pillay, and Ramsey, and gives a counterexample to Petrykowski's conjecture.
Read the full abstract

Optimization and Control

Can Acceleration in Gradient-Norm Minimization Be Anytime? Sharp Last-Iterate Limits in Smooth Convex Optimization

In smooth convex optimization, the gradient norm is a directly observable measure of stationarity. Accelerating a first-order method that minimizes the gradient norm is known to be more delicate than accelerating the minimization of function values. Optimal accelerated methods such as OGM-G (Kim & Fessler, 2021) are known to exist for any prescribed finite horizon, but their coefficients depend explicitly on the length of that horizon, i.e. the number of iterations. We ask what kind of acceleration is feasible when the stopping horizon is unknown to the method, i.e. for horizon-independent methods. Diakonikolas & Wang (2022) conjectured that an $\Omega(N^{-1})$ lower bound on the squared gradient norm holds at every horizon $N$ for any nonadaptive, horizon-independent linear-span first-order method. We disprove this pointwise conjecture by exhibiting a method that achieves near-$N^{-2}$ last-iterate guarantees on a density-one set of horizons. We show instead that an $\Omega(N^{-1})$ lower bound must hold for infinitely many horizons. More precisely, if $\mathcal G_N(\mathcal A)$ denotes the bound on the squared gradient norm after $N$ iterations for a method $\mathcal A$, we prove that $\limsup_{N\to\infty}N \mathcal G_N(\mathcal A) \ge 1/2$ for any method $\mathcal A$. This bound is sharp: the constant $1/2$ is exactly attained by the horizon-independent gradient-descent schedule of Rotaru et al. (2026). In addition, we show that the above two extreme behaviors cannot be achieved by the same method: any method $\mathcal A$ with an $o(N^{-1})$ guarantee on a subsequence of iterates must satisfy $\limsup_N N \mathcal G_N(\mathcal A)=\infty$. In contrast, best-so-far output admits a uniform $O(N^{-2})$ guarantee.
Read the full abstract

Chemical Physics

cuSOAP: a GPU-accelerated Generator of Smooth Overlap of Atomic Positions Descriptor

The Smooth Overlap of Atomic Positions (SOAP) descriptor is one of the most widely adopted representations of atomic environments in molecular machine learning, but the cost of evaluating it and its derivatives remains a principal bottleneck of SOAP-based interatomic potentials, particularly at the large radial and angular basis sizes demanded by complex, multi-species condensed-phase environments. We present cuSOAP, a GPU-accelerated generator of atom-wise SOAP vectors and their analytic derivatives. Built on PyTorch, cuSOAP evaluates closed-form expressions or quadratures for the projection coefficients and their Cartesian gradients for Gaussian-type-orbital and polynomial radial bases, through fused CUDA and Triton kernels that eliminate the multi-gigabyte intermediates of a naive tensor formulation. The package is a drop-in replacement for the CPU-based reference DScribe, reproducing its constructor signature, feature ordering, and output to within ${\sim}10^{-6}$, and accepts structures directly as Atomistic Simulation Environment (ASE) Atoms objects. On an NVIDIA Grace–Blackwell (GB200) node, a single Blackwell GPU generates the full descriptor-plus-Jacobian workload for a 1000-molecule water cluster up to two orders of magnitude faster than DScribe, i.e., from $13\times$ to $227\times$ across the entire $(n_{\max}, l_{\max})$ hyperparameter map, with the speedup growing with the angular band limit $l_{\max}$. Descriptor-only generation scales as $t \propto n^{1.34}$, close to linear, up to a million-atom water cluster, which is processed in 19.5s on one GPU and, through a shared-memory multiprocessing driver, in 5.7s on the four GPUs of the node, demonstrating $\sim$90% parallel efficiency with no sign of saturation as devices are added. These results bring on-the-fly SOAP evaluation for large-scale condensed-phase simulation within reach.
Read the full abstract

All authors Hongyu Yan, Yuqing Xia, Yong Wei, Minghan Chen, Hanning Chen

Artificial Intelligence

CUSP: Decomposable Collective Uncertainty for Multi-Agent Multimodal Reasoning

Aggregating heterogeneous vision-language models (VLMs) can improve multimodal reasoning, but neither an individual model's confidence nor that of the aggregated answer measures reliability at the system level. We present CUSP (Collective Uncertainty through Semantic Opinion Pooling), a training-free uncertainty quantification framework that maps multiple VLM responses to a shared semantic response space, pools them into a pooled semantic opinion, and reports two complementary system-level signals: collective uncertainty, the dispersion of the pooled opinion, and Jensen-Shannon divergence (JSD), the conflict among the model-level opinions. Within this pooled semantic opinion, the unnormalized collective entropy decomposes exactly into the mean of the models' individual semantic entropies and the JSD, separating total dispersion from model conflict. Requiring neither token logits nor calibration labels, CUSP applies to open-weight and commercial VLMs alike. In static multi-VLM ensembles, collective uncertainty is the strongest signal in the small-model regime (0.764 AUROC for prediction-error detection, 0.889 AUARC for abstention), outperforming uncertainty baselines majority voting and naive selection by 4.7 to 15.8 points and widening its margin as the ensemble grows; JSD is strongest in the evaluated commercial regime (0.819 AUROC, 0.910 AUARC) and ranks hard-answer model conflict with AUROC up to 0.982. The pooled prediction also improves accuracy over the average single model by 5.6 to 13.0 points. Over the full trajectory of a multi-step, multi-agent system, subagent collective uncertainty ranks system failures above chance (0.619 AUROC) and gives the best abstention ordering among the evaluated signals (0.699 AUARC).
Read the full abstract

All authors Chung-En Johnny Yu, David Garcia, Brian Jalaian, Nathaniel D. Bastian

Computation and Language

Intra-Prompt Parallel Decoding for Common-Context Question Answering

In common-context question answering (CCQA) tasks, multiple input questions share a common context to base their answers from. However, Large Language Models typically generate each answer using an independent prompt. While existing batching and caching techniques help improve parallelism and reduce repeated computations, the separation of questions across prompts limits the achievable speedup, as modern GPUs are underutilized due to a memory bottleneck during attention. We present Intra-Prompt Parallel Decoding (IPPD), a novel inference method that answers multiple common-context questions in parallel within a single prompt. IPPD directly addresses the bottleneck by efficiently sharing both memory and computation during the attention process, as the next token for every question is decoded in a single inference step. IPPD uses virtual position IDs and attention mask manipulation to generate the same output as standard prompting without requiring fine-tuning or any changes to the LLM architecture. Since all parallelism occurs within a prompt, IPPD is fully compatible with batched inference, even when each prompt features a different context. Our experiments show that IPPD delivers up to 7X the effective throughput as standard decoding without quality degradation, and outperforms prefix caching with PagedAttention in most settings.
Read the full abstract

All authors Theodore Glavas, Nikhita Vedula, Dushyanta Dhyani, Antonios Valkanas, Yilun Zhu, Shervin Malmasi

Medical Physics

A unified dielectric-tensor model of macular form birefringence and dichroism

Optical anisotropy of the Henle fiber layer (HFL) underlies two macular polarization phenomena, form birefringence (the macular cross) and dichroism (Haidinger's brushes), which have conventionally been treated as unrelated. We model the HFL as a radial diattenuating retarder and derive a single complex dielectric tensor: effective-medium theory for parallel cylinders gives the real part, and cylindrical averaging of the measured transmembrane xanthophyll tilt gives the imaginary part. The birefringent slow axis is radial and the absorption axis tangential, perpendicular by structural necessity. The model reproduces measured dichroic ratios of 1.04–1.14 for a single calibrated oriented-pigment fraction, and supplies the input parameters for full electromagnetic treatments.
Read the full abstract

All authors Gary P. Misson, Dusan Sarenac, Dmitry. A. Pushin, Shelby E. Temple

Group Theory

Commensurability and quasi-isometry classification for one vertex one loop tubular groups

A tubular group has a graph of groups decomposition with $\mathbb{Z}^2$ vertex groups and $\mathbb{Z}$ edge groups. This paper gives a classification of one vertex one loop tubular groups $G_{(k,\ell),(m,n)} = \langle a,b,t: [a,b]=1, t a^m b^n t^{-1} = a^k b^\ell \rangle$ up to commensurability and quasi-isometry. We show that tubular groups where the images of the edge maps have nonzero “intersection number" are all commensurable and so quasi-isometric. When the intersection number is zero, there are two quasi-isometry classes and infinitely many commensurability classes. The nonzero and zero intersection number tubular groups are not quasi-isometric.
Read the full abstract

Social and Information Networks

Reddit Community Interventions and Cross-Platform Response: Migration and Policy-Response Discourse on Voat

Platform moderation can disrupt online communities, but the responses that follow do not necessarily take the form of direct migration. Communities may relocate, reorganize, discuss moderation decisions, express censorship concerns, or continue activity on alternative platforms without clear evidence of user-level movement. This study examines Reddit community bans and quarantines and their relationship with activity and migration-related discourse on Voat. Using data from the Multi-Platform Aggregated Dataset of Online Communities (MADOC), we analyze interventions affecting FatPeopleHate, GreatAwakening, MillionDollarExtreme, and CringeAnarchy. We combine computational discourse screening with a stratified independent human-validation procedure and population-adjusted estimates across 90-day pre- and post-intervention windows. The results reveal heterogeneous cross-platform responses. FatPeopleHate experienced a substantial increase in overall Voat activity while the relative prevalence of policy-response and broader relevant discourse declined. CringeAnarchy showed declines in direct migration and policy-response discourse following its permanent ban, although some estimates were sensitive to uncertain human labels. GreatAwakening instead exhibited a modest increase in policy-response discourse without clear evidence of increased direct migration. These findings suggest that platform interventions do not produce a uniform migration process. Overall destination-platform activity and migration-related discourse may move in different directions, highlighting the importance of distinguishing behavioral migration from broader discussion of moderation, censorship, governance, and community rebuilding.
Read the full abstract

Nuclear Theory

Beyond Viscosity Matching: Microscopic Relaxation in Anisotropic Flow

When a gas of scattering particles is disturbed, each angular harmonic of its velocity distribution decays at its own rate. Hydrodynamics, and with it the shear viscosity, depends on only one of these rates, the one for the quadrupole harmonic. A system that scatters only a few times before it stops interacting remembers all of them. This is the situation of small collision systems such as $p+p$, $p+$Pb and O$+$O, where kinetic models tuned to the same viscosity nevertheless predict different anisotropic flow. We show that this difference is not a model uncertainty but a measurement of the remaining rates: to first order in the number of collisions, ratios of flow responses equal ratios of relaxation rates, independently of the initial geometry. For the QCD collision kernel, where the rates also depend on momentum, we obtain the full spectrum of relaxation modes in closed form and show that dilute flow, viscosity, and late-time decay probe three different averages of it. A momentum-resolved kinetic solver confirms that the angular hierarchy is lost first at soft momenta and that the flow response settles at a value no single-rate model produces. The same structure governs two-dimensional electron and atomic Fermi gases.
Read the full abstract

Quantum Physics

Characterizing Privacy Risks of Quantum Machine Learning with Emergent Quantum-Native Access

Quantum Machine Learning (QML) has shown rapid advances by utilizing quantum computing for machine learning tasks. Meanwhile, the privacy risks accompanying QML is also starting to be studied, which inherit privacy leakage channels from "classical" ML and also quantum-unique risks. Existing work on privacy-preserving QML largely focuses on a QML-as-a-service scenario, which generally assumes that the QML model owner provides only classical bit outputs to queries, while users (and adversaries) have only classical computing abilities. However, this view is increasingly challenged in a quantum-native world of quantum-capable users/adversaries, which may have access to both quantum computing abilities and access to quantum information output from service providers. In this paper, we aim to bridge this gap by examining membership inference attacks against QML models by demonstrating that increasing quantum access and quantum computing abilities provides provable theoretical privacy leakage and empirical adversarial gain. However, the probabilistic nature of QML introduces a gap between theoretical and empirical adversarial advantage. These results show that existing research on privacy leakage in QML models underestimates privacy leakage in emergent quantum-native access regimes, and we hope to establish a first step in examining potential privacy leakages for QML in the quantum-native world.
Read the full abstract

Cryptography and Security

XAI-SDN: An Explainable Entropy-Guided Machine Learning Framework for Real-Time DDoS Detection in Software Defined Networks

One of the biggest risks faced by Software Defined Networks (SDN) is the Distributed Denial of Service (DDoS) attack in which a compromised controller can make an entire network unusable. To address these challenges, we suggest an entropy-guided machine learning framework, called XAI-SDN, for real-time DDoS detection in SDN environments which is lightweight and explainable. The framework extends the flow features extracted by CICFlowMeter with eight Shannon entropy metrics obtained by an $\mathcal{O}(1)$ rolling algorithm and uses a Random Forest classifier with SHAP TreeExplainer for providing transparency at the prediction level. On a fixed temporal split, XAI-SDN achieves an accuracy of 99.9987%, a macro F1-score of 99.9621%, and an AUC-ROC of 1.0000 on the full 3.59 million flows of the CIC-DDoS2019 SYN benchmark. The pipeline sustains 0.0165~ms per flow (60{,}606 flows/s) without the use of SHAP and 0.5122~ms per flow (1{,}953 flows/s) with full support of SHAP under the 99.14% prevalence of DDoS traffic, which is a step towards achieving a balance between the detection performance and operational transparency in next-generation SDN security.
Read the full abstract

All authors Adeel Ahmad, Ali Akarma, Ahmad Ali, Hammad Muneer, Toqeer Ali Syed

Metric Geometry

Sandwich Slice: Optimal Transport Potentials Are Optimal Generalized Slicers

Sliced optimal transport replaces a transport problem in Rd by one-dimensional problems along scalar slices, and has been used both to lower bound the Wasserstein distance W1 (max-sliced distances) and, more recently, to upper bound it by lifting a one-dimensional optimal plan back to the ambient supports (min-sliced plans). We show that for 1-Lipschitz slicers these two directions are two sides of a single sandwich inequality and we characterize when it collapses to equality. If the slicer class contains an optimal Kantorovich-Rubinstein potential, then the same slicer simultaneously maximizes the max-sliced lower objective and minimizes the min-sliced upper objective, and both equal W1. We define the upper objective through a tie-broken transport problem, which removes the dependence on the one-dimensional ordering and yields the identities without any uniqueness assumption on the slice-optimal coupling. The result requires the slicer to be nonlinear in general, consistent with the known failure of exact linear max-sliced/W1 equivalence in dimension greater than or equal to 2, and identifies Kantorovich-Rubinstein potentials as optimal generalized slicers.
Read the full abstract

All authors Rocio Diaz Martin, Xinran Liu, Matthew Thorpe, Soheil Kolouri

Quantum Gases

Generating quantum error correcting codes from topological pre-thermal scars

Quantum many-body scars are rare nonthermal states embedded within otherwise thermalizing spectra, making them promising candidates for storing and manipulating quantum information. However, a general framework that can both identify such states without prior knowledge of their microscopic scarring mechanism and determine when they enable a quantum error correcting code has remained elusive. Here, we develop an end-to-end framework connecting nonthermal-state discovery to certifiable quantum error correction using a many-body spectral localizer. By searching for states jointly localized in energy and a diagnostic observable in systems with a many-body grading symmetry, the spectral localizer identifies candidate scars within dense many-body spectra in tandem with a topological index whose shifts determine their graded subspace. For suitable pairs of same-graded scars, the grading renders grading-odd errors exactly detectable, while their localization can suppress the logical action of sufficiently local grading-even error products, yielding an approximate Knill–Laflamme condition and an operator-algebra quantum error-correction structure. We demonstrate the scar-identification framework in one- and two-dimensional density-difference-dependent bosonic models and in the PXP model, and explicitly construct the resulting approximate quantum code in the one-dimensional bosonic system. Our results establish a systematic route from identifying nonthermal many-body states to exploiting them for protecting quantum information, and position pseudospectral methods as a broadly applicable tool for discovering useful behaviors in interacting quantum systems.
Read the full abstract

All authors William N. Faugno, Frank Barrows, Terry A. Loring, Nathan Goldman, Alexander Cerjan

Machine Learning

Analysis of Respiratory Sinus Arrhythmia with Neural Networks

The paper introduces a neural network-based approach for analyzing ECG signals to estimate respiratory rate by leveraging the phe- nomenon of Respiratory Sinus Arrhythmia (RSA). Our method employs a deep learning model trained to predict respiratory waveforms directly from ECG input data. To achieve this, we developed and evaluated three different neural network architectures capable of automatically extract- ing relevant features from ECG signals without the need for manual preprocessing. The proposed approach offers a robust and scalable solu- tion for non-invasive respiratory monitoring, with potential applications in healthcare and wearable technology
Read the full abstract

Analysis of PDEs

Global asymptotic stability of KdV-Burgers fronts in a weakly two-dimensional model

We study front-type solutions of nonlinear dispersive-dissipative PDEs modeling the propagation of undular bores in a channel in two dimensions. The system extends the Korteweg-de Vries-Burgers (KdVB) equation by incorporating weak transverse motion, and admits the KdVB fronts as one-dimensional solutions. We investigate their stability under general two-dimensional perturbations. We prove that a one-dimensional front is a global asymptotic attractor in the weakly two-dimensional setting when the channel is sufficiently narrow in the transverse direction and the relative dispersion parameter lies in a range for stability to one-dimensional perturbations. Particularly, the front is unique up to spatial translations. The proof extends the energy method for temporally-modulated perturbed solutions, developed previously in the one-dimensional setting, to accommodate the transverse dynamics.
Read the full abstract

Materials Science

Bridging simulation length scales with cellular automata

Multiscale simulation requires coupling physics models that operate at different characteristic length and time scales, because no single method spans the range needed for most problems of interest. Moving to a coarser-grained representation unlocks longer length and time scales, but it discards the microscopic interactions that build morphology. A fine-grained model can be initialized from an arbitrary packing and left to self-assemble into a physically meaningful structure; a lower-resolution model cannot, and must inherit its starting configuration from a higher-fidelity simulation. The length scales accessible to the coarse-grained model are therefore set not by the coarse-grained method itself, but by the largest fine-grained configuration that can be affordably equilibrated. A representative example is the scale-up from particle-based molecular dynamics (MD) to a lattice-based representation such as kinetic Monte Carlo (kMC). In this work, we present a cellular automata (CA) approach for generating arbitrarily large lattice starting configurations. CA is a natural fit for this task: short-ranged local rules drive the evolution of a lattice, and their repeated application gives rise to emergent long-range order, thematically mirroring how short-ranged interactions in MD produce self-assembled morphology. We use a configuration from a higher-fidelity simulation as a training set and learn the CA rules from it via logistic regression. As a proof of principle, we develop these rules for a hydrated anion exchange membrane (AEM), generate new starting configurations, and benchmark their performance in mesoscale kMC simulations against an MD-derived "ground truth." We then demonstrate the ability to generate substantially larger lattices and show that their behavior in kMC is consistent with that of the smaller CA benchmark configurations.
Read the full abstract

All authors John J. Karnes, Esteban D. Gadea, Shakkira Erimban, Ignacio J. Bombau, Valeria Molinero