Peer Revenue Now accepting positions

Papers

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

All of arXiv

Monday 9 November

Data Analysis and Statistics

Distinguishing thyroid nodule malignancy using digital palpation

Accurate pre-operative diagnosis of thyroid nodules remains a major clinical challenge. Current diagnostic techniques rely on fine-needle aspiration and demonstrate high rates of inconclusive results, often leading to avoidable thyroidectomies with lifelong implications. Here, we introduce a mechanics-based diagnostic method that leverages Digital Palpation to determine malignancy and identify tissue subtypes in human thyroid nodules. From a cohort of patients undergoing thyroidectomies, we evaluated ex vivo thyroid specimens and processed the resulting mechanical data from 36 tests in 10 thyroid nodules. Principal Component Analysis reveals that these experimental results are distributed within a three-dimensional space of interpretable scalar features reflecting tissue grittiness, strength, and toughness, which map closely to descriptions of malignant and benign nodules given by clinicians. Our results further demonstrate that malignant nodules exhibit distinct mechanical signatures compared to their benign counterparts. This revealed a statistically significant difference (p = 9 x 10^-4) and separability (95% lower-bound Hedges' g = 1.08) between benign and malignant nodules along a direction in principal component space representing effective stiffness. Preliminary clustering patterns were also observed among malignant subtypes, including follicular, oncocytic, and papillary thyroid carcinomas. Together, these results highlight the potential of Digital Palpation as a novel mechanics-based method for tissue characterization and motivate its further development toward diagnostic applications.
Read the full abstract

All authors Brendan M. Unikewicz, Joseph E. Bonavia, Ekaterina L. Koelliker, Abigail Klein, Katie C. Spaeth, Hannah M. Varner, S. Kiana Naghibzadeh, Yoav Horev, Sareh Parangi, Tal Cohen

Optimization and Control

Differential Flatness of Nonlinear Consensus Networks: A Graph-Theoretic Perspective

We study the property of differential flatness for networks of identical flat agents coupled through a consensus protocol, where a single exogenous input acts on one agent and the output of another agent is measured. We distinguish whether the network admits a flat output at all from whether the measured task output is a flat output. For the former, we show that flatness at a synchronized equilibrium requires controllability of the Laplacian leader–follower pair, which is also sufficient for linear agents but not for nonlinear ones. For the latter, we show that the relative degree of the task output is determined by its graph distance to the input node, so that it is a flat output if and only if the graph is a path with the input and output nodes at its ends. Numerical examples illustrate how flatness enables finite-time transitions between consensus configurations through a single input.
Read the full abstract

Mesoscale and Nanoscale Physics

Quotient Index for Symmetry-Protected Zeros and the Tenfold Way in Mackey's Construction of Super Corepresentations

Symmetries that anticommute with a Hamiltonian can protect zero modes by obstructing their removal through symmetry-compatible perturbations. We formulate this obstruction as a quotient of the additive group of symmetry representations by the image of the map that forgets a compatible representation grading. To compute the quotient, we use Mackey's method to construct all irreducible super corepresentations of finite graded symmetry groups, allowing antiunitary operations and projective factor systems. Starting from the unitary Hamiltonian-preserving normal subgroup, this construction yields two complex and ten real structural types, characterized by residual Altland–Zirnbauer symmetry algebras. The residual and induced graded modules have isomorphic graded commutants, and their represented symmetry algebras share a graded Morita class that is super opposite to that of the commutant. We derive character criteria that identify these types. Exhaustive enumeration of the restriction relations gives independent sector contributions $0$, $Z$, or $Z_2$, with only integer factors in the purely unitary case. In the finite semisimple setting, the complete index vanishes if and only if all zero modes can be lifted by a symmetry-compatible perturbation. Applications to one-dimensional end states and interpolation cylinders of disordered boundary Dirac theories recover protected multiplet counts and connect the quotient to bulk topology and sigma-model topological terms using bulk–boundary correspondence, texture-index theorems, and fermion-phase calculations.
Read the full abstract

Earth and Planetary Astrophysics

The Neptunian ridge depends on the choice of linear or logarithmic period

The Neptunian ridge has been reported as an overdensity of planets with radii of $\simeq$5-8 R$_\oplus$ and orbital periods of $\simeq$3-6 d, located between the Neptunian desert and the savanna. It was first identified in Kepler data and has recently been reanalysed using our TESS occurrence rates, where it appears as an approximately twofold excess relative to neighbouring regions. We show that the reported excess arises from expressing the occurrence density per unit period, $\mathrm{d}N/\mathrm{d}P$, rather than per unit logarithmic period, $\mathrm{d}N/\mathrm{d}\ln P$, where the logarithmic form is standard in planet demographics. Because the savanna interval used for comparison is twice as wide in period as the ridge, a population uniform in $\ln P$ naturally produces a factor-of-two contrast in $\mathrm{d}N/\mathrm{d}P$. When expressed per unit $\ln P$, the published TESS and Kepler ridge-to-savanna contrasts decrease from 1.6-2.2 to 0.8-1.2, all consistent with unity. A simulated population with a desert but no intrinsic excess reproduces the appearance of a ridge in $\mathrm{d}N/\mathrm{d}P$ precisely. Reanalysing the TESS-SPOC occurrence rates of 5.5-8.5 R$_\oplus$ planets around FGK stars, we find a ridge-to-savanna ratio of $0.91^{+0.21}_{-0.16}$ per unit $\ln P$, which does not support the claimed overdensity at the ridge. Taken together, these tests imply that the published ridge is a result of the scale choice used, rather than a definitive property of the observed planet population.
Read the full abstract

Optimization and Control

Completely monotone memory systems

Linear time-invariant (LTI) relaxation systems are characterized by a completely monotone impulse response. This paper proposes an extension of this property to nonlinear systems: a system is said to have completely monotone memory if a positive increment in the past input leads to a completely monotone increment in the future output. We show that this property constitutes a strong form of monotonicity, while also imposing stability constraints that closely resemble fading memory, contraction, and incremental stability. We also show that completely monotone memory is equivalent to classical relaxation for LTI systems and to a combination of monotonicity and stability for a class of scalar state-affine systems. Finally, we illustrate the relevance of our framework to the gating dynamics of conductance-based models.
Read the full abstract

Functional Analysis

Gelfand and sampling numbers of weighted Wiener spaces

The aim of this paper is to establish sharp asymptotic bounds on the Gelfand numbers of weighted Wiener spaces $\mathcal{A}^r_\theta$ measured in $L_q$. This is done by using results for Gelfand numbers of sequence spaces and applying them to each layer of a hyperbolic cross to obtain results for Gelfand numbers between different (weighted) Wiener spaces. These can then be used to establish bounds in $L_q$ by combining them with known results and techniques for Besov spaces. This significantly narrows a gap in the current research, since the Gelfand numbers provide a lower bound on the (non-linear) sampling numbers, whereas upper bounds on this quantity have been studied in multiple papers via an upper bound given by the best m-term approximation.
Read the full abstract

Computer Vision

WanderBench: Can World Models Explore Dynamic Worlds from Video?

Video world models conditioned on an input video make it possible to explore real, recorded dynamic scenes, enabling re-shooting of footage, free-viewpoint replay, and real-to-sim environments for embodied agents. Such uses require the generated rollout to stay faithful to the recorded world, yet existing benchmarks start from a single image or a text prompt, and video-conditioned generation is evaluated mainly on perceptual quality and camera following. These evaluations do not directly assess whether exploration preserves subject identity, event progression, and scene consistency upon revisiting a location. We introduce WanderBench, a benchmark for exploring dynamic worlds from video. It comprises 745 single-shot source videos of real and AI-generated content in three difficulty tiers (3-25 s), paired with camera routes given as keyboard actions, and evaluates six dimensions with 24 human-validated sub-metrics. Benchmarking eight models, we find that as clips grow longer, models keep the static scene and visual quality but lose camera control and the source dynamics, and that none keeps revisited places consistent. We further propose WanderLand, a training-free framework that conditions each frame of a frozen model on the same-instant source and, on revisits, on its own earlier static content. WanderLand improves five models on every dimension, by up to 8.3 points in source-dynamics fidelity.
Read the full abstract

All authors Yuhang Chen, Jiheng Li, Jiale Wang, Zai Wang, Guangyue Liu

Cosmology and Nongalactic Astrophysics

BAO and Type Ia Supernovae in LTB Cosmology: Constraints on Radial Inhomogeneity and Acoustic-Ruler Evolution

We investigate how baryon acoustic oscillations (BAO) and Type Ia supernovae constrain radial density profiles and acoustic-ruler evolution in Lemaitre-Tolman-Bondi (LTB) cosmology with a cosmological constant. Extending a homogeneous Friedmann-Robertson-Walker analysis requires calculating both distances and unequal radial and transverse stretching of a single initial ruler. Combining transverse BAO with supernovae isolates relative ruler evolution without an absolute calibration, while radial BAO supplies complementary information. We connect this test to exact nonlinear models with a low-density region compensated by surrounding excess matter, varying its depth, width, distance from the observer, and background matter density. Using DESI DR2 BAO and DES-Dovekie supernovae with full covariance and identical calibration freedom, the LTB fit improves $\chi^2$ by 8.523 over flat $\Lambda$CDM for three additional parameters, mainly through the supernovae. DESI strongly rejects the particular deep profile favored by supernovae and angular BAO alone. However, a narrow depression near the observer and a broader, shallower depression farther away fit almost equally well, with $\Delta\chi^2=0.647$. The combined data therefore constrain particular geometries without uniquely reconstructing the density profile or ruler evolution. The inference is conditional on supernova selection and interpreting released BAO summaries through local ruler transport at effective redshifts. Homogeneous simulations show that the fit improvement does not establish a detection of inhomogeneity. Finally, in the local approximation, common LTB ruler evolution shifts angular BAO and the transverse component of three-dimensional BAO together and cannot alone explain their discrepancy.
Read the full abstract

Systems and Control

Degradation-Aware Participation of Battery Storage in Frequency Regulation Markets

As data centers make the load on the power grid fluctuate faster, frequency regulation grows more valuable. Battery energy storage systems (BESSs) are well suited to provide frequency regulation, but providing this service speeds up aging, especially during operation at extreme state of charge (SoC). In the PJM interconnection, the grid operator supplies a regulation signal that a participating BESS is expected to follow, within a capacity that it must bid before the signal is known. Accordingly, we decompose the degradation-aware frequency regulation problem into a bidding layer and an operations layer. For bidding, we propose an adaptive strategy, assuming perfect tracking in which the bid depends on the SoC expected at the start of the bid period by means of choosing basepoints. For operation, the BESS further protects its battery by deliberately giving up a perfect performance score, to keep its SoC away from the extremes. On the PJM regulation signal and prices of 2023, adaptive bids lengthen the battery life by $31\%$ against bidding full capacity, for $2.2\%$ of the revenue. Giving up a perfect score in operation lengthens the battery life of constant bids by $10\%$ to $27\%$, at a change in revenue of at most $0.2\%$.
Read the full abstract

Statistical Mechanics

Resolving the backbone of multiscale interactions through mutual information rates

Natural systems are often composed of processes evolving on multiple timescales and coupled through complex interaction networks. Inferring such networks is especially challenging when timescales are widely separated, since resolving the fastest processes while accurately sampling the slowest ones is generally prohibitive. In this work, we propose a framework to reconstruct a multiscale interaction tree by combining a rescaled version of the mutual information rate, its conditional counterpart, and the ordering of timescales. We derive an exact theory for stationary Gaussian processes, finding that these measures identify directly connected processes, with the asymptotic mutual information rate distinguishing fast-to-slow from slow-to-fast couplings. We show that this reconstruction can be achieved from the spectral matrix and illustrate its robustness to finite trajectory lengths, moderate timescale separation, and weak nonlinearities. Our work paves the way for exploiting the presence of multiple timescales in real-world systems as a source of structural identifiability rather than an obstacle to network inference.
Read the full abstract

Optimization and Control

Optimal Actuator Support Design for Infinite-Horizon Linear-Quadratic Parabolic Control Problems

We study an optimal actuator design problem for infinite-horizon LQ control problems governed by parabolic equations. The control acts only on a small measurable subregion of the spatial domain, and the objective combines the infinite-horizon control value with a cost for the actuator region. Using the Riccati representation of the value function, we derive a comparison identity for two actuator supports and obtain an exact expansion of the objective. This leads to a topological derivative with respect to local changes in the support. We prove a localized Riccati sensitivity estimate to control the remainder and show that the infinite-horizon term in the derivative can be computed by solving a Lyapunov equation. This yields a computable support-update algorithm with a measure-controlled Armijo search. Under explicit uniform remainder and admissibility assumptions, its first-order stationarity defect tends to zero. Numerical experiments illustrate these results.
Read the full abstract

Artificial Intelligence

SciMARS: Multi-Agent Recursive Self-Improvement for Closed-Loop Scientific Discovery

We propose SciMARS, a framework for recursive task-local adaptation of multi-agent harnesses in closed-loop scientific discovery. Multi-agent systems coordinate analysis, experimental design, and verification as scientific evidence evolves. However, two problems limit their effectiveness: (i) without shared memory, agents lose observations and repeat work; (ii) fixed role procedures, handoffs, and memory policies perpetuate mistakes despite retained evidence. We address these problems with a shared evidence graph and separate component editors. SciMARS adapts roles and coordination; SciMARS + Memory adds a third editor for how agents record, retrieve, and receive evidence. Candidate probes guide composition and selection of the next harness. We evaluate scientific discovery through experimental-data replay and a digital twin. SciMARS + Memory resolves 12/18 GPT-6-Astra and 2/18 GLM-5.3 attempts on Autonomous Science Bench, compared with 8/18 and 1/18 for the base multi-agent team. On ScienceAgentBench and AppWorld, it solves 14/38 and 11/20 tasks, compared with 11/38 and 5/20 for the base team. These results support combining shared memory with harness adaptation across scientific and interactive tasks.
Read the full abstract

Analysis of PDEs

Single-point determination of a potential and a nonlinearity in a semilinear biharmonic equation

In this paper, we study the inverse problem of determining the coefficient $q$ and the nonlinearity $F$ in the semilinear biharmonic equation $\Delta^2 u+q(x)u=F(x,u)$ with Navier boundary data. The measurement consists of $\partial_\nu u(x_0)$ at a fixed boundary point $x_0$. We apply the mixed variation approach to linearize the nonlinear problem and construct semiclassical boundary-localized solutions for the linearized equation. Under suitable analyticity, nonresonance, and nondegeneracy assumptions on $q$ and $F$, we prove that this boundary point measurement uniquely determines both $q$ and $F$. We also establish boundary determination of all normal derivatives of $q$ and of the first nonzero Taylor coefficient of $F$ on a nonempty relatively open subset of the boundary. When the linear potential $q$ is known, we prove that either component of the Navier boundary input alone suffices to determine $F$, and that the polynomial nonlinearity can be uniquely determined without requiring any analyticity conditions. For suitable finite-dimensional classes of nonanalytic potentials, we also obtain, for fixed $F$, the stable determination of $q$ from finitely many boundary inputs and the associated point measurements. In addition, as an important supplement, we provide nontrivial examples verifying the main assumptions, as well as counterexamples concerning the recovery of the linear coefficient. Finally, we present numerical experiments that demonstrate the theoretical results.
Read the full abstract

All authors Song-Ren Fu, Li Li, Hongyu Liu, Zhikun Zhang

Statistics Theory

A Strict Upper Bound for Three-Point Weighted Selection in Planar Vector-Valued Regression

We prove $F_{\mathrm{weighted}}(2,2,3)<11/6$ for minimum-Frobenius-norm least-squares regression with two input and two output coordinates, improving the upper bound $15/8$. More precisely, a universal $\varepsilon_0>0$ satisfies $\mathcal T_3\le(5/6-\varepsilon_0)E$ for every finite residual frame. The proof combines an exact scalar certificate, rigidity of equality in planar selection, and a compactness argument that controls disappearing weights. Together with the known lower bound, this gives $13/8\le F_{\mathrm{weighted}}(2,2,3)<11/6$. No numerical value of $\varepsilon_0$ is claimed.
Read the full abstract

Mathematical Physics

The finite projective classification of complex spinors via momentum maps

We describe the projective orbits of complex spinors from the point of view of the momentum map. We treat the irreducible spin modules in odd dimensions 1,3,5,7,9,11 and the half-spin modules in even dimensions 2,4,6,8,10,12,14, which are the only cases with finitely many orbits. In the Clifford–Fock model, we construct representatives as sums of pure basis spinors, and the coefficients are chosen so that the representative is a critical point of the squared norm of the momentum map. The resulting momentum values in a fixed Weyl chamber distinguish the orbits. Completeness follows from Vinberg's theory of graded Lie algebras and stabilizer actions, independently of the classical spinor classifications. We recover the classical counts of 1,1,1,2,3,5 orbits in the odd cases and 1,1,1,2,2,4,9 in the even cases, and compute the projective dimension of every orbit. In dimension fourteen we also determine the closure order of the orbits, using the fact that the minimum of the squared momentum norm increases strictly under degeneration.
Read the full abstract

Superconductivity

Direct imaging of vortex lattices across fractional fillings in two-dimensional Josephson junction arrays

Josephson junction arrays are predicted to host a rich sequence of commensurate vortex lattices. However, direct imaging of these configurations remains limited, while transport measurements cannot resolve the underlying vortex arrangements or how order breaks down. Using scanning superconducting quantum interference device (SQUID) microscopy, we resolve individual vortices and directly image this long-predicted hierarchy spanning filling factors $f$ = 1/8 to 1/2 in two-dimensional Au-Nb Josephson junction arrays, including several states with no discernible transport signatures. At $f$ = 1/4, two structurally distinct vortex orders with 4$\times$4 and 2$\times$4 unit cells coexist, consistent with their energetic degeneracy in the frustrated XY model. Away from commensurate filling, we resolve defect structures tied to the underlying vortex-lattice order: the $f$ = 1/2 checkerboard retains its order around isolated excess vortices and vacancies, whereas the $f$ = 1/3 staircase fragments into domains separated by domain walls. These results reveal how frustration, competing configurations, and defects shape the spatial coherence of commensurate vortex order in engineered superconducting lattices.
Read the full abstract

All authors Cequn Li, Austin R. Kaczmarek, Omri Lesser, Maciej W. Olszewski, Katja C. Nowack

Statistical Mechanics

Memory in biological Lévy walks from the dynamic-entropy formalism

Lévy walks are recognized by the histogram of their flight lengths, while the renewal hypothesis on which the textbook model rests, that every flight is drawn afresh, is seldom tested. We test it with the dynamic-entropy formalism. Maximizing the causal entropy of a sequence of logarithmic flight lengths at fixed lagged correlations yields a chain of continuous spins on a compact interval, whose depth, selected by the data, tells a renewal walker from a Markovian one and from one with a longer memory. With one lag the chain is exactly solvable. A duality that exchanges long and short flights locates a first-order condensation of the flights at the upper cutoff, with the logarithmic range of the flights in the role of system size, and a positive memory makes the histogram heavier-tailed than the rule that generates each flight. For groups, the same construction separates the memory of each individual from delayed interactions and from drives common to all units, which an analysis of the group average confounds, and it gives the collective speed a critical point. On open data, renewal is rejected in seven of ten conditions of Drosophila larvae and in every termite, whose memory reaches a few flights back and lowers the exponent of the histogram by more than half a unit below that of the one-step rule. Wild baboons align with where their nearest neighbors, those ahead above all, were heading 10 to 30 s earlier, and homing pigeons are coupled across their small flocks. Zebrafish T cells, a negative control, remember their own speed, but no delayed coupling between the speeds of different cells survives the controls.
Read the full abstract

Mesoscale and Nanoscale Physics

Topological states in fractals through a self-similar tensor network encoding

Tensor network methods have been instrumental in addressing exponentially large many-body problems. Most recently, these techniques have also been applied to single-particle tight-binding Hamiltonians, allowing observables to be evaluated on exponentially large lattices. Here, we show that the scale-separated nature of these methods allows for a natural encoding of self-similar structures, such as the Sierpiński gasket, a scale-invariant fractal with Hausdorff dimension $\log 3/\log 2 \approx 1.58$, yielding an exact and compact representation of Hamiltonians on such lattices. Using this encoding, we study a Chern insulator, the flux kagome model, confined to this fractal geometry, and develop methods to investigate its topological states within a tensor network framework. This allows us to reach much higher fractal generations than would otherwise be possible with dense-matrix methods, resolving the density of states of a deep fractal with over 100 million sites. Furthermore, the larger structures allow us to resolve topology in smaller gaps, suggesting that these methods could be instrumental in understanding the interplay between topology and non-integer dimensionality.
Read the full abstract

All authors Zebedeus F. Osseweijer, Jagoda Malag, Jose L. Lado, Tiago V. C. Antão, Anouar Moustaj

Physics and Society

Teamknitt: Configuring, Observing and Intervening in LLM-Agent Teams through the Lens of Dynamical-Systems and Human-Team Theory

Most frameworks for teams of large language model (LLM) agents fix the team's shape inside the software: who talks to whom, how much is passed along, and what a human can do once a run has started. Human-team research has long asked those questions: where team phenomena live (in members or in their interaction), how team-level properties arise, and what separates what a team does from how a team is. I propose using that theory as the organizing lens for configuring, observing and intervening in LLM-agent teams, without assuming its findings transfer. I introduce teamknitt, a framework and open reference implementation in which a user configures a team of 1-16 AI agents for task completion, solution generation, achieving a goal or a research mission; selects one of fourteen deterministic routing topologies or a custom directed graph; chooses a sharing profile that limits what may travel along permitted routes; and binds sources, representations, work dependencies, handoffs and an expertise directory. The recorded process can be observed through non-system views (member reports and behavior) and system views (distributed problem-solving, differentiated transactive memory, interactive coordination, team dynamics). Interventions are explicit and human-led: reviewed reconfiguration between runs, planned disturbances, and private feedback, pause, resume or stop applied at dependency-wave boundaries, with "queued", "applied", "assembled", "understood" and "improved" kept as separate claims. Emergent states such as trust and shared understanding are declared unmeasured. This is an idea and system-description paper with falsifiable hypotheses on topology, information flow and intervention timing, including whether the fractal scaling of communication output between members, and between members on different models, relates to task and team performance. No comparative result is reported.
Read the full abstract

Optimization and Control

Learning inexact alternating minimization for convex problems with quadratic coupling

Alternating minimization is a standard tool for optimization problems with two coupled variable blocks. Its cost is dominated by the inner subproblem solvers, which rarely have a closed-form solution. In this paper, we propose to learn an inexact alternating scheme in which neural networks approximate the subproblem solution at each iteration. For convex problems with quadratic couplings between the variable blocks that satisfy a quadratic growth condition, we derive a worst-case bound on the optimality gap that exposes the per-iteration error that each network is trained to minimize in a greedy fashion. Combining the bound with a PAC-Bayes argument, we obtain a fixed-horizon, high-probability bound on the expected optimality gap over the distribution of problem instances. We apply the learned scheme to robust matrix completion and low-dose CT reconstruction with dictionary regularization where we achieve a 5x-13x speedup over the inexact first-order baseline at matched accuracy. A brief end-to-end fine-tuning stage extends this up to 30x and outperforms other learned schemes. The code is available at https://github.com/paulhausner/learned-alternating-minimization.
Read the full abstract

All authors Paul Häusner, Jevgenija Rudzusika, Jens Sjölund, Ozan Öktem

Earth and Planetary Astrophysics

Kinematic Signatures of Stellar Flybys in Protoplanetary Discs

Recent high-resolution molecular line observations of protoplanetary discs have revealed large-scale non-Keplerian features in the gas kinematics of several systems, which can be interpreted as signatures of warps in the disc. Warps can arise from a range of physical mechanisms, and it remains unclear whether the observed kinematic residuals can be uniquely attributed to warping, or to different competing effects. Stellar flybys are one such mechanism capable of both warping a disc and directly perturbing its kinematics. Using 3D smoothed particle hydrodynamics simulations, we study the evolution of protoplanetary discs perturbed by inclined and coplanar stellar flybys, focusing on timescales of several thousand years after periastron rather than the immediate aftermath of the encounter. We find that inclined stellar flybys initially induce warping and twisting in the disc, but that this geometric perturbation is damped on a timescale set by the disc viscosity. Despite this geometric damping, synthetic ALMA observations of the CO kinematics show that complex non-Keplerian spiral features persist in the velocity residuals long after the warped morphology has relaxed, including in coplanar encounters where no warp is ever induced. Fitting these residuals with a simple warped-disc model, the degree of warping in the disc is systematically overestimated. We therefore conclude that kinematic residuals alone are insufficient to establish that a disc is warped, as flybys induce kinematic complexity independently of warps. We further show that scattered light observations provide an independent diagnostic, showing spiral arms and lateral asymmetries.
Read the full abstract

All authors Vasundhara R. Prasad, Cristiano Longarini, Carolin N. Kimmig, Cathie J. Clarke

Numerical Analysis

Investigating parameter dependence for counterexamples to Payne's nodal line conjecture using HPC

The behavior of the nodal line of the second eigenfunction of the negative Laplacian has been widely studied. Dahne could show numerically for one domain with six holes in two dimensions that it is possible to reject Payne's nodal line conjecture. We investigate the behavior of the nodal line if it does not touch the boundary, specifically the general shape, its prediction, its dependence on the parameters, and the numbers of holes. In this paper, we could address these issues numerically with a very efficient parallelized algorithm showing the rejection for various domains with six and five holes.
Read the full abstract

Number Theory

Independence of Hasse–Weil $L$-functions over rings of Dirichlet series

We study linear and algebraic independence of Hasse–Weil $L$-functions of abelian varieties over $\mathbb{Q}$, with Dirichlet series as coefficients. For nonzero varieties that are pairwise non-isogenous over $\overline{\mathbb{Q}}$, we prove that their $L$-functions with arbitrary Dirichlet twists, together with all their derivatives and $1$, are linearly independent when the coefficient series use only primes from a set of natural density zero. For untwisted $L$-functions, this independence is equivalent to the condition that the varieties are pairwise non-isogenous over $\mathbb{Q}$. If the individual monodromy groups are connected, the result for varieties that are pairwise non-isogenous over $\overline{\mathbb{Q}}$ extends to larger coefficient rings: the prime supports of coefficients are unrestricted outside a prescribed set of positive upper natural density and meet it in a set of natural density zero. We also characterize algebraic independence of untwisted $L$-functions in terms of the multiplicities of the simple isogeny factors of the varieties. The proofs combine an arithmetic-function lemma of Kaczorowski–Molteni–Perelli with Frobenius density arguments and require no analytic continuation or automorphy.
Read the full abstract

Earth and Planetary Astrophysics

Long-term average collision frequencies between two bodies undergoing Lidov-Kozai evolution

The existing Öpik-type collision frequency framework that accounts for the projectile's Lidov-Kozai (LK) secular evolution holds the target's semimajor axis, eccentricity, and inclination fixed and prescribes uniform precession of its longitude of the ascending node and argument of periapsis. This treatment is not readily applicable when the target's eccentricity, inclination, or orbital precession rates undergo appreciable secular variations. Here, we extend the framework of Liang and Liu (2026, arXiv:2608.30972) to address one such case, in which both the projectile and target undergo LK evolution. Using a doubly averaged quadrupole model to describe the secular dynamics of both bodies, we include the target's orbital evolution when evaluating the collision contribution of each orbital intersection configuration. In the numerical example considered, the projectile survival curve predicted using the calculated collision frequency shows an overall decay trend close to that obtained from direct N-body simulations using REBOUND. In addition, we examine whether the long-term mean collision frequency depends on initial states while holding both bodies' semimajor axes, Kozai constants and closed Hamiltonian level curves fixed. Our theoretical analysis indicates that this frequency is independent of initial states when the parameters describing their relative state are jointly equidistributed at the start of successive projectile LK cycles. Numerical cases with different initial states support this conclusion.
Read the full abstract

Databases

Dynamic Graph Indexing for Label-Filtered Approximate Nearest Neighbor Search

Approximate nearest neighbor search (ANNS) over high-dimensional vectors is a core primitive in modern retrieval systems. In practice, objects are often also associated with labels (e.g., attributes or keywords), and retrieval must combine vector similarity with explicit label-based filtering, giving rise to label-filtered ANNS. This task is challenging because label filtering and ANNS favor very different index structures, selective filters fragment the admissible search space, and updates complicate index maintenance. To address these challenges, we propose DYLAN, a dynamic composite index for label-filtered ANNS supporting containment, overlap, and their conjunctions and disjunctions. DYLAN organizes label sets by shared prefixes in a frequency-ordered trie and selectively shares HNSW-style graph indices across trie-induced subsets, so that each trie region has a constant-factor covering index while each object participates in only logarithmically many indices. To mitigate fragmentation, DYLAN performs a cross-region multi-graph search that exploits ancestor-level indices as bridges between otherwise isolated trie regions. To support insertions, it adopts a threshold-based maintenance scheme that rebuilds shared indices only when necessary, yielding low amortized update cost. Extensive experiments with real data show that DYLAN consistently improves query throughput at comparable recall, by up to two orders of magnitude on challenging label-rich datasets, while retaining competitive memory, construction, and update costs.
Read the full abstract

All authors Mengxu Jiang, Zhi Yang, Fangyuan Zhang, Yin Yang, Sibo Wang

Quantum Physics

Singularising quantum channels preserving kappa-additivity

We construct quantum channels by averaging unitary shift conjugations with respect to $\kappa$-additive probability measures on measurable abelian groups. The channels act on the duals of operator algebras generated by shifts and bounded measurable multiplication operators. They preserve $\kappa$-additivity of quantum states and admit a weak-$*$ integral representation on this class. If the averaging measure vanishes on singletons, every output state is singular. In the appendix, we also establish the corresponding Fubini's theorem for $\kappa$-additive measures.
Read the full abstract

Rings and Algebras

Lie algebras graded by the root system $AG_2$

We study Lie algebras graded by the generalized root system $AG_2$, which is obtained from $G_2$ by adjoining the doubles of its short roots, over fields of characteristic different from $2$ and $3$. We show that every such Lie algebra with a grading subalgebra of type $A_2$ is, modulo its center, the Tits-Kantor-Koecher Lie algebra of a structurable algebra equipped with an idempotent $t$ satisfying $t+\overline{t}=1$. Conversely, every structurable algebra with such an idempotent gives rise to an $(AG_2,A_2)$-graded Lie algebra. We describe the corresponding Peirce decomposition and characterize the additional algebraic structure required for an $(AG_2,G_2)$-grading. Using the classification of central simple structurable algebras, we determine the central simple examples when the characteristic is also different from $5$. Finally, we compare our constructions with the theory of Tits hexagons. For the central simple examples in which all root spaces are non-zero, we determine which ones are Tits-hexagonalizable; in certain cases, this imposes an additional balancing condition on the idempotent.
Read the full abstract

Probability

Exponential mixing for the randomized two-vortex model

This paper investigates the mixing of passive scalars advected by a randomized two-vortex model on $\mathbb{T}^2$, in which the vortices interact through a regularized Biot–Savart kernel and are driven by independent Brownian motions. We establish quenched exponential mixing for the Lagrangian flow, characterized by the exponential correlation decay of mean-zero $H^s$ observables. The proof proceeds by establishing ergodicity of the associated two-point Markov process. We combine a uniform contraction estimate with controllability arguments to verify the Lyapunov drift and minorization conditions, allowing us to apply Harris-type theorem.
Read the full abstract

Theoretical Economics

A Game-Theoretic Framework for Resource Allocation under Institutional Asymmetry

We develop a two-player knapsack game to investigate resource allocation under individual capacity constraints, competition over shared resources, and institutional asymmetry. Players select feasible subsets of resources, while their payoffs depend on competitive losses, allocation rules, and asymmetric institutional benefits and costs. We derive the aggregate welfare function and show that allocation coefficients redistribute payoffs without directly affecting total welfare, whereas institutional parameters influence welfare through their net contribution. Consequently, institutional arrangements with identical welfare effects may induce different strategic incentives and Nash equilibrium outcomes. We analyze how capacity constraints and institutional priority shape equilibrium configurations and their welfare properties. A central result concerns the distinction between the social efficiency and strategic stability of shared resources. We derive conditions under which sharing is socially preferable to exclusive allocation yet remains strategically unstable because the non-priority player can increase their payoff by abandoning the resource. Conversely, sharing may remain strategically stable even when exclusive allocation yields higher aggregate welfare. These findings demonstrate that social efficiency and strategic stability need not coincide under institutional asymmetry.
Read the full abstract

Cryptography and Security

ML Models Are Not Ciphers: Power Analysis for Weight Extraction Is Harder Than You Think

Can power-analysis techniques reveal secret weights of machine-learning models? Multiple recent works demonstrate proof-of-concept attacks that can partially recover such weights. To that end, these works build on techniques from cryptographic engineering, treating the model as a “cipher” and its secret weights as keys. We question the basic assumption behind such straightforward adaptations of cryptographic attacks and highlight the challenges arising when applying past techniques to fully recover all layers of deep neural networks. Unlike past works, which extract only a few weights, typically from shallow layers, we investigate the recovery of complete networks. Moreover, we target fully trained and functional models, rather than untrained, randomly initialized networks. To focus on fundamental challenges, we assume an unrealistically strong side-channel adversary that can directly observe the output of the leakage model without any added noise. With this adversary, we show that machine-learning models differ significantly from cryptographic implementations, giving rise to unique challenges that have not been observed in the prior art. We observe two main differences between machine-learning models and ciphers. First, in contrast to the high-diffusion nonlinear integer arithmetic prevalent in cryptographic schemes, machine-learning models tend to use limited-precision floating-point arithmetic and low-diffusion linear operations. We show that these operations introduce difficulties in weight recovery, often requiring an unexpectedly large number of traces, even under ideal channel conditions. Second, we observe that, unlike ciphers, which aim to increase the data entropy, machine-learning models aim to reduce it. This reduces data variability as the algorithm progresses, hampering weight recovery in deep layers of the model. (details can be found in the paper)
Read the full abstract

All authors Mohammed Qaiss, Peter Horvath, Michal Shagam, Lejla Batina, Zhuoran Liu, Eyal Ronen, Mahmood Sharif, Yuval Yarom