October 2025 arXiv papers — page 149

Sarah Harkins Dayton, Hayden Everett, Ioannis Schizas, David L. Boothe

Convolutional neural networks (CNNs) have been established as the main workhorse in image data processing; nonetheless, they require large amounts of data to train, often produce overconfident predictions, and frequently lack the ability to quantify the uncertainty of their predictions. To address these concerns, we propose a new Bayesian topological CNN tha

Yolanda G. C. Frensch, François Bouchy, Gaspare Lo Curto, Alexandrine L'Heureux

Gas giant planets orbiting low-mass stars are uncommon outcomes of planet formation. Increasing the sample of well-characterised giants around early M dwarfs will enable population-level studies of their properties, offering valuable insights into their formation and evolutionary histories. We aim to characterise giant exoplanets transiting M dwarfs identifi

T. Emil Rivera-Thorsen, Brian Welch, Taylor Hutchison, Matthew J. Hayes

At present, the best opportunity for detailed Lyman Continuum escape studies is in gravitationally lensed galaxies at z >~ 2. Only one such galaxy currently exists in the literature with sufficient spatial magnification: The Sunburst Arc at redshift z = 2.37. Here, we present rest-frame optical JWST NIRSpec integral field observations of the Sunburst Arc tha

Zhaochen Yu, Ling Yang, Jiaru Zou, Shuicheng Yan

Recently, the emergence of agentic RL has showcased that RL could also effectively improve the agentic reasoning ability of LLMs, yet the key design principles and optimal practices remain unclear. In this work, we conduct a comprehensive and systematic investigation to demystify reinforcement learning in agentic reasoning from three key perspectives: data,

Maksat Satybaldiev, Manuel Linares, Vittoria Vecchiotti

Compact binary millisecond pulsars (also known as spiders) allow us to probe pulsar winds in their innermost regions, between the light cylinder (radius $\sim10^{7}$ cm) and the companion star (at $\sim10^{11}$ cm). Their flux is known to vary along the orbit, from radio to X-rays. During the past decade, gamma-ray orbital modulation (GOM) has been discovere

Ruida Wang, Jiarui Yao, Rui Pan, Shizhe Diao

Solving math problems through verifiable languages such as Lean has significantly impacted both the mathematics and computer science communities. Current state-of-the-art models are often trained with expensive online Reinforcement Learning (RL) or expert iteration. However, these approaches rely on fixed problem sets, which causes inefficient training and l

Adrian Beker

Given positive integers $n$ and $m$, let $p_n(m)$ be the probability that a uniform random permutation of $[n]$ has order exactly $m$. We show that, as $n \to \infty$, the maximum of $p_n(m)$ over all $m$ is asymptotic to $1/n$, the probability of an $n$-cycle. Furthermore, for sufficiently large $n$, we show that the maximum is attained precisely if $m$ is

Matteo Mordacchini, Emanuele Carlini, Patrizio Dazzi

We address the problem of computing a Minimum Weighted Vertex Cover (MWVC) in a decentralized network. MWVC, a classical NP-hard problem, is foundational in applications such as network monitoring and resource placement. We propose a fully decentralized protocol where each node makes decisions using only local knowledge and communicates with its neighbors. T

Wei Huang, Yi Ge, Shuai Yang, Yicheng Xiao

We propose QeRL, a Quantization-enhanced Reinforcement Learning framework for large language models (LLMs). While RL is essential for LLMs' reasoning capabilities, it is resource-intensive, requiring substantial GPU memory and long rollout durations. QeRL addresses these issues by combining NVFP4 quantization with Low-Rank Adaptation (LoRA), accelerating rol

Lingfei Qian, Xueqing Peng, Yan Wang, Vincent Jim Zhang

Although Large Language Model (LLM)-based agents are increasingly used in financial trading, it remains unclear whether they can reason and adapt in live markets, as most studies test models instead of agents, cover limited periods and assets, and rely on unverified data. To address these gaps, we introduce Agent Market Arena (AMA), the first lifelong, real-

Arjun Sahney, Ram Gorthi, Cezary Łastowski, Javier Vega

We present Operand Quant, a single-agent, IDE-based architecture for autonomous machine learning engineering (MLE). Operand Quant departs from conventional multi-agent orchestration frameworks by consolidating all MLE lifecycle stages -- exploration, modeling, experimentation, and deployment -- within a single, context-aware agent. On the MLE-Benchmark (2025

Chenghao Xiao, Hou Pong Chan, Hao Zhang, Weiwen Xu

Recent multimodal embedding approaches leveraging multimodal large language models (MLLMs) fine-tuned with contrastive learning (CL) have shown promising results, yet the underlying reasons behind their superiority remain underexplored. This work argues that a crucial advantage of MLLM-based approaches stems from implicit cross-modal alignment achieved durin

Samuel G. Gessow, Brett T. Lopez

We present an analysis on the convergence properties of the so-called geometric heat flow equation for computing geodesics (extremal curves) on Riemannian manifolds. Computing geodesics numerically in real time has become an important capability across several fields, including control and motion planning. The geometric heat flow equation involves solving a

Ayman Shehata, M. Tawfik, Ayman M. Mahmoud, Nada Mostafa

The aim of the present study is to establish some properties for q-Bessel matrix polynomials such as several q-differential matrix equation, q-differential matrix relations and q-recurrence matrix relations, and integral representation, q-Laplace and q-Mellin transforms with the help of q-Analysis. Furthermore, we give connections between q-Horn's matrix fun

Taira Tsuchiya

In two-player zero-sum games, the learning dynamic based on optimistic Hedge achieves one of the best-known regret upper bounds among strongly-uncoupled learning dynamics. With an appropriately chosen learning rate, the social and individual regrets can be bounded by $O(\log(mn))$ in terms of the numbers of actions $m$ and $n$ of the two players. This study

Boyang Zheng, Nanye Ma, Shengbang Tong, Saining Xie

Latent generative modeling, where a pretrained autoencoder maps pixels into a latent space for the diffusion process, has become the standard strategy for Diffusion Transformers (DiT); however, the autoencoder component has barely evolved. Most DiTs continue to rely on the original VAE encoder, which introduces several limitations: outdated backbones that co

Maggie Wang, Stephen Tian, Aiden Swann, Ola Shorinwa

Learning robotic manipulation policies directly in the real world can be expensive and time-consuming. While reinforcement learning (RL) policies trained in simulation present a scalable alternative, effective sim-to-real transfer remains challenging, particularly for tasks that require precise dynamics. To address this, we propose Phys2Real, a real-to-sim-t

Zicheng Liu, Lige Huang, Jie Zhang, Dongrui Liu

The increasing autonomy of Large Language Models (LLMs) necessitates a rigorous evaluation of their potential to aid in cyber offense. Existing benchmarks often lack real-world complexity and are thus unable to accurately assess LLMs' cybersecurity capabilities. To address this gap, we introduce PACEbench, a practical AI cyber-exploitation benchmark built on

Jinyu Zhang, Haitao Lin, Jiashu Hou, Xiangyang Xue

Estimating an object's 6D pose, size, and shape from visual input is a fundamental problem in computer vision, with critical applications in robotic grasping and manipulation. Existing methods either rely on object-specific priors such as CAD models or templates, or suffer from limited generalization across categories due to pose-shape entanglement and multi

Hector Iglesias, Leanne Durkan, Deirdre Shoemaker

In the past few decades, the waveform community has made advances in producing waveforms that span the inspiral-merger-ringdown of comparable-mass-ratio black hole binaries using advances in post-Newtonian and numerical relativity (NR) theory along with state-of-the-art gravitational wave models. Current methods in NR have shown progress towards producing st

C. Y. Tan, W. Cerny, A. B. Pace, J. A. Sharp

We report the discovery of three Milky Way satellite candidates: Carina IV, Phoenix III, and DELVE 7, in the third data release of the DECam Local Volume Exploration survey (DELVE). The candidate systems were identified by cross-matching results from two independent search algorithms. All three are extremely faint systems composed of old, metal-poor stellar

Hang Liu, Yuman Gao, Sangli Teng, Yufeng Chi

Enabling humanoid robots to exploit physical contact, rather than simply avoid collisions, is crucial for autonomy in unstructured environments. Traditional optimization-based planners struggle with contact complexity, while on-policy reinforcement learning (RL) is sample-inefficient and has limited multi-task ability. We propose a framework combining a lear

Jennifer R. DeRosa, Tian Qiu, D. Vale Cofer-Shabica, Joseph E. Subotnik

We investigate the applicability of the Condon approximation (i.e. the notion that the diabatic coupling is invariant to geometry) in the context of both electron transfer (ET) and triplet energy transfer (TET) and compare the two cases. Although it is well appreciated that diabatic couplings usually arise from the interactions of electronic wavefunction tai

Daniel K. Mark, Federica M. Surace, Thomas Schuster, Adam L. Shaw

Gauge theories describe the fundamental forces of nature. However, high-energy dynamics, such as the formation of quark-gluon plasmas, is notoriously difficult to model with classical methods. Quantum simulation offers a promising alternative in this regime, yet experiments have mainly probed low energies. Here, we observe the formation of a ballistic plasma

Songrun He, Linying Lv, Asaf Manela, Jimmy Wu

We introduce a family of chronologically consistent, instruction-tuned large language models to eliminate lookahead bias. Each model is trained only on data available before a clearly defined knowledge-cutoff date, ensuring strict temporal separation from any post-cutoff data. The resulting framework offers (i) a simple, conversational chat interface, (ii) f

Chuan He, Zhaosong Lu

We study convex composite optimization problems, where the objective function is given by the sum of a prox-friendly function and a convex function whose subgradients are estimated under heavy-tailed noise. Existing work often employs gradient clipping or normalization techniques in stochastic first-order methods to address heavy-tailed noise. In this paper,

Dipkamal Bhusal, Michael Clifford, Sara Rampazzi, Nidhi Rastogi

Interpreting deep neural networks through concept-based explanations offers a bridge between low-level features and high-level human-understandable semantics. However, existing automatic concept discovery methods often fail to align these extracted concepts with the model's true decision-making process, thereby compromising explanation faithfulness. In this

Stephon Alexander, Bruno Alexandre, Michael Fine, João Magueijo

We propose a graviGUT unification scheme based on the simple orthogonal group $\mathrm{SO}(1,9 ,\mathbb{C})$ that resolves the chiral duplication of weak isospin in Pati--Salam models. In the conventional $SU(4)\times SU(2)_+\times SU(2)_-$ framework, the unobserved second chiral $SU(2)$ is typically removed by ad hoc high-energy scale breaking. Here we inst

Nihar Gargava, Vlad Serban, Maryna Viazovska, Ilaria Viglino

We prove an asymptotic formula for the number of fixed rank matrices with integer coefficients over a number field K/Q and bounded norm. As an application, we derive an approximate Rogers integral formula for discrete sets of module lattices obtained from lifts of algebraic codes. This in turn implies that the moment estimates of random lattices with a numbe

Yaroslav Kopylov, Vadim Leshkov

We consider Lambek's invariants $\text{Ker}$ and $\text{Im}$ for commutative squares in Grandis homological categories. We prove that Nomura's null sequences exist in such categories and find sufficienct conditions for their exactness. We also prove the coincidence of left and right homology in Grandis homological categories.

Sunilkumar V Rajat, Sandeep Gautam, Arko Roy

We investigate the ferromagnetic-paramagnetic phase transition in coherently (Rabi) coupled Bose-Einstein condensates at zero and finite temperatures, exploring different routes to the transition by tuning the Rabi coupling or increasing the temperature at a fixed coupling. Using the Hartree-Fock-Bogoliubov theory within the Popov approximation, we map out t

Ludovica Maga, Mathias Peirlinck, Lise Noël

Hearts subjected to volume overload (VO) are prone to detrimental anatomical and functional changes in response to elevated mechanical loading, ultimately leading to heart failure. Experimental findings now emphasize that organ-scale changes following VO cannot be explained by myocyte growth alone, as traditionally proposed in the literature. Collagen degrad

Marcelo Lopes Ferro, Armando M. V. Corro

In this work, we provide a local classification of certain special classes of surfaces determined by the prescription of the radial mean curvature in terms of the height and angle functions. Moreover, we introduce a special class of hypersurfaces, and we also provide a local classification of these three-dimensional hypersurfaces whose second mean curvature

Takayuki Hibi, Seyed Amin Seyed Fakhari

Let $S=K[x_1, \ldots,x_n]$ denote the polynomial ring in $n$ variables over a field $K$ and $I(G) \subset S$ the edge ideal of a finite graph $G$ on $n$ vertices. Given a vector $\mathfrak{c}\in\mathbb{N}^n$ and an integer $q\geq 1$, we denote by $(I(G)^q)_{\mathfrak{c}}$ the ideal of $S$ generated by those monomials belonging to $I(G)^q$ whose exponent vect

Noah L. Estrada, Marc A. Cantu, Arturo Rodriguez, Andrew R. Ybarra

The performance of convergent and divergent nozzles is critical in aerospace propulsion systems, where the efficient expansion of high-temperature, high-pressure gases directly impacts thrust generation. In this study, we investigate a series of nozzle geometries using numerical simulations in ANSYS Fluent, guided by classical compressible flow theory, initi

Natalie Lang, Atsutse K. Kludze, Nir Shlezinger, Yasaman Ghasempour

The ever-growing demand for ultra-high data rates, massive connectivity, and joint communication-sensing capabilities in future wireless networks is driving research into sub-terahertz (sub-THz) communications. While these frequency bands offer abundant spectrum, they also pose severe propagation and hardware design challenges, motivating the search for alte

Elliot Fox, Aude Gehrmann-De Ridder, Thomas Gehrmann, Nigel Glover

We present a calculation of the thrust distribution in Higgs decays to quarks and gluons, $H\to b\bar{b}$, $H\to c\bar{c}$, and $H\to gg$, including the resummation of large logarithmic corrections that arise in the two-particle limit at next-to-next-to-leading logarithmic (NNLL) accuracy, and match it to fixed-order results for three-particle decays at next

Caitlin Callaghan, David J Reinkensmeyer

The accuracy with which the human proprioceptive system estimates hand speed is not well understood. To investigate this, we designed an experiment using hobby-grade mechatronics parts and integrated it as a laboratory exercise in a large remote laboratory course. In a simple joint position reproduction task, participants (N = 191) grasped a servomotor-drive

Steven B. Damelin, Ruiwen Shu

In this paper, we consider the one-dimensional interaction energy $\frac{1}{2}\int_{\mathbb{R}}(W*\rho)(x)d\rho(x) + \int_{\mathbb{R}}U(x)d\rho(x)$ where the interaction potential $W(x)= -\frac{|x|^b}{b},\,1\le b \le 2$ and the external potential $U(x)=\frac{|x|^4}{4}$, and $\rho$ is a compactly supported probability measure on the real line. Our main result

Shijie Xia, Yuhan Sun, Pengfei Liu

Recently, Large Language Models (LLMs) have been applied to scientific equation discovery, leveraging their embedded scientific knowledge for hypothesis generation. However, current methods typically confine LLMs to the role of an equation proposer within search algorithms like genetic programming. In this paper, we present SR-Scientist, a framework that ele

Yi Yang, Kefan Gu, Yuqing Wen, Hebei Li

While Vision-Language-Action (VLA) models have demonstrated impressive capabilities in robotic manipulation, their performance in complex reasoning and long-horizon task planning is limited by data scarcity and model capacity. To address this, we introduce ManiAgent, an agentic architecture for general manipulation tasks that achieves end-to-end output from

Florian Obermüller, Gordon Fraser

When learning to program, students are usually assessed based on the code they wrote. However, the mere completion of a programming task does not guarantee actual comprehension of the underlying concepts. Asking learners questions about the code they wrote has therefore been proposed as a means to assess program comprehension. As creating targeted questions

Panos Tsimpos, Youssef Marzouk

We study dynamic measure transport for generative modeling: specifically, flows induced by stochastic processes that bridge a specified source and target distribution. The conditional expectation of the process' velocity defines an ODE whose flow map achieves the desired transport. We ask \emph{which processes produce straight-line flows} -- i.e., flows whos

Guo Ye

Two body tunneling problems are hard to treat analytically due to the incompatibility between tunneling and perturbation theory. The lack of classical solutions of the Euclidean Lagrangian of continuous systems further thwarts semi-classical expansions. To develop an analytic theory which provides insight on interacting two-particle tunneling, we use new res

Mark L. Lewis

In this paper, we prove a property of kernels of Brauer characters. We propose a candidate for the kernels of Isaacs' partial characters, and we show that this candidate has the same property.

Daniel Berhane Araya, Duoduo Liao

Financial markets face growing threats from misinformation that can trigger billions in losses in minutes. Most existing approaches lack transparency in their decision-making and provide limited attribution to credible sources. We introduce FinVet, a novel multi-agent framework that integrates two Retrieval-Augmented Generation (RAG) pipelines with external

Prasanna Mayilvahanan, Ricardo Dominguez-Olmedo, Thaddäus Wiedemer, Wieland Brendel

With the advent of DeepSeek-R1, a new wave of reinforcement learning (RL) methods has emerged that seem to unlock stronger mathematical reasoning. However, a closer look at the open-source ecosystem reveals a critical limitation: with sufficiently many draws (e.g., $\texttt{pass@1024}$), many existing base models already solve nearly all questions on widely

Xin Gui, King Zhu, JinCheng Ren, Qianben Chen

In recent years, the research focus of large language models (LLMs) and agents has shifted increasingly from demonstrating novel capabilities to complex reasoning and tackling challenging tasks. However, existing evaluations focus mainly on math/code contests or general tasks, while existing multi-domain academic benchmarks lack sufficient reasoning depth, l

Federica Bertolotti, Ervin Hadziosmanovic

In this paper, we investigate the asymptotic behavior of the integral simplicial volume of cyclic covers of manifolds that fiber over the circle with fiber given by an $n$-dimensional torus. By studying the integral filling volume -- an invariant introduced by Frigerio and the first author -- for the monodromy, we establish both lower and upper bounds for th

Yuxuan Xue, Xianghui Xie, Margaret Kostyrko, Gerard Pons-Moll

Generating realistic and controllable 3D human avatars is a long-standing challenge, particularly when covering broad attribute ranges such as ethnicity, age, clothing styles, and detailed body shapes. Capturing and annotating large-scale human datasets for training generative models is prohibitively expensive and limited in scale and diversity. The central

Pradyumna Yalandur Muralidhar, Yuxuan Xue, Xianghui Xie, Margaret Kostyrko

Reconstructing metrically accurate humans and their surrounding scenes from a single image is crucial for virtual reality, robotics, and comprehensive 3D scene understanding. However, existing methods struggle with depth ambiguity, occlusions, and physically inconsistent contacts. To address these challenges, we introduce PhySIC, a framework for physically p

Ahmad Z. Fino, Berikbol T. Torebek

This paper investigates the critical behavior of global solutions to a parabolic equation with a Hartree-type nonlinearity of the form $$\left\{\begin{array}{ll} u_{t}+(-\Delta)^{\frac{\beta}{2}} u= (\mathcal{K}\ast |u|^{p})|u|^{q},&\qquad x\in \mathbb{R}^n,\,\,\,t>0, u(x,0)=u_{0}(x),& \qquad x\in \mathbb{R}^n,\end{array} \right.$$ where $\beta\in(0,2]$, $n\

Yinan Chen, Jiangning Zhang, Teng Hu, Yuxiang Zeng

Instruction-guided video editing has emerged as a rapidly advancing research direction, offering new opportunities for intuitive content transformation while also posing significant challenges for systematic evaluation. Existing video editing benchmarks fail to support the evaluation of instruction-guided video editing adequately and further suffer from limi

Jingyuan Xing, Mingru Yang, Zhipeng Li, Xiaofen Xing

Autoregressive (AR) frameworks have recently achieved remarkable progress in zero-shot text-to-speech (TTS) by leveraging discrete speech tokens and large language model techniques. Despite their success, existing AR-based zero-shot TTS systems face two critical limitations: (i) an inherent speed-quality trade-off, as sequential token generation either reduc

Jacob B. Fiedler

We prove a new lower bound on the algorithmic information content of points lying on a line in $\mathbb{R}^n$. More precisely, we show that a typical point $z$ on any line $\ell$ satisfies \begin{equation*} K_r(z)\geq \frac{K_r(\ell)}{2} + r - o(r) \end{equation*} at every precision $r$. In other words, a randomly chosen point on a line has (at least) half o

E. Abasov, L. Dudko, E. Iudin, A. Markina

We apply a unified machine-learning framework based on Normalizing Flows (NFs) for the event-by-event reconstruction of invisible momenta and the subsequent evaluation of spin-sensitive observables in top-quark pair and dark-matter (DM) associated production processes. Building on recent studies in single-top + DM topologies, we extend the research to $t\bar

Alexey Dubinsky

We investigate the propagation and absorption of Standard Model fields -- scalar, electromagnetic, and Dirac -- on a 3+1-dimensional brane embedded in a higher-dimensional Schwarzschild--de Sitter (SdS) spacetime. Using the effective four-dimensional projection of the Tangherlini metric, we compute grey-body factors (GBFs) and absorption cross-sections for e

Egun Im, Ghina Badih, Laetitia Kurzawa, Andreas Buttenschön

Left-right axis specification establishes embryonic laterality through asymmetric signaling cascades originating at the cellular scale. We previously reported the presence of a directionality bias in confined pairs of endothelial (and fibroblast) cells exhibiting persistent circular motion, with cytoskeletal contractility modulating the direction. The relati

T. M. Kowalewski, A. D. Ayangeakaa, N. Sensharma, R. V. F. Janssens

The electromagnetic properties of low-lying states in $^{70}$Ge were investigated via multi-step Coulomb excitation of a $^{70}$Ge beam impinging on a $^{208}$Pb target at the ATLAS facility of the Argonne National Laboratory. A total of 27 transitional elements and six diagonal matrix elements coupling 11 low-lying states, were extracted from the measured c

Felix Zhou

We study the sublinear space continual release model for edge-differentially private (DP) graph algorithms, with a focus on the densest subgraph problem (DSG) in the insertion-only setting. Our main result is the first continual release DSG algorithm that matches the additive error of the best static DP algorithms and the space complexity of the best non-pri

Zhanyu Liu, Shiyao Wang, Xingmei Wang, Rongzhou Zhang

The powerful generative capacity of Large Language Models (LLMs) has instigated a paradigm shift in recommendation. However, existing generative models (e.g., OneRec) operate as implicit predictors, critically lacking the capacity for explicit and controllable reasoning-a key advantage of LLMs. To bridge this gap, we propose OneRec-Think, a unified framework

Yijia Fang, Gennian Ge, Yang Shu, Qian Xu

In a seminal work, Cheng and Xu showed that if $S$ is a square or a triangle with a certain property, then for every positive integer $r$ there exists $n_0(S)$ independent of $r$ such that every $r$-coloring of $\mathbb{E}^n$ with $n\ge n_0(S)$ contains a monochromatic or a rainbow congruent copy of $S$. Geh\'{e}r, Sagdeev, and T\'{o}th formalized this dimen

Emil Abasov, Lev Dudko, Daniil Gorin, Oleg Vasilevskii

We present StatTestCalculator (STC), a new open-source statistical analysis tool designed for analysis high energy physics experiments. STC provides both asymptotic calculations and Monte Carlo simulations for computing the exact statistical significance of a discovery or for setting upper limits on signal model parameters. We review the underlying statistic

Peijian Zeng, Guan Wang, Haohao Gu, Xiaoguang Hu

Solving large-scale Partial Differential Equations (PDEs) on complex three-dimensional geometries represents a central challenge in scientific and engineering computing, often impeded by expensive pre-processing stages and substantial computational overhead. We introduce Low-Rank Query-based PDE Solver (LRQ-Solver), a physics-integrated framework engineered

Francesca Scarcella, Bradley J. Kavanagh

Three black-hole low-mass X-ray binaries (LMXBs) in the Milky Way show rates of period decay which cannot be easily explained by standard mechanisms. Recently, it has been claimed that the anomalous period decays in two of these systems may be explained by dynamical friction due to very high dark matter (DM) densities around the black holes. We critically as

Yu Zhang

We present a reduced-scaling auxiliary-field quantum Monte Carlo (AFQMC) framework designed for large molecular systems and ensembles, with or without coupling to optical cavities. Our approach leverages the natural block sparsity of Cholesky decomposition (CD) of electron repulsion integrals in molecular ensembles and employs tensor hypercontraction (THC) t

Lucy D'Agostino McGowan

This paper provides clear and practical guidance on the specification of imputation models when multiple imputation is used in conjunction with doubly robust estimation methods for causal inference. Through theoretical arguments and targeted simulations, we demonstrate that if a confounder has missing data, the corresponding imputation model must include all

Valery Asiryan

In this paper we study the even monic degree-8 cuboid polynomial $P_{a,u}(t)$ introduced by R.A. Sharipov in the first-cuboid specialization of his cuboid equations. For nonzero integers $a,u$ with $u^2\neq a^2$ we prove that $P_{a,u}(t)$ is irreducible in $\mathbb{Z}[t]$ (equivalently, in $\mathbb{Q}[t]$), thus confirming Sharipov's irreducibility conjectur

Krittin Chaowakarn, Paramin Sangwongngam, Nang Htet Htet Aung, Chalie Charoenlarpnopparut

Recent studies in 3D object detection for autonomous vehicles aim to enrich features through the utilization of multi-modal setups or the extraction of local patterns within LiDAR point clouds. However, multi-modal methods face significant challenges in feature alignment, and gaining features locally can be oversimplified for complex 3D object detection task

Tobias Preintner, Weixuan Yuan, Adrian König, Thomas Bäck

Combining large language models with evolutionary computation algorithms represents a promising research direction leveraging the remarkable generative and in-context learning capabilities of LLMs with the strengths of evolutionary algorithms. In this work, we present EvoCAD, a method for generating computer-aided design (CAD) objects through their symbolic

Alexander E. Ulanov, Bastian Ruhnke, Theia El Sharkawy, Thibault Wildi

Synchronised laser oscillators are essential for probing the fastest processes in chemistry, materials science, and biology down to atto-second timescales. Tight synchronisation is also crucial at scientific facilities such as free-electron lasers or radio-telescopes, and increasingly relevant to communication and information technologies in multi-node netwo

Ryan Hazlett, Jennifer Mead, Eli Visbal, Greg L. Bryan

We present an extension of our semi-analytic model that follows the formation of Population III stars and their metal-enriched descendants, incorporating dark matter halo merger trees from cosmological $N$-body simulations and feedback from reionization. Our extended model is calibrated using two complementary cosmological hydrodynamical simulations: Aeos, w

Sepideh Mahabadi, Mohammad Roghani, Jakub Tarnawski, Ali Vakilian

In this work we consider the Metric Steiner Forest problem in the sublinear time model. Given a set $V$ of $n$ points in a metric space where distances are provided by means of query access to an $n\times n$ distance matrix, along with a set of $k$ terminal pairs $(s_1,t_1), \dots, (s_k,t_k)\in V\times V$, the goal is to find a minimum-weight subset of edges

Qianben Chen, Jingyi Cao, Jiayu Zhang, Tianrui Qin

Large language models split into two families: reasoning-centric LLMs, which strengthen internal chain-of-thought reasoning but cannot invoke external tools, and agentic LLMs, which learn to interact with environments and leverage tools but often lag in deep reasoning. This divide arises from fundamentally different training objectives, leading to mismatched

Drew Springham, Edith Elkind, Bart de Keijzer, Maria Polukarov

In multiwinner approval elections with many candidates, voters may struggle to determine their preferences over the entire slate of candidates. It is therefore of interest to explore which (if any) fairness guarantees can be provided under reduced communication. In this paper, we consider voters with one-dimensional preferences: voters and candidates are ass

Yichen Liu, Aerim Si

Semitoric systems are a special type of 4-dimensional integrable system where one of the functions is the moment map of a Hamiltonian $S^1$-action. While their classification is well understood thanks to the work of Pelayo and V{\~u} Ng\d{o}c, relatively few explicit examples are known. Recently, Le Floch and Palmer introduced semitoric transition families i

Eduardo Esteves, Antonio Nigro, Pedro Rizzo

We parameterize by a fine moduli space all degenerations of linear series to a singular curve which is the union of two smooth components meeting transversally at a single point. For this we introduce a novel object in the study of degenerations of linear series, which is the continuous linear series. Our moduli space can be regarded as a Hilbert quotient, i

Oliver H. Wang

Smooth structures on high dimensional manifolds are classified by maps to the infinite loop space $TOP/O$. The homotopy groups of this space are known to be finite. Given a compact Lie group $G$, this space can be regarded as an equivariant infinite loop space and equivariant maps from a locally linear, high dimensional $G$-manifold to $TOP/O$ classify stabl

Nick S. Blunt, Aleksei V. Ivanov, Andreas Juul Bay-Smidt

Trotter product formulas are a natural and powerful approach to perform quantum simulation. However, the error analysis of product formulas is challenging, and their cost is often overestimated. It is established that Trotter error can be bounded in terms of spectral norms of nested commutators of the Hamiltonian partitions [Childs et al., Phys. Rev. X 11, 0

Siheng Xiong, Ali Payani, Faramarz Fekri

Inference-time scaling enhances the reasoning ability of a language model (LM) by extending its chain-of-thought (CoT). However, existing approaches typically generate the entire reasoning chain in a single forward pass, which often leads to CoT derailment, i.e., the reasoning trajectory drifting off course due to compounding errors. This problem is particul

Y. Sun, Y. Ahn, D. Sapkota, H. S. Arachchige

Multiferroic materials, in which electric polarization and magnetic order coexist and couple, offer rich opportunities for both fundamental discovery and technology. However, multiferroicity remains rare due to conflicting electronic requirements for ferroelectricity and magnetism. One route to circumvent this challenge is to exploit the noncollinear orderin

Zehao Chen, Rong Pan, Haoran Li

Human writers often begin their stories with an overarching mental scene, where they envision the interactions between characters and their environment. Inspired by this creative process, we propose a novel approach to long-form story generation, termed hybrid bottom-up long-form story generation, using multi-agent simulations. In our method, agents interact

François Schwarzentruber

In these lecture notes, we first recall the connection between graph neural networks, Weisfeiler-Lehman tests and logics such as first-order logic and graded modal logic. We then present a modal logic in which counting modalities appear in linear inequalities in order to solve verification tasks on graph neural networks. We describe an algorithm for the sati

Elliot L. Epstein, Rose Wang, Jaewon Choi, Markus Pelger

Statistical arbitrage exploits temporal price differences between similar assets. We develop a framework to jointly identify similar assets through factors, identify mispricing and form a trading policy that maximizes risk-adjusted performance after trading costs. Our Attention Factors are conditional latent factors that are the most useful for arbitrage tra

Xurong Xie, Zhucun Xue, Jiafu Wu, Jian Li

Knowledge distillation (KD) is a key technique for compressing large-scale language models (LLMs), yet prevailing logit-based methods typically employ static strategies that are misaligned with the dynamic learning process of student models. These methods typically treat all tokens indiscriminately and apply a single, fixed temperature, resulting in suboptim

Fatemeh Mohammadi, Sebastian Seemann

We study the geometric and algebraic structure of Vandermonde cells, defined as images of the standard probability simplex under the Vandermonde map given by consecutive power sum polynomials. Motivated by their combinatorial equivalence to cyclic polytopes, which are well-known examples of positive geometries and tree amplituhedra, we investigate whether Va

Feng Zhang, Haoyou Deng, Zhiqiang Li, Lida Li

Photo enhancement plays a crucial role in augmenting the visual aesthetics of a photograph. In recent years, photo enhancement methods have either focused on enhancement performance, producing powerful models that cannot be deployed on edge devices, or prioritized computational efficiency, resulting in inadequate performance for real-world applications. To t

Yves Auad, Florian Castioni, Jassem Baaboura, Malo Bézard

Direct electron detection is revolutionizing electron microscopy by offering lower noise, reduced point-spread function, and increased quantum efficiency. Among these advancements, the Timepix3 hybrid-pixel direct electron detector stands out for its unique ability to output temporal information about individual hits within its pixel array. Its event-based a

Cristian Joana, Zi-Yan Yuwen

Primordial black holes (PBHs) are a compelling dark matter candidate and a unique probe of small-scale cosmological fluctuations. Their formation is usually attributed to large positive curvature perturbations, which collapse upon Hubble re-entry during radiation domination. In this work we investigate instead the role of negative curvature perturbations, co

Piotr Tatarczak, Tomasz Fąs, Jan Pawłowski, Aleksandra Krystyna Dąbrowska

Single-photon emitters (SPEs) in two-dimensional materials are highly promising candidates for quantum technologies. SPEs in hexagonal boron nitride (hBN) have been widely investigated, but mostly in exfoliated or powder samples that require an activation process, making it difficult to compare studies and reproduce results. Here, we address this problem and

Xietao Zhou, Steven G. Gilmour

Unreplicated two-level factorial designs are often used in screening experiments to determine which factors out of a large plausible set are active. A theorem regarding the generalized word count pattern is stated and proved for unreplicated designs. It is shown that a phenomenon regarding optimal designs seen in the recent literature can be explained by the

Shiqi Zhang, Xinbei Ma, Yunqing Xu, Zouying Cao

Large Language Models (LLMs) exhibit strong reasoning abilities for planning long-horizon, real-world tasks, yet existing agent benchmarks focus on task completion while neglecting time efficiency in parallel and asynchronous operations. To address this, we present ParaCook, a benchmark for time-efficient collaborative planning. Inspired by the Overcooked ga

Souvik Kundu, Kedar Damle

Motivated by recent work that mapped the low-temperature properties of a class of frustrated spin $S=1$ kagome antiferromagnets with competing exchange and single-ion anisotropies to the fully-packed limit (with each vertex touched by exactly one dimer or nontrivial loop) of a system of dimers and nontrivial (length $s > 2$) loops on the honeycomb lattice, w

Yicheng Xu, Yue Wu, Jiashuo Yu, Ziang Yan

Multimodal Large Language Models (MLLMs) hold promise for accelerating scientific discovery by interpreting complex experimental procedures. However, their true capabilities are poorly understood, as existing benchmarks neglect the fine-grained and long-horizon nature of authentic laboratory work, especially in wet-lab settings. To bridge this gap, we introd

Leonard Bruns, Axel Barroso-Laguna, Tommaso Cavallari, Áron Monszpart

Scene coordinate regression (SCR) has established itself as a promising learning-based approach to visual relocalization. After mere minutes of scene-specific training, SCR models estimate camera poses of query images with high accuracy. Still, SCR methods fall short of the generalization capabilities of more classical feature-matching approaches. When imagi

Indrajith Ekanayake, Sanjula De Alwis

In online retail, customer acquisition typically incurs higher costs than customer retention, motivating firms to invest in churn analytics. However, many contemporary churn models operate as opaque black boxes, limiting insight into the determinants of attrition, the timing of retention opportunities, and the identification of high-risk customer segments. A

Umit Dogan Daglum, Maria Stamenova, Ersoy Sasioglu, Stefano Sanvito

We present an ab initio investigation of magnetic exchange interactions using the spin-spiral method implemented in the VASP code, with a comparative analysis of the self-consistent (SC) and magnetic force theorem (MFT) approaches. Using representative 3d ferromagnets (Fe, Co, Ni) and Mn-based full Heusler compounds, we compute magnon dispersion relations di

Huiyin Xue, Nafise Sadat Moosavi, Nikolaos Aletras

The success of Transformer language models is widely credited to their dot-product attention mechanism, which interweaves a set of key design principles: mixing information across positions (enabling multi-token interactions), sequence-dependent activations (where attention weights adapt to each input), a specific mathematical form (dot-product similarities

Marc Brinner, Sina Zarrieß

We propose SemCSE-Multi, a novel unsupervised framework for generating multifaceted embeddings of scientific abstracts, evaluated in the domains of invasion biology and medicine. These embeddings capture distinct, individually specifiable aspects in isolation, thus enabling fine-grained and controllable similarity assessments as well as adaptive, user-driven

Bo Cheng, Xu Wang, Jinda Liu, Yi Chang

Low-Rank Adaptation (LoRA) has emerged as one of the most widely used parameter-efficient fine-tuning (PEFT) methods for adapting large language models (LLMs) to downstream tasks. While highly effective in single-task settings, it struggles to efficiently leverage inter-task knowledge in complex multi-task learning scenarios, often requiring substantial task