December 2025 arXiv papers — page 2

Qinan Wang, David O. Jones, Justin D. R. Pierel, Matthew R. Siebert

Upcoming cosmological surveys will obtain numerous rest-frame ultraviolet (UV) observations of Type Ia supernovae (SNe Ia), yet there is concern about how standardizable SNe Ia are in the UV. In this work, we train a robust optical--UV SED model for SNe Ia (SALT3-UV) with the open-source model-training software $\texttt{SALTshaker}$. We incorporate a spectro

Diji Yang, Yi Zhang

Despite their scale and success, modern transformers are usually trained as single-minded systems: optimization produces a deterministic set of parameters, representing a single functional hypothesis about the data. Motivated by the analogy to human populations, in which population-level intelligence emerges from diverse individual behaviors, we propose Popu

Yotam Shomroni

We develop a theory of polymatroids on Stallings core graphs, which provides a new technique for proving lower bounds on stable invariants of words and subgroups in free groups $F$, and for upper bounds on their probability for mapping, under a random homomorphism from $F$ to a finite group $G$, into some subgroup of $G$. As a result, we prove the gap conjec

Yidong Chen, Nima Lashkari, Kwing Lam Leung

We write down the global Hamiltonian of conformal field theory (CFT) in finite volume in terms of the modular Hamiltonian of the vacuum reduced to a local ball-shaped region, and use it to propose local approximations to the global Minkowski Hamiltonian in quantum field theory (QFT). The proposed Hamiltonians are motivated by the operator-algebraic property

Gabriela Moisescu-Pareja, Gavin McCracken, Harley Wiltzer, Vincent Létourneau

The Clock and Pizza interpretations, associated with architectures differing in either uniform or learnable attention, were introduced to argue that different architectural designs can yield distinct circuits for modular addition. In this work, we show that this is not the case, and that both uniform attention and trainable attention architectures implement

Wei Wang, Nengneng Yu, Sixian Xiong, Zaoxing Liu

Modern ML training and inference now span tens to tens of thousands of GPUs, where network faults can waste 10--15\% of GPU hours due to slow recovery. Common network errors and link fluctuations trigger timeouts that often terminate entire jobs, forcing expensive checkpoint rollback during training and request reprocessing during inference. We present R$^2$

Laura Casabella, Alessio Sammartano

An orthogonal n-frame is an ordered set of n pairwise orthogonal vectors. The set of all orthogonal n-frames in a d-dimensional quadratic vector space is an algebraic variety V(d,n). In this paper, we investigate the variety V(d,n) as well as the quadratic ideal I(d,n) generated by the orthogonality relations, which cuts out V(d,n). We classify the irreducib

Eren Volkan Küçük

We formalise the self-referential definition of physical laws using monotone operators on a lattice of theories, resolving the pathologies of naive set-theoretic formulations. By invoking Tarski fixed point theorem, we identify physical theories as the least fixed points of admissibility constraints derived from Galois connections. We demonstrate that QED an

Tianzhi He, Farrokh Jazizadeh

This study presents a conceptual framework and a prototype assessment for Large Language Model (LLM)-based Building Energy Management System (BEMS) AI agents to facilitate context-aware energy management in smart buildings through natural language interaction. The proposed framework comprises three modules: perception (sensing), central control (brain), and

Yin Tang, Cristian Voinea, Liangdong Hu, Zlatko Papić

Supersymmetric conformal field theories (SCFTs) form a unique subset of quantum field theories which provide powerful insights into strongly coupled critical phenomena. Here, we present a microscopic and non-perturbative realization of the three-dimensional $\mathcal{N}=1$ superconformal Ising critical point, based on a Yukawa-type coupling between a 3D Isin

Chao Peng, Bin Wang, Zhilei Long, Jinfang Sheng

Retrieval-augmented generation (RAG) is highly sensitive to the quality of selected context, yet standard top-k retrieval often returns redundant or near-duplicate chunks that waste token budget and degrade downstream generation. We present AdaGReS, a redundancy-aware context selection framework for token-budgeted RAG that optimizes a set-level objective com

Giulio Bonelli, Anton Shchechkin, Alessandro Tanzini

We derive bilinear tau forms of the canonically quantized Painlevé equations, thereby relating them to those previously obtained from the $\mathbb{C}^2/\mathbb{Z}_2$ blowup relations for the $\mathcal{N}=2$ supersymmetric gauge theory partition functions on a general $Ω$-background. We fully fix the refined Painlevé/gauge theory dictionary by formulating the

Richard H. Bamler, Yi Lai

We introduce a new approach for analyzing ancient solutions and singularities of mean curvature flow that are locally modeled on a cylinder. Its key ingredient is a general mechanism, called the \emph{PDE--ODI principle}, which converts a broad class of parabolic differential equations into systems of ordinary differential inequalities. This principle bypass

I Komang Januariyasa, Francesco Reda, Nikolai Liubimtsev, Marina Saphiannikova

Microstructured surfaces are central to photonics, biointerfaces, and coating technologies, but they are typically fabricated through multistep workflows involving masks, molds, and postprocessing. Azopolymers offer a direct light-driven route to surface structuring, yet holographic photopatterning of flat films has mainly remain limited to smooth, shallow,

Scott Melville

This talk gives a short introduction to the ``UV/EFT correspondence", which uses scattering amplitudes to relate the Effective Field Theory (EFT) coefficients probed by low-energy measurements to properties of the underlying high-energy (UV) completion. This includes recent ``positivity bounds" on EFT coefficients, which are the low-energy signatures of caus

Mario Ferraro, Bertrand Kibler, Pierre Béjot, Frédéric Gérome

This paper reviews the field of extreme nonlinear optics in optical fibers, highlighting key phenomena and advancements. It discusses multiple ionization effects caused by femtosecond laser pulses that generate plasma and induce permanent material modifications, as well as plasma luminescence and its dependence on material imperfections. The formation and dy

Christopher M. Hans, Ningyi Liu

Many regularization priors for Bayesian regression assume the regression coefficients are a priori independent. In particular this is the case for standard Bayesian treatments of the lasso and the elastic net. While independence may be reasonable in some data-analytic settings, incorporating dependence in these prior distributions provides greater modeling f

Alice Moayyedi

An $\alpha$-thin tree $T$ of a graph $G$ is a spanning tree such that every cut of $G$ has at most an $\alpha$ proportion of its edges in $T$. The Thin Tree Conjecture proposes that there exists a function $f$ such that for any $\alpha > 0$, every $f(\alpha)$-edge-connected graph has an $\alpha$-thin tree. Aside from its independent interest, an algorithm wh

Yan Chen, Lihua Lei

Many applications involve estimating the mean of multiple binomial outcomes as a common problem -- assessing intergenerational mobility of census tracts, estimating prevalence of infectious diseases across countries, and measuring click-through rates for different demographic groups. The most standard approach is to report the plain average of each outcome.

Nisrine Charaf, Faouzi Triki

In this paper, we study the observability of compactly perturbed infinite dimensional systems. Assuming that a given infinite-dimensional system with self-adjoint generator is exactly observable we derive sufficient conditions on a compact self adjoint perturbation to guarantee that the perturbed system stays exactly observable. The analysis is based on a ca

Rina Anno, Vladimir Baranovsky, Timothy Logvinenko

We prove an orbifold type decomposition theorem for the Hochschild homology of the symmetric powers of a small DG category $\mathcal{A}$. In noncommutative geometry, these can be viewed as the noncommutative symmetric quotient stacks of $\mathcal{A}$. We use this decomposition to show that the total Hochschild homology of the symmetric powers of $\mathcal{A}

Zheyan Wan, Juven Wang

Discrete finite-group global symmetries may suffer from nonperturbative 't-Hooft anomalies. Such global anomalies can be canceled by anomalous symmetry-preserving topological quantum field theories (TQFTs), which contain no local point operators but only extended excitations such as line and surface operators. In this work, we study mixed gauge-gravitational

S. V. Borodachov, P. G. Boyvalenkov, P. D. Dragnev, D. P. Hardin

We identify universal polar dual pairs of spherical codes $C$ and $D$ such that for a large class of potential functions $h$ the minima of the discrete $h$-potential of $C$ on the sphere occur at the points of $D$ and vice versa. Moreover, the minimal values of their normalized potentials are equal. These codes arise from the known sharp codes embedded in th

Yiming Wang, Shouvik Sur, Chia-Chuan Liu, Qimiao Si

Strange metals represent a foundational problem in quantum condensed matter physics, and heavy fermion systems provide a canonical setting to advance a general understanding. The concept of a Kondo destruction quantum critical point is widely invoked to describe the competition of the Kondo effect and the local-moment magnetism. Here, we develop a unified fi

K. S. Babu, P. S. Bhupal Dev, Anil Thapa

We develop a general effective field theory (EFT) framework for neutrino-dark matter (DM) interactions, and apply it to systematically find all possible gauge-invariant ultraviolet (UV) completions at a given EFT operator dimension. Our goal here is to find simple UV-complete models that can realize potentially large neutrino-DM interactions, while being con

Yabo Chen, Yuanzhi Liang, Jiepeng Wang, Tingxi Chen

World models aim to endow AI systems with the ability to represent, generate, and interact with dynamic environments in a coherent and temporally consistent manner. While recent video generation models have demonstrated impressive visual quality, they remain limited in real-time interaction, long-horizon consistency, and persistent memory of dynamic scenes,

Alexander C. Li, Ananya Kumar, Deepak Pathak

Discriminative approaches to classification often learn shortcuts that hold in-distribution but fail even under minor distribution shift. This failure mode stems from an overreliance on features that are spuriously correlated with the label. We show that generative classifiers, which use class-conditional generative models, can avoid this issue by modeling a

J. L. Iguain, F. O. Sanchez-Varreti, M. A. Frechero

We present a systematic characterization of the fractal conduction pathways governing ionic transport in a non-crystalline solid below the glass-transition temperature. Using classical molecular dynamics simulations of lithium metasilicate, we combine mobility-resolved dynamical analysis with a real-space description of the regions explored by lithium ions.

Manisha Dhillon, Kuldeep Kumar Kataria, Shyan Ghosh

In this paper, we study a multivariate gamma subordinator whose components are independent gamma processes subject to a random time governed by an independent negative binomial process. We derive the explicit expressions for its joint Laplace-Stieltjes transform, its probability density function and the associated governing differential equations. Also, we s

Guillaume Pignon-Ywanne

In this article, we introduce a new class of smooth partially proper rigid analytic varieties over a $p$-adic field that satisfy Poincar\'e duality for \'etale cohomology with mod $p$-coefficients : the varieties satisfying "primitive comparison with compact support". We show that almost proper varieties, as well as p-adic (weakly admissible) period domains

Omar Said, Christopher Tossas-Betancourt, Mary K. Olive, Jimmy C. Lu

Pulmonary arterial hypertension (PAH) is a progressive cardiopulmonary disease that leads to increased pulmonary pressures, vascular remodeling, and eventual right ventricular (RV) failure. Pediatric PAH remains understudied due to limited data and the lack of targeted diagnostic and therapeutic strategies. In this study, we developed and calibrated multi-sc

Nasim Borazjanizadeh, James McClelland

Transformer language models can generate strikingly natural text by modeling language as a sequence of tokens, but by relying primarily on surface-level co-occurrence statistics they fail to form globally consistent latent representations of entities and events, which contributes to poor relational generalization (the reversal curse), contextualization error

Elynn Chen, Yuefeng Han, Jiayu Li, Ke Xu

We introduce a Modewise Additive Factor Model (MAFM) for matrix-valued time series that captures row-specific and column-specific latent effects through an additive structure, offering greater flexibility than multiplicative frameworks such as Tucker and CP factor models. In MAFM, each observation decomposes into a row-factor component, a column-factor compo

Nicholas Agia

It is often assumed that the most general transformation between two inertial reference frames is affine linear in their Cartesian coordinates, an assumption which is however not true. We provide a complete derivation of the most general inertial frame transformation, which is indeed nonlinear; along the way, we shall find that the conditions of preserving t

Timo Kaufmann, Yannick Metz, Daniel Keim, Eyke Hüllermeier

Binary choices, as often used for reinforcement learning from human feedback (RLHF), convey only the direction of a preference. A person may choose apples over oranges and bananas over grapes, but which preference is stronger? Strength is crucial for decision-making under uncertainty and generalization of preference models, but hard to measure reliably. Meta

Johanna Düntsch, Felix Günther

This paper develops a discrete theory of real Riemann surfaces based on quadrilateral cellular decompositions (quad-graphs) and a linear discretization of the Cauchy-Riemann equations. We construct a discrete analogue of an antiholomorphic involution and classify the topological types of discrete real Riemann surfaces, recovering the classical results on the

Gregory Bowers, Eve Chase, William Ford, Daniel Coupland

A Detector Response Matrix (DRM) is a discrete representation of an instrument's Detector Response Function (DRF), which quantifies how many discrete energy depositions occur in a detector volume for a given distribution of particles incident on the detector. For simple radiation detectors that can count such energy depositions (such as scintillators, Propor

Danny Hermelin, Danny Segev, Dvir Shabtay

We consider a recently introduced fair repetitive scheduling problem involving a set of clients, each asking for their associated job to be daily scheduled on a single machine across a finite planning horizon. The goal is to determine a job processing permutation for each day, aiming to minimize the maximum total completion time experienced by any client. Th

Emmanuel Gnandi, Fortuné Massamba

This paper investigates timelike conformal vector fields on closed Lorentzian $3$-manifolds and shows that, although these fields form a broader class than Killing fields, their behavior in dimension three is nonetheless remarkably rigid. After performing a conformal change of the metric so that the vector field becomes unit and Killing, we analyze the geome

Lacy M. Greening, Santanu S. Dey, Alan L. Erera

Multicommodity capacitated network design (MCND) models can be used to optimize the consolidation of shipments within e-commerce fulfillment networks. In practice, fulfillment networks require that shipments with the same origin and destination follow the same transfer path. This unsplittable flow requirement complicates the MCND problem, requiring integer p

Chenguang Liu, Antonis Papapantoleon, Jasper Rou

The aim of this article is to provide a firm mathematical foundation for the application of deep gradient flow methods (DGFMs) for the solution of (high-dimensional) partial differential equations (PDEs). We decompose the generalization error of DGFMs into an approximation and a training error. We first show that the solution of PDEs that satisfy reasonable

Gabriel Siqueira, Alexsandro Oliveira Alexandrino, Zanoni Dias

Genome rearrangement distances are an established method in genome comparison. Works in this area may include various rearrangement operations representing large-scale mutations, gene orientation information, the number of nucleotides in intergenic regions, and weights reflecting the expected frequency of each operation. In this article, we model genomes con

Siddhant Agarwal, Adya Dhuler, Polly Ruhnke, Melvin Speisman

Over the past years, memes have evolved from being exclusively a medium of humorous exchanges to one that allows users to express a range of emotions freely and easily. With the ever-growing utilization of memes in expressing depressive sentiments, we conduct a study on identifying depressive symptoms exhibited by memes shared by users of online social media

Haozhe Jiang, Nika Haghtalab, Lijie Chen

Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive models for faster inference via parallel token generation. We provide a rigorous foundation for this advantage by formalizing a model of parallel sampling and showing that DLMs augmented with polynomial-length chain-of-thought (CoT) can simulate any parallel sampling a

Xianghong Chen

In this note, we present a characterization of semistable unitary operators on $L^2(\mathbb{R})$, under the assumption that the operator is (i) translation-invariant, (ii) symmetric, and (iii) locally uniformly continuous (LUC) under dilation. As a consequence, we characterize one-parameter groups formed by such operators, which are of the form $e^{i\beta t|

Nilima Nigam

How do the geometric properties of a domain impact the spectrum of an operator defined on it? How do we compute accurate and reliable approximations of these spectra? The former question is studied in spectral geometry, and the latter is a central concern in numerical analysis. In this short expository survey we revisit the process of eigenvalue approximatio

Wee Liang Gan, Khoa Ta

Fix a finite field $\mathbb{F}$. Let $\mathrm{VI}$ be a skeleton of the category of finite dimensional $\mathbb{F}$-vector spaces and injective $\mathbb{F}$-linear maps. We study $\mathrm{VI}^m$-modules over a noetherian commutative ring in the nondescribing characteristic case. We prove that if a finitely generated $\mathrm{VI}^m$-module is generated in deg

Sajad Salami

The splitting field of an elliptic surface $\mathcal{E}/\mathbb{Q}(t)$ is the smallest finite extension $\mathcal{K} \subset \mathbb{C}$ such that all $\mathbb{C}(t)$-rational points are defined over $\mathcal{K}(t)$. In this paper, we provide a symbolic algorithmic approach to determine the splitting field and a set of $68$ linearly independent generators f

Yuchen Wu, Jiahe Li, Fabio Tosi, Matteo Poggi

We present FoundationSLAM, a learning-based monocular dense SLAM system that addresses the absence of geometric consistency in previous flow-based approaches for accurate and robust tracking and mapping. Our core idea is to bridge flow estimation with geometric reasoning by leveraging the guidance from foundation depth models. To this end, we first develop a

Will Barker

In the study of alternative or extended theories of gravity, Dirac's Hamiltonian constraint algorithm is invaluable for enumerating the propagating modes and gauge symmetries. For gravity, this canonical approach is frequently applied as a means for finding pathologies such as strongly coupled modes; more generally it facilitates the reconstruction of gauge

Jungeun Park, Douglas Rizzolo

We study fixed point biased involutions that avoid a pattern. For every pattern of length three we obtain limit theorems for the asymptotic distribution of the (appropriately centered and scaled) number of fixed points of a random fixed point biased involution avoiding that pattern. When the pattern being avoided is either $321$, $132$, or $213$, we find a p

Artyom Lisitsyn, Umut Oktem, Melissa Sherman-Bennett, Jaroslav Trnka

On-shell diagrams are gauge invariant quantities which play an important role in the description of scattering amplitudes. Based on the principles of generalized unitarity, they are given by products of elementary three-point amplitudes where the kinematics of internal on-shell legs are determined by cut conditions. In the ${\cal N}=4$ Super Yang-Mills (SYM)

Jonas Stadelmann, Julian Übelher, Mafalda Ramôa, Bharath Sambasivam

The adaptive derivative-assembled problem-tailored variational quantum eigensolver (ADAPT-VQE) provides a promising approach for simulating highly correlated quantum systems on quantum devices, as it strikes a balance between hardware efficiency, trainability, and accuracy. Although ADAPT-VQE avoids many of the shortcomings of other VQEs, it is sometimes hin

Lukas Anzeletti, Oleg Butkovsky, Máté Gerencsér, Alexander Shaposhnikov

We present a versatile framework to study strong existence and uniqueness for stochastic differential equations (SDEs) in Hilbert spaces with irregular drift. We consider an SDE in a separable Hilbert space $H$ \begin{equation*} dX_t= (A X_t + b(X_t))dt +(-A)^{-\gamma/2}dW_t,\quad X_0=x_0 \in H, \end{equation*} where $A$ is a self-adjoint negative definite o

Valeri V. Makarov

Using the censored catalog of 103,169 resolved Gaia DR3 binary stars with accurate astrometric data for each component, a new observable, object-specific parameter is computed for each pair: the projected orbital momentum. This parameter is the product of four functions of physical characteristics: total mass, semimajor axis, eccentricity, and inclination an

Ágnes Kúsz

We prove that the local limit of the weighted spanning trees on any simple connected high degree almost regular sequence of electric networks is the Poisson(1) branching process conditioned to survive forever, by generalizing [NP22] and closing a gap in their proof. We also study the local statistics of the WST's on high degree almost balanced sequences, whi

Zhenyu Cui, Jiahuan Zhou, Yuxin Peng

Lifelong person Re-IDentification (L-ReID) exploits sequentially collected data to continuously train and update a ReID model, focusing on the overall performance of all data. Its main challenge is to avoid the catastrophic forgetting problem of old knowledge while training on new data. Existing L-ReID methods typically re-extract new features for all histor

Seunghoon Paik, Kangjie Zhou, Matus Telgarsky, Ryan J. Tibshirani

We introduce \textit{basic inequalities} for first-order iterative optimization algorithms, forming a simple and versatile framework that connects implicit and explicit regularization. While related inequalities appear in the literature, we isolate and highlight a specific form and develop it as a well-rounded tool for statistical analysis. Let $f$ denote th

Wei Tang, Benoît Tuybens, Jutho Haegeman

The continuous matrix product state (cMPS) ansatz is a promising numerical tool for studying quantum many-body systems in continuous space. Although it provides a clean framework that allows one to directly simulate continuous systems, the optimization of cMPS is known to be a very challenging task, especially in the case of multi-component systems. In this

Luis Adrián Cabrera-Diego

Classifying legal documents is a challenge, besides their specialized vocabulary, sometimes they can be very long. This means that feeding full documents to a Transformers-based models for classification might be impossible, expensive or slow. Thus, we present a legal document classifier based on DeBERTa V3 and a LSTM, that uses as input a collection of 48 r

Jeffrey Bergfalk, Iian B. Smythe

We consider classification problems for manifolds and discrete subgroups of Lie groups from a descriptive set-theoretic point of view. This work is largely foundational in conception and character, recording both a framework for general study and Borel complexity computations for some of the most fundamental classes of manifolds. We show, for example, that f

Bijendra Kumar Vishvakarma, Shubham Kala

We investigate the strong gravitational lensing and shadow properties of the black hole in the context of bumblebee gravity, characterized by a global monopole charge $\kappa\eta^2$ and a Lorentz symmetry breaking parameter $\gamma$. We compute the deflection angles of light passing near the black hole in strong deflection limit, and estimate key lensing obs

Francesco Becattini, Daniele Roselli, Xin-Li Sheng

We present an \emph{ab initio} calculation within quantum statistical field theory and linear response theory, of the dissipative correction to the momentum spectrum of scalar particles emitted at decoupling (freeze-out) from a relativistic fluid assuming the initial state to be in local thermodynamic equilibrium. We obtain an expansion of the Wigner functio

Josué R. León-Torres, Patrick Hendra, Yugant Mukeshbhai Hadiyal, Christopher Spiess

Quantum imaging with undetected light has recently emerged as a technique in which quantum correlations and nonlinear interferometry are combined to decouple illumination and detection paths. This approach has been more recently extended and combined with digital phase-shifting holography and off-axis holography to extract both the amplitude and phase inform

Yi-Han Yu, Xin-Yi Li, Kai Xu, Heng Fan

A common challenge in superconducting quantum circuits is the trade-off between strong coupling and computational subspace integrity. We present Mathieu control, which uses a non-resonant two-photon drive to create a selective nonlinear frequency shift. This shift modifies interactions while preserving qubit states, enabling continuous tuning of the ZZ coupl

Gyung Hyun Je, Colin Raffel

While large language models (LLMs) demonstrate reasonable zero-shot capability across many downstream tasks, fine-tuning is a common practice to improve their performance. However, a task's data efficiency--i.e., the number of fine-tuning examples needed to achieve a desired level of performance--is often unknown, resulting in costly cycles of incremental an

Lee A. Burton

Materials in which electrons occupy interstitial sites as anions are called electrides and exhibit unusual dimensionality-dependent electronic behavior. These properties make electrides attractive for catalysis, transparent conductors, and emergent quantum phenomena, yet their theoretical treatment remains challenging. In conventional materials, the ground-s

Yegor Zenkevich

We investigate BPS states in 4d N=4 supersymmetric Yang-Mills theory and the corresponding (p, q) string networks in Type IIB string theory. We propose a new interpretation of the algebra of line operators in this theory as a tensor product of vector representations of a quantum toroidal algebra, which determines protected spin characters of all framed BPS s

Justin McInroy, Abdul Wajid Mir

Axial algebras of Monster type are a class of non-associative algebras which generalise the Griess algebra, whose automorphism group is the largest sporadic simple group, the Monster. The $2$-generated algebras, which are the building blocks from which all algebras in this class can be constructed, have recently been classified by Yabe; Franchi and Mainardis

Luca Collorone, Mert Kiray, Indro Spinelli, Fabio Galasso

Realistic visual simulations are omnipresent, yet their creation requires computing time, rendering, and expert animation knowledge. Open-vocabulary visual effects generation from text inputs emerges as a promising solution that can unlock immense creative potential. However, current pipelines lack both physical realism and effective language interfaces, req

Yohan Park, Hyunwoo Ha, Wonjun Jo, Tae-Hyun Oh

Vision Language Models (VLMs) are increasingly adopted as central reasoning modules for embodied agents. Existing benchmarks evaluate their capabilities under ideal, well-lit conditions, yet robust 24/7 operation demands performance under a wide range of visual degradations, including low-light conditions at night or in dark environments, a core necessity th

Jacob L. Bourjaily

At tree-level, scattering amplitudes involving only gluons or gravitons are unaffected by supersymmetry, allowing them to be efficiently encoded by and extracted from those of maximally supersymmetric (N=4,8) theories. This fails beyond tree-level, of course, but much less than would be expected. We show that all the leading singularities of (sub-maximally o

Bahadır Utku Kesgin, Gülsüm Yaren Durdu, Uğur Teğin

Optical computing could reduce the energy cost of artificial intelligence by leveraging the parallelism and propagation speed of light. However, implementing nonlinear activation, essential for machine learning, remains challenging in low-power optical systems dominated by linear wave physics. Here, we introduce an optical spiking neural network that uses op

Isaac H. Kim, Tuomas Laakkonen

Since an n-qubit circuit consisting of CNOT gates can have up to $\Omega(n^2/\log{n})$ CNOT gates, it is natural to expect that $\Omega(n^2/\log{n})$ Toffoli gates are needed to apply a controlled version of such a circuit. We show that the Toffoli count can be reduced to at most n. The Toffoli depth can also be reduced to O(1), at the cost of 2n Toffoli gat

Prosenjit Howlader, Churn-Jung Liau

We introduce a two-sort weighted modal logic for possibilistic reasoning with fuzzy formal contexts. The syntax of the logic includes two types of weighted modal operators corresponding to classical necessity ($\Box$) and sufficiency ($\boxminus$) modalities and its formulas are interpreted in fuzzy formal contexts based on possibility theory. We present its

Adolfo Holguin

We revisit the problem of computing extremal and non-extremal three point functions of semiclassical probes with single trace operators and point out certain inconsistencies in previous approaches in the literature. We clarify the roles of wavefunctions and averaging over moduli, concluding that holographic computations may be performed with or without avera

C. Dudeja, J. Roy, U. Panda, S. Bhattacharyya

We present results from multi-epoch observations of the repeating fast radio burst FRB 20201124A with the upgraded Giant Metrewave Radio Telescope (uGMRT) during its active phase between 8 May and 28 May 2021. The bursts exhibit significant morphological diversity, including multiple sub-bursts, downward frequency drifts, and intrinsic widths ranging from 1.

Yuchao Hou, Zixuan Zhang, Jie Wang, Wenke Huang

As a critical application of computational intelligence in remote sensing, deep learning-based synthetic aperture radar (SAR) image target recognition facilitates intelligent perception but typically relies on centralized training, where multi-source SAR data are uploaded to a single server, raising privacy and security concerns. Federated learning (FL) prov

Barna Zajzon, Younes Bouhadjar, Maxime Fabre, Felix Schmidt

Sequential structure is a key feature of multiple domains of natural cognition and behavior, such as language, movement and decision-making. Likewise, it is also a central property of tasks to which we would like to apply artificial intelligence. It is therefore of great importance to develop frameworks that allow us to evaluate sequence learning and process

Pietro Valigi, M. Ángeles Serrano, Claudio Castellano, Lorenzo Cirigliano

The graphicality problem -- whether or not a sequence of integers can be used to create a simple graph -- is a key question in network theory and combinatorics, with many important practical applications. In this work, we study the graphicality of degree sequences distributed as a power-law with a size-dependent cutoff and as a double power-law with a size-d

Cristina P. Martin-Linares, Jonathan P. Ling

Sparse autoencoders (SAEs) aim to disentangle model activations into monosemantic, human-interpretable features. In practice, learned features are often redundant and vary across training runs and sparsity levels, which makes interpretations difficult to transfer and reuse. We introduce Distilled Matryoshka Sparse Autoencoders (DMSAEs), a training pipeline t

Yunxi Tang, Tianqi Yang, Jing Huang, Xiangyu Chu

Deformable linear objects (DLOs) manipulation presents significant challenges due to DLOs' inherent high-dimensional state space and complex deformation dynamics. The wide-populated obstacles in realistic workspaces further complicate DLO manipulation, necessitating efficient deformation planning and robust deformation tracking. In this work, we propose a no

Rafał Potempa, Michał Kordasz, Józef P. Cyran, Kamil Wereszczyński

This work presents a Python framework named after the General Equation of Quantum Image Encoding (GEQIE). The framework creates the image-encoding state using a unitary gate, which can later be transpiled to target quantum backends. The benchmarking results, simulated with different noise levels, demonstrate the correctness of the already implemented encodin

James M Parks

This is a study of the construction of particular regular sub-n-gons T in regular n-gons P using a special system of chords of P. In particular, some of these sub-n-gons have areas which are integer divisors of the area of the given n-gon P. Initially, the study will concentrate on chords which are from a vertex to special points of one of the opposite sides

Bingyang Hu, Xiaojing Zhou

Motivated by the work of Cheng-Fang-Wang-Yu on the hypersingular Bergman projection, we develop a real-variable framework for hypersingular operators in regimes where strong-type bounds fail on the critical line. Our main new ingredient is the Forelli-Rudin method: a dyadic mechanism, inspired by complex-analytic Forelli-Rudin type arguments, that yields sha

Itallo Patrick Castro Alves Da Silva, Emanuel Adler Medeiros Pereira, Erick de Andrade Barboza, Baldoino Fonseca dos Santos Neto

Compressed deep learning models are crucial for deploying computer vision systems on resource-constrained devices. However, model compression may affect robustness, especially under natural corruption. Therefore, it is important to consider robustness evaluation while validating computer vision systems. This paper presents a comprehensive evaluation of compr

Chen Chen, Jiuyang Liang, Zhenli Xu, Qianru Zhang

The Born-Mayer-Huggins (BMH) potential, which combines Coulomb interactions with dispersion and short-range exponential repulsion, is widely used for ionic materials such as molten salts. However, large-scale molecular dynamics simulations of BMH systems are often limited by computation, communication, and memory costs. We recently proposed the random batch

Colin Scheibner, Lindsay M. Smith, William Bialek

We use large language models (LLMs) to uncover long-ranged structure in English texts from a variety of sources. The conditional entropy or code length in many cases continues to decrease with context length at least to $N\sim 10^4$ characters, implying that there are direct dependencies or interactions across these distances. A corollary is that there are s

Hangcheng Zhao, Ron Berman

Generative AI can adversely impact news publishers by lowering consumer demand. It can also reduce demand for newsroom employees, and increase the creation of news "slop." However, it can also form a source of traffic referrals and an information-discovery channel that increases demand. We use high-frequency granular data to analyze the strategic response of

Yuto Kuwayama, Yongming Liang, Kentaro Nagamine, Yuri Oku

The recently reported Cosmic Himalayas (CH) -- an extreme quasar overdensity at z~2 -- poses an apparent challenge to the Lambda CDM framework, with a reported significance of 16.9-sigma under Gaussian assumptions. Such an event appears improbably rare, with a formal probability of P ~ 10^-68. In this work, we investigate whether CH-like structures can natur

Siyuan Hu, Kevin Qinghong Lin, Mike Zheng Shou

Building intelligent agents capable of dexterous manipulation is essential for achieving human-like automation in both robotics and digital environments. However, existing GUI agents rely on discrete click predictions (x,y), which prohibits free-form, closed-loop trajectories (e.g. dragging a progress bar) that require continuous, on-the-fly perception and a

Alessia andò, Giusy Bosco, Dimitri Breda, Davide Liessi

We consider the problem of discretizing evolution operators of linear delay equations with the aim of approximating their spectra, which is useful in investigating the stability properties of (nonlinear) equations via the principle of linearized stability. We develop a general convergence analysis based on a reformulation of the operators by means of a fixed

Peter J. Cho, Robert J. Lemke Oliver, Asif Zaman

Let $G$ be a finite group. Let $K/k$ be a Galois extension of number fields with Galois group isomorphic to $G$, and let $C \subseteq \mathrm{Gal}(K/k) \simeq G$ be a conjugacy invariant subset. It is well known that there exists an unramified prime ideal $\mathfrak{p}$ of $k$ with Frobenius element lying in $C$ and norm satisfying $\mathrm{N}\mathfrak{p} \l

Tong Wei, Kumar Vijay Mishra, Bhavani Shankar M. R., Björn Ottersten

Near-field sensing with extremely large-scale antenna arrays (ELAAs) in practical 6G systems is expected to operate over broad bandwidths, where delay, Doppler, and spatial effects become tightly coupled across frequency. The purpose of this and the companion paper (Part I) is to develop the unified Cram'er--Rao bounds (CRBs) for sensing systems spanning fro

Marzieh Eidi, Nina Otter

Similarity notions between vertices in a graph, such as structural and regular equivalence, are one of the main ingredients in clustering tools in complex network science. We generalise structural and regular equivalences for undirected hypergraphs and provide a characterisation of structural and regular equivalences of undirected graphs and hypergraphs thro

Kristina Giesel, Hongguang Liu

The absence of Birkhoff's theorem in effective quantum gravity models leads to a fundamental ambiguity in the vacuum sector, where a priori no unique vacuum solution exists. As a result, phenomenological investigations of the physical implications of these models have been made more difficult. We address this challenge by establishing a constructive framewor

András Antos, András Millinghoffer, Péter Antal

Many modern AI and ML problems require evaluating partners' contributions through shared yet asymmetric, computationally intensive processes and the simultaneous selection of the most beneficial candidates. Sequential approaches to these problems can be unified under a new framework, Sequential Support Network Learning (SSNL), in which the goal is to select

Tong Wei, Kumar Vijay Mishra, Bhavani Shankar M. R., Björn Ottersten

Extremely large-scale antenna arrays (ELAAs) envisioned for 6G enable high-resolution sensing. However, the ELAAs worked in extremely high frequency will push operation into the near-field region, where spherical wavefronts invalidate classical far-field models and alter fundamental estimation limits. The purpose of this and the companion paper (Part II) is

AMAP AI Agent Team, Yulan Hu, Xiangwen Zhang, Sheng Ouyang

We present STAgent, an agentic large language model tailored for spatio-temporal understanding, designed to solve complex tasks such as constrained point-of-interest discovery and itinerary planning. STAgent is a specialized model capable of interacting with ten distinct tools within spatio-temporal scenarios, enabling it to explore, verify, and refine inter

Domingos S. P. Salazar

Thermodynamic uncertainty relations (TURs) bound the precision of currents by entropy production, but quantum transport of noncommuting (non-Abelian) charges challenges standard formulations because different charge components cannot be monitored within a single classical frame. We derive a process-level matrix TUR starting from the operational entropy produ