November 2025 arXiv papers — page 46

Luca A. Lanzendörfer, Florian Grötschla

Neural audio codecs have gained recent popularity for their use in generative modeling as they offer high-fidelity audio reconstruction at low bitrates. While human listening studies remain the gold standard for assessing perceptual quality, they are time-consuming and impractical. In this work, we examine the reliability of existing objective quality metric

Gianluca Giacchi, Davide Tramontana

In this work, we investigate the microlocal properties of the evolutions of Schr\"odinger equations using metaplectic Wigner distributions. So far, only restricted classes of metaplectic Wigner distributions, satisfying particular structural properties, have allowed the analysis of microlocal properties. We first extend the microlocal results to all metaplec

Sowmitra Das

In this letter, we introduce a method to synthesize an $n$-qubit Clifford unitary $C$ from the stabilizer tableau of its inverse $C\dag$, using ancilla qubits and measurements. The procedure uses ancillary $|+\rangle$ states, controlled-Paulis, $X$-basis measurements and single-qubit Pauli corrections on the data qubits (based on the measurement results). Th

Salman Sajad Wani, Aatif Kaisar Khan, Saif Al-Kuwari, Mir Faizal

Timing and phase resolution in satellite QKD, kilometre-scale gravitational-wave detectors, and space-borne clock networks hinge on quantum-speed limits (QSLs), yet benchmarks omit relativistic effects for coherent and squeezed probes. We derive first-order relativistic corrections to the Mandelstam-Tamm and Margolus-Levitin bounds. Starting from the Foldy-W

J. V. M. Silveira, H. C. Costa, G. S. Spezzatto, T. L. Prado

In this study, we employ the recently developed recurrence microstate probabilities as features to improve accuracy of several well-established machine learning (ML) algorithms. These algorithms are applied to classify discrete and continuous dynamical systems, as well as colored noise. We demonstrate that the dynamical characteristics quantified by this met

Hongchen Wang, Rafael Espinosa Castañeda, Jay R. Werber, Yao Fehlis

Active learning (AL) accelerates scientific discovery by prioritizing the most informative experiments, but traditional machine learning (ML) models used in AL suffer from cold-start limitations and domain-specific feature engineering, restricting their generalizability. Large language models (LLMs) offer a new paradigm by leveraging their pretrained knowled

Sakib Ahmed, Oscar Pizarro

Unmanned Aerial Vehicles (UAVs) are crucial in Search and Rescue (SAR) missions due to their ability to monitor vast maritime areas. However, small objects often remain difficult to detect from high altitudes due to low object-to-background pixel ratios. We propose an altitude-aware dynamic tiling method that scales and adaptively subdivides the image into t

Steven Peh

Large Language Models (LLMs) remain vulnerable to prompt injection attacks, representing the most significant security threat in production deployments. We present Prompt Fencing, a novel architectural approach that applies cryptographic authentication and data architecture principles to establish explicit security boundaries within LLM prompts. Our approach

Roberto Garrone

Multi-agent systems often operate under feedback, adaptation, and non-stationarity, yet many simulation studies retain static decision rules and fixed control parameters. This paper introduces a general adaptive multi-agent learning framework that integrates: (i) four dynamic regimes distinguishing static versus adaptive agents and fixed versus adaptive syst

Emily Oliphant, Emmanouil Kioupakis, Wenhao Sun

Atomic orbitals underpin our understanding of electronic structure, providing intuitive descriptions of bonding, charge transfer, magnetism, and correlation effects. Despite their utility, an atomic basis that is adaptable, strictly localized on atomic centers, and enables accurate tight-binding interpolation has remained elusive. Here, we introduce Crystal

Sabia Asghar, Qiyao Peng, Fred Vermolen, Cornelis Vuik

The efficient inversion of matrix polynomials is a critical challenge in computational mathematics. We design a procedure to determine the inverse of matrices polynomial of multidimensional Laplace matrices. The method is based on eigenvector and eigenvalue expansions. The method is consistent with previous expressions of the inverse discretized Laplacian in

Chenyang Qiu, Zongli Lin

This paper studies a class of distributed optimization problems with coupled equality constraints in networked systems. Many existing distributed algorithms rely on solving local subproblems via the $\operatorname{argmin}$ operator in each iteration. Such approaches become computationally burdensome or intractable when local cost functions are complex. To ad

Ilya O. Ryzhov, John Gunnar Carlsson, Yinchu Zhu

Socioeconomic segregation often arises in school districting and other contexts, causing some groups to be over- or under-represented within a particular district. This phenomenon is closely linked with disparities in opportunities and outcomes. We formulate a new class of geographical partitioning problems in which the population is heterogeneous, and it is

Pedro Marun, Saharon Shelah, Corey Bacal Switzer

Abstractly, the generic extensions after $\aleph_\omega$-many Cohen reals and $\aleph_{\omega+1}$-many Cohen reals must be different for reasons of uniform density the relevant Boolean algebras. Nevertheless this is not satisfying and it would be nice to pin the difference between the two models down to some mathematical or combinatorial principle. In this p

Nikola Crnković, Goran Duplančić, Saad Nabeebaccus, Kornelija Passek-K.

We investigate the hard exclusive photoproduction of photon-meson pairs at leading-twist and leading-order in perturbative QCD, and focus on pseudoscalar mesons $\text{M} \in \{\pi^\pm, \pi^0, \eta, \eta'\}$. Compact analytical expressions are obtained for the amplitudes involving quark generalized parton distributions, with the two-gluon components of the $

Mobina Mehrazar, Mohammad Amin Yousefi, Parisa Abolfath Beygi, Behnam Bahrak

Large language models (LLMs) are increasingly used to generate self-explanations alongside their predictions, a practice that raises concerns about the faithfulness of these explanations, especially in low-resource languages. This study evaluates the faithfulness of LLM-generated explanations in the context of emotion classification in Persian, a low-resourc

Chengwei Zhou, Vipin Chaudhary, Gourav Datta

The computational overhead of Vision Transformers in practice stems fundamentally from their deep architectures, yet existing acceleration strategies have primarily targeted algorithmic-level optimizations such as token pruning and attention speedup. This leaves an underexplored research question: can we reduce the number of stacked transformer layers while

Vikram Ramavarapu, João Alfredo Cardoso Lamy, Mohammad Dindoost, David A. Bader

Community detection, or network clustering, is used to identify latent community structure in networks. Due to the scarcity of labeled ground truth in real-world networks, evaluating these algorithms poses significant challenges. To address this, researchers use synthetic network generators that produce networks with ground-truth community labels. RECCS is o

Theodor Hagström, Lars Herre

Decarbonisation, decentralisation, and intermittency are driving the development of flexibility markets towards shorter market time units (MTU). Shorter MTUs and shorter gate closures lower the entrance barriers of demand side aggregators that face significant uncertainty on longer time scales. We study the business case for aggregated EV fleets participatin

Chenyang Qiu, Zongli Lin

We study distributed convex optimization with two ubiquitous forms of coupling: consensus constraints and global affine equalities. We first design a linearized method of multipliers for the consensus optimization problem. Without smoothness or strong convexity, we establish non-ergodic sublinear rates of order O(1/\sqrt{k}) for both the objective optimality

Wilbert J. Smit, Thomas Gibaud, Sébastien Manneville, Thibaut Divoux

Colloidal gels form through the sol-gel transition of attractive particle suspensions, where local aggregation leads to a space-spanning network with solid-like properties. Their microstructure and mechanical properties are highly sensitive to external perturbations, which can substantially alter the pathway of network formation. Here, we investigate how non

Jonathan Jenvrin

We establish an explicit lower bound for the N\'eron-Tate height on elliptic curves with complex multiplication, for nontorsion points defined over the maximal abelian extension of a number field. Building on a strategy developed by Amoroso, David, and Zannier, we provide an alternative proof of a theorem originally due to Baker. The novelty in our approach

Marlon Estanislau

Let $G$ be a finite $p$-group with normal subgroup $N$, and $R$ a complete discrete valuation ring in mixed characteristic. We characterize permutation $RG$-modules in terms of modules for $RN$ and $R[G/N]$. The result generalizes both the seminal detection theorem for permutation modules due to Weiss, who characterizes those permutation $RG$-modules that ar

Lukas Molnar, Jin Cheng, Gabriele Fadini, Dongho Kang

Loco-manipulation demands coordinated whole-body motion to manipulate objects effectively while maintaining locomotion stability, presenting significant challenges for both planning and control. In this work, we propose a whole-body model predictive control (MPC) framework that directly optimizes joint torques through full-order inverse dynamics, enabling un

Gabriele Cesar Iwashima, Claudia Susie Rodrigues, Claudio Dipolitto, Geraldo Xexéo

Large language models (LLMs) may generate outputs that are misaligned with user intent, lack contextual grounding, or exhibit hallucinations during conversation, which compromises the reliability of LLM-based applications. This review aimed to identify and analyze techniques that align LLM responses with conversational goals, ensure grounding, and reduce hal

Chenyang Qiu, Yangyang Qian, Zongli Lin, Yacov A. Shamash

This paper studies distributed convex optimization with both affine equality and nonlinear inequality couplings through the duality analysis. We first formulate the dual of the coupling-constraint problem and reformulate it as a consensus optimization problem over a connected network. To efficiently solve this dual problem and hence the primal problem, we de

Dennis Bodewits, John W. Noonan, Michael S. P. Kelley, Carrie E. Holt

Comet C/2025 K1 (ATLAS) reached perihelion at 0.33 au on 2025 October 8. Daily monitoring by the LCO Outbursting Objects Key Project revealed a major activity increase between November 2 and 4, accompanied by rapid changes in coma morphology. Serendipitous HST/STIS acquisition images obtained on November 8-10 captured the comet only days after this event and

Ziteng Sun, Adrian Benton, Samuel Kushnir, Asher Trockman

Post-training quantization is an effective method for reducing the serving cost of large language models, where the standard approach is to use a round-to-nearest quantization level scheme. However, this often introduces large errors due to outliers in the weights. Proposed mitigation mechanisms include applying adaptive rounding, random rotation transformat

Omar Alama, Darshil Jariwala, Avigyan Bhattacharya, Seungchan Kim

Open-vocabulary semantic segmentation (OVSS) underpins many vision and robotics tasks that require generalizable semantic understanding. Existing approaches either rely on limited segmentation training data, which hinders generalization, or apply zero-shot heuristics to vision-language models (e.g CLIP), while the most competitive approaches combine multiple

A. Massarenti, M. Mella

We study the dimension and identifiability of neurovarieties associated to polynomial neural networks. We give an independent geometric proof that the linear bounds $d_i\geq 2n_i-1$ on the activation degrees imply non defectiveness for any number of outputs, a dimension statement previously obtained from finite identifiability. The proof is based on a direct

Yorick Andeweg, John Kitching, Matthew T. Hummon

We implement a feedback protocol to suppress the AC Stark shift in a two-photon rubidium optical frequency reference, reducing its sensitivity to optical power variations by a factor of 1000. This method alleviates the tradeoff between short-term and long-term stability imposed by the AC Stark shift, enabling us to simultaneously achieve instabilities of $3\

Josef Richter, Masudul Haque, Lucas Sá

We develop an approach for understanding the dynamics of open quantum systems by analyzing individual quantum trajectories in the eigenbasis of the Liouvillian superoperator. From trajectory-eigenstate overlaps, we construct a quasiprobability distribution that characterizes the degree of localization of the trajectories in the Liouvillian eigenbasis. Contra

Charles Fleming, Luca Muscariello, Vijoy Pandey, Ramana Kompella

Large Language Models (LLMs) have demonstrated remarkable performance improvements and the ability to learn domain-specific languages (DSLs), including APIs and tool interfaces. This capability has enabled the creation of AI agents that can perform preliminary computations and act through tool calling, which is now being standardized via protocols like MCP.

Amir Hossein Houshmand Almani, Ali Mortezapour, Alireza Nourmandipour

We investigate how detuning and auxiliary qubits collaboratively enhance quantum synchronization in a dissipative multi-qubit system that is coupled to a structured reservoir. Our findings indicate that while detuning is ineffective in Markovian environments, it emerges as a powerful control parameter in the non-Markovian regime, where environmental memory f

Aristides Kontogeorgis, Orestis Lygdas

We compute explicit bases for the de Rham cohomology of cyclic covers of the projective line defined over an algebraically closed field of characteristic $p\geq 0$. For both Kummer and Artin-Schreier extensions, we describe precise $k$-bases for the cohomology groups $H^{1}(X,\mathcal{O}_{X})$ and $H^{0}(X,\Omega_{X})$, and we use these to construct an expli

Josep Martí, Pedro Luis Luque-Escamilla

V4641 Sagittarii (V4641 Sgr) is a unique Galactic microquasar system featuring a stellar-mass black hole accreting matter from a massive companion. One of its intriguing features is the presence of relativistic radio jets almost perpendicular to the observed extended gamma-ray emission, implying significant propagation effects or interactions with the Galact

Chin-Chia Michael Yeh, Uday Singh Saini, Junpeng Wang, Xin Dai

The ubiquity of time series data creates a strong demand for general-purpose foundation models, yet developing them for classification remains a significant challenge, largely due to the high cost of labeled data. Foundation models capable of in-context learning (ICL) offer a powerful solution, adapting to new tasks with minimal examples and reducing the nee

Chin-Chia Michael Yeh, Uday Singh Saini, Xin Dai, Xiran Fan

Payment networks form the backbone of modern commerce, generating high volumes of transaction records from daily activities. Properly modeling this data can enable applications such as abnormal behavior detection and consumer-level insights for hyper-personalized experiences, ultimately improving people's lives. In this paper, we present TREASURE, TRansforme

E. V. Gorbar, A. I. Momot

For dwarf galaxies modeled as deformed Plummer spheres and orbitally moving in ultralight dark matter halo of the Milky Way, the torque induced by the dynamical friction force is determined. The impact of this torque, as well as the torque produced by the gravitational force of the Milky Way, on the internal kinematics of dwarf galaxies is studied. Possible

Thomas Marshall Vielmetti, Devansh R Agrawal, Dimitra Panagou

We present Multi-Agent gatekeeper, a framework that provides provable safety guarantees for leader-follower formation control in cluttered 3D environments. Existing methods face a trad-off: online planners and controllers lack formal safety guarantees, while offline planners lack adaptability to changes in the number of agents or desired formation. To addres

Niklas Haas, Sören Schmitt, Rob van Stee

We consider the online buffer minimization in multiprocessor systems with conflicts problem (in short, the buffer minimization problem) in the recently introduced flow model. In an online fashion, workloads arrive on some of the $n$ processors and are stored in an input buffer. Processors can run and reduce these workloads, but conflicts between pairs of pro

Sergiy V. Lukyanov

The time evolution of the Wigner distribution function for a single-particle excitation in a Fermi system was studied within the framework of the diffusion approximation of kinetic theory by numerically solving a nonlinear diffusion equation with constant kinetic coefficients. A method was proposed to separate the dissipative processes into contributions fro

Sreenivas Raguraman, Homero Reyes Pulido, Christopher Hutchinson, Arun Devaraj

Vacancy engineering, the intentional control of atomic-scale vacancies in metals and alloys, is emerging as a powerful yet underexplored strategy for tailoring microstructures and optimizing performance across diverse applications. By enabling excess vacancy populations through quenching, severe deformation, thermomechanical treatments, or additive manufactu

Paula Lackie, Elliot Pickens, Dashiell Coyier

The DataSquad at Carleton College addresses a common problem at small liberal arts colleges: limited capacity for data services and few opportunities for students to gain practical experience with data and software development. Academic Technologist Paula Lackie designed the program as a work-study position that trains undergraduates through structured peer

Abdurahman Ali Mohammed, Wallapak Tavanapong, Catherine Fonder, Donald S. Sakaguchi

Cell counting in biomedical imaging is pivotal for various clinical applications, yet the interpretability of deep learning models in this domain remains a significant challenge. We propose a novel prototype-based method for interpretable cell counting via density map estimation. Our approach integrates a prototype layer into the density estimation network,

Tao-Zhi Yang, Zhao-Yu Zuo, Shi-ping Guo, Xu Ding

In this paper, we present a detailed analysis of the light variation of KIC 5623923 using high-precision time-series data from the $Kepler$ mission. The analysis reveals this target is an eclipsing binary system with $\delta$ Scuti type pulsations from the primary component, rather than from the secondary as previously reported. The frequency analysis of thr

Vivek Chavan, Yasmina Imgrund, Tung Dao, Sanwantri Bai

We introduce IndEgo, a multimodal egocentric and exocentric dataset addressing common industrial tasks, including assembly/disassembly, logistics and organisation, inspection and repair, woodworking, and others. The dataset contains 3,460 egocentric recordings (approximately 197 hours), along with 1,092 exocentric recordings (approximately 97 hours). A key f

Bilal Ahmed, Joseph G. Makin

Inverse problems are fundamental to science and engineering, where the goal is to infer an underlying signal or state from incomplete or noisy measurements. Recent approaches employ diffusion models as powerful implicit priors for such problems, owing to their ability to capture complex data distributions. However, existing diffusion-based methods for invers

Eugene Lavretsky

In this paper, a state feedback control design with min/max operational limiting constraints is developed for multi-input-multi-output linear time invariant systems. Specifically, servo-tracking control problems with input and output constraints are considered. For static servo-controllers, the output design limits are imposed component-wise on the system se

Christina Reissel, Devin Lai, Shivanshu Dwivedi, Edgard Bonilla

The unprecedented sensitivity of the Laser Interferometer Gravitational-Wave Observatory, which enables the detection of distant astrophysical sources, also renders the detectors highly susceptible to low-frequency ground motion. Persistent microseisms in the 0.1-0.3 Hz band couple into the instruments, degrade lock stability, and contribute substantially to

Tabia Tanzin Prama, Christopher M. Danforth, Peter Sheridan Dodds

Large Language Models (LLMs) have demonstrated strong translation abilities through prompting, even without task-specific training. However, their effectiveness in dialectal and low-resource contexts remains underexplored. This study presents the first systematic investigation of LLM-based machine translation (MT) for Sylheti, a dialect of Bangla that is its

Yuchen Bi, Jie Zhou

For an integral $2$-varifold $V\subset \mathbb{S}^3$ with square-integrable mean curvature, unit density, and support of genus at least $1$, assume that its Willmore energy satisfies \[ \mathcal{W}(V)\le 2\pi^2+\delta^2,\qquad \delta<\delta_0\ll1. \] We show that the support $\Sigma=\operatorname{spt}V$ is, after applying a suitable conformal transformation

Ying Bao, Jessie Liu

This paper investigates how content moderation affects content creation in an ideologically diverse online environments. We develop a model in which users act as both creators and consumers, differing in their ideological affiliation and propensity to produce toxic content. Affective polarization, i.e., users' aversion to ideologically opposed content, inter

Megala Anandan, Mária Lukáčová-Medvid'ová

We develop structure-preserving finite volume schemes for the barotropic Euler equations in the low Mach number regime. Our primary focus lies in ensuring both the asymptotic-preserving (AP) property and the discrete entropy stability. We construct an implicit-explicit (IMEX) method with suitable acoustic/advection splitting including implicit numerical diff

Zachary Halberstam, Carl Schildkraut

Given a word $w$, what is the maximum possible number of appearances of $w$ reading contiguously along any of the directions in $\{-1, 0, 1\}^d \setminus \{\mathbf{0}\}$ in a large $d$-dimensional grid (as in a word search)? Patchell and Spiro first posed a version of this question, which Alon and Kravitz completely answered for a large class of "well-behave

Benjamin Stockton, Michele Santacatterina, Soutrik Mandal, Charles M. Cleland

Sequential parallel comparison design (SPCD) clinical trials aim to adjust active treatment effect estimates for placebo response to minimize the impact of placebo responders on the estimates. This is potentially accomplished using a two stage design by measuring treatment effects among all participants during the first stage, then classifying some placebo a

Parsa Madinei, Ryan Solgi, Ziqi Wen, Jonathan Skaza

We introduce INTERLACE, a novel framework that prunes redundant layers in VLMs while maintaining performance through sample-efficient finetuning. Existing layer pruning methods lead to significant performance drop when applied to VLMs. Instead, we analyze triplets of consecutive layers to identify local redundancy, removing the most redundant of the first tw

Jiarui Wang, Mahyar Fazlyab

This paper presents the Safe Sequential Quadratically Constrained Quadratic Programming (SS-QCQP) algorithm, a first-order method for smooth inequality-constrained nonconvex optimization that guarantees feasibility at every iteration. The method is derived from a continuous-time dynamical system whose vector field is obtained by solving a convex QCQP that en

Tianhai Luo, Katie R. Gann, Cameron A. Gorsak, Ming-Chiang Chang

Among ultrawide bandgap semiconductors, beta-Ga2O3 is particularly promising for high power and frequency applications. For devices, n-type concentrations above 10^19 cm^-3 are required. Ge is a promising alternative n-type dopant with an ionic radius similar to Ga. Homoepitaxial 010 beta-Ga2O3 films were implanted with Ge to form 50 and 100 nm box concentra

Wuhuan Deng, Scott Nestler

Plate discipline is an important feature of a hitter's success. Hitter who are able to recognize good pitches to swing at and balls to take are generally recognized as disciplined hitters. Although there are some metrics that can provide insight into the patience of a hitter, most do not capture the ability of a batter to take balls. In this research, we int

Rishab Sharma, Iman Saberi, Elham Alipour, Jie JW Wu

Financial applications of large language models (LLMs) require factual reliability and computational efficiency, yet current systems often hallucinate details and depend on prohibitively large models. We propose FISCAL (Financial Synthetic Claim-Document Augmented Learning), a modular framework for generating synthetic data tailored to financial fact-checkin

Luis Ferreirinha, Iberia Medeiros

Cyber-Physical Systems have played an essential role in our daily lives, providing critical services such as power and water, whose operability, availability, and reliability must be ensured. The C programming language, prevalent in CPS development, is crucial for system control where reliability is critical. However, it is also commonly susceptible to vulne

Souradip Poddar, Chia-Tung Ho, Ziming Wei, Weidong Cao

Conventional AI-driven AMS design automation algorithms remain constrained by their reliance on high-quality datasets to capture underlying circuit behavior, coupled with poor transferability across architectures, and a lack of adaptive mechanisms. This work proposes HeaRT, a hierarchical circuit reasoning-based agentic framework for automation loops and a s

Henry C. Hammer, Hassan A. Bukhari, Yogendra Limbu, Brett M. Wasick

Harnessing rare-earth ions in oxides for quantum networks requires integration with bright emitters in III-V semiconductors, but local disorder and interfacial noise limit their optical coherence. Here, we investigate the microscopic origins of the ensemble spectrum in Er$^{3+}$:TiO$_2$ epitaxial thin films on GaAs and GaSb substrates. Ab initio calculations

Istiak Ahmed, Galib Ahmed, K. Shahriar Sanjid, Md. Tanzim Hossain

OncoVision is a multimodal AI pipeline that combines mammography images and clinical data for better breast cancer diagnosis. Employing an attention-based encoder-decoder backbone, it jointly segments four ROIs - masses, calcifications, axillary findings, and breast tissues - with state-of-the-art accuracy and robustly predicts ten structured clinical featur

Nikky Avila, Hank He, Reza Rastegar, Jamie Tolan

Carbon matching aims to improve corporate carbon accounting by tracking emissions rather than energy consumption and production. We present a mathematical derivation of carbon matching using marginal emission rates, where the unit of matching is tons of carbon emitted. We present analysis and open source notebooks showing how marginal emissions can be calcul

Deni Salja

Persistent homology is a popular technique in topological data analysis that tracks the lifespans of homological features in a nested sequence of spaces. This data is typically presented in a multi-set called a persistence diagram or a barcode. For single parameter filtrations with homology coefficient taken in a principal ideal domain, the persistence diagr

Jiaxin Shi, Michalis K. Titsias

We derive a new theoretical interpretation of the reweighted losses that are widely used for training diffusion models. Our method is based on constructing a cascade of time-dependent variational lower bounds on the data log-likelihood, that provably improves upon the standard evidence lower bound and results in reduced data-model KL-divergences. Combining s

Ahmed Awadallah, Yash Lara, Raghav Magazine, Hussein Mozannar

Progress in computer use agents (CUAs) has been constrained by the absence of large and high-quality datasets that capture how humans interact with a computer. While LLMs have thrived on abundant textual data, no comparable corpus exists for CUA trajectories. To address these gaps, we introduce FaraGen, a novel synthetic data generation system for multi-step

Eric R. Bittner

Squeezed reservoirs provide a powerful means of engineering nonclassical noise and controlling irreversible dynamics in open quantum systems. Here we develop a comprehensive analysis of two coupled harmonic oscillators driven by independent squeezed baths, focusing on the emergence of coherence-driven entropy flow and the structure of exceptional points (EPs

Xinhai Hou, Shaoyuan Xu, Manan Biyani, Moyan Li

Agentic vision-language models are increasingly trained to "think with images" by calling image operations. However, we show that high final-answer accuracy often hides unfaithful visual reasoning: models may invoke tools on irrelevant regions or ignore tool outputs entirely, yet still guess the correct answer. In this work, we first propose a faithfulness e

Biswaraj Palit, Abhijeet Borkar, Agata Różańska, Alex Markowitz

The Changing-Look active galactic nucleus Mkn 590 is currently in a rejuvenated state, exhibiting a contemporaneous flux rise across X-rays, UV, optical and cm wavelengths. In this study, we present three new optical spectra obtained with the Nordic Optical Telescope, alongside three 1.4 GHz continuum measurements from the Giant Meterwave Radio Telescope, ac

Peter Brearley, Philipp Pfeffer

Dissipation and irreversibility are central to many physical systems, yet they lead to non-unitary dynamics that are challenging to realise on quantum processors. High-order operator splitting is an attractive approach for simulating unitary dynamics, yet conventional product formulas introduce negative time steps at high orders that are ill-conditioned for

D. Karas, K. Badgley, Z. Chen, V. Chernenok

The Proton Improvement Plan II (PIP-II) project is a vital upgrade to the Fermilab accelerator complex. The magnet pulse rate of the PIP-II Injection system requires an increase from the current rate of 15 Hz to 20 Hz as well as a roughly 30% increase in the magnetic field of the new Orbital Bump (ORBUMP) pulsed dipole magnets in the Booster. The ORBUMP magn

Sepideh Koohfar

Time-series forecasting remains difficult in real-world settings because temporal patterns operate at multiple scales, from broad contextual trends to fast, fine-grained fluctuations that drive critical decisions. Existing neural models often struggle to represent these interacting dynamics, leading to unstable predictions and reduced reliability in downstre

Shantanu Rahman, Nayeb Hasin, Mainul Islam, Golam Sarowar

Autonomous vehicles are becoming popular day by day not only for autonomous road traversal but also for industrial automation, farming and military. Most of the standard vehicles follow the Ackermann style steering mechanism. This has become to de facto standard for large and long faring vehicles. The local planner of an autonomous vehicle controls the low-l

Stephen C. Gravereaux, Sheikh Rabiul Islam

This study examines whether Low-Rank Adaptation (LoRA) fine-tuned Large Language Models (LLMs) can approximate the performance of fully fine-tuned models in generating human-interpretable decisions and explanations for malware classification. Achieving trustworthy malware detection, particularly when LLMs are involved, remains a significant challenge. We dev

Mahmud Suhaimi Ibrahim, Shantanu Rahman, Muhammad Samin Hasan, Minhaj Uddin Ahmad

Collision-free path planning is the most crucial component in multi-UAV formation-flying (MFF). We use unlabeled homogenous quadcopters (UAVs) to demonstrate the use of a flow network to create complete (inter-UAV) collision-free paths. This procedure has three main parts: 1) Creating a flow network graph from physical GPS coordinates, 2) Finding a path of m

Abolfazl Moslemi, Hossein Peyvandi

Early and accurate detection of Alzheimer's disease (AD) is crucial for enabling timely intervention and improving outcomes. However, developing reliable machine learning (ML) models for AD diagnosis is challenging due to limited labeled data, multi-site heterogeneity, and class imbalance. We propose a Transformer-based diagnostic framework that combines dif

Lishuo Pan, Mattia Catellani, Thales C. Silva, Lorenzo Sabattini

Control barrier functions (CBFs) are an effective model-based tool to formally certify the safety of a system. With the growing complexity of modern control problems, CBFs have received increasing attention in both optimization-based and learning-based control communities as a safety filter, owing to their provable guarantees. However, success in transferrin

Ana León Contreras, Ricardo Amorín, Mario Llerena, Vital Fernández

We present a study of the ionized gas kinematics and feedback properties in a sample of 14 low-mass, UV-luminous Lyman Break Analogs (LBAs) at redshifts z~0.1-0.3. These compact, strongly star-forming galaxies serve as local analogs of high-redshift starbursts. Using high-resolution VLT/X-shooter spectra, we model the optical emission-line profiles, includin

Angelo Gaspar Diniz Nogueira, Kayua Oleques Paim, Hendrio Bragança, Rodrigo Brandão Mansilha

The ever-increasing number of Android devices and the accelerated evolution of malware, reaching over 35 million samples by 2024, highlight the critical importance of effective detection methods. Attackers are now using Artificial Intelligence to create sophisticated malware variations that can easily evade traditional detection techniques. Although machine

Manil Shrestha, Edward Kim

Multi-hop question answering over knowledge graphs remains computationally challenging due to the combinatorial explosion of possible reasoning paths. Recent approaches rely on expensive Large Language Model (LLM) inference for both entity linking and path ranking, limiting their practical deployment. Additionally, LLM-generated answers often lack verifiable

Jennifer Grannen, Michelle Pan, Kenneth Llontop, Cherie Ho

Foundation models (FM) have unlocked powerful zero-shot capabilities in vision and language, yet their reliance on internet pretraining data leaves them brittle in unstructured, real-world settings. The messy, real-world data encountered during deployment (e.g. occluded or multilingual text) remains massively underrepresented in existing corpora. Robots, as

Tara Fetherolf, Sadie G. Welter, Colby M. Ostberg, Stephen R. Kane

Planetary atmospheric energy budgets primarily depend on stellar incident flux. However, stellar variability can have major consequences for the evolution of planetary climates. In this work, we evaluate how stellar variability influences the equilibrium temperature and water retention of planets within the Habitable Zone (HZ). We present a sample of 9 stars

Stelios Stefas, George Zoupanos

Within the gauge-theoretic approach of gravity, the gauging of an enlarged symmetry of the tangent space in four dimensions allows gravity to be unified with internal interactions. We study the unification of the Conformal and Noncommutative (Fuzzy) Gravities with Internal Interactions based on the $SO(10)$ GUT.

Damodar Panigrahi, Raj Patel, Shaswata Mitra, Sudip Mittal

Modern enterprise systems face escalating cyber threats that are increasingly dynamic, distributed, and multi-stage in nature. Traditional intrusion detection and response systems often rely on static rules and manual workflows, which limit their ability to respond with the speed and precision required in high-stakes environments. To address these challenges

D. Baranov, O. Pochinka

An isotopy between two diffeomorphisms means the existence of an arc connecting them in the space of diffeomorphisms. Among such arcs there are so-called stable arcs, which do not qualitatively change under small perturbations. In the present paper we consider a set of gradient-like diffeomorphisms f of 2-torus whose induced isomorphism given by a matrix $\b

Wuhuan Deng

Runs Batted IN (RBI) records the number of runs a hitter directly drives in during their plate appearances and reflects a batter's ability to convert opportunities into scoring. Because producing runs determines game outcomes, RBI has long served as a central statistic in evaluating offensive performance. However, traditional RBI treats all batted-in runs eq

Ruimin Feng, Xingxin He, Ronald Mercer, Zachary Stewart

Purpose: To investigate whether a vision-language foundation model can enhance undersampled MRI reconstruction by providing high-level contextual information beyond conventional priors. Methods: We proposed a semantic distribution-guided reconstruction framework that uses a pre-trained vision-language foundation model to encode both the reconstructed image a

João M. Alendouro Pinho, Simão S. Cardoso, Yuliy V. Bludov, João M. Viana Parente Lopes

In this paper we present a theoretical examination of second-harmonic generation (SHG) in a graphene monolayer integrated within an attenuated total internal reflection (ATR) configuration. By embedding graphene in this optical setup, we explore the enhancement in the nonlinear optical response, particularly focusing on the efficiency of SHG. Our analysis re

Niccolò Brembilla, Yinbin Ma, Pietro Belotti, Federico Malucelli

Inspired by prior work by Tian and by Cao and Xu, this paper presents an efficient computer-aided framework to characterize the fundamental limits of coded caching systems under the constraint of linear coding. The proposed framework considers non-Shannon-type inequalities which are valid for representable polymatroids (and hence for linear codes), and lever

Haokun Zhou

Data-driven weather prediction models implicitly assume that the statistical relationship between predictors and targets is stationary. Under anthropogenic climate change, this assumption is violated, yet the structure of the resulting concept drift remains poorly understood. Here we introduce concept drift of simple forecast models as a diagnostic of atmosp

Max Nilsson, Anton Åkerman, Pontus Giselsson

The Douglas-Rachford splitting method is a classical and widely used algorithm for solving monotone inclusions involving the sum of two maximally monotone operators. It was recently shown to be the unique frugal, no-lifting resolvent-splitting method that is unconditionally convergent in the general two-operator setting. In this work, we show that this uniqu

Abhi Chivukula, Jay Somasundaram, Vijay Somasundaram

LLM-based coding agents are increasingly common but still face challenges in context management, latency, reliability, reproducibility, and scalability. We present Agint, an agentic graph compiler, interpreter, and runtime that incrementally and hierarchically converts natural-language instructions into typed, effect-aware code DAGs. Agint introduces explici

Peize Ding, Nishchhal Verma, Raquel Queiroz

Transition metal dichalcogenides (TMDs) exhibit unconventional Landau level (LL) spectra that cannot be fully captured by an effective mass approximation or a minimal two-band Dirac model. Namely, TMDs show an anomalous, upward-sloping zeroth LL in the valence band and an asymmetric orbital magnetization between electron and hole bands. In this paper, we emp

Marie Martig, Francesca Pinna, Jesús Falcón-Barroso, Ignacio Martín-Navarro

We present a detailed analysis of the vertical and radial structure of mono-age stellar populations in three edge-on lenticular galaxies (FCC 153, FCC 170, and FCC 177) in the Fornax cluster, using deep MUSE observations. By measuring the half-mass radius (R$_{50}$) and half-mass height (z$_{50}$) across 1 Gyr-wide age bins, we trace the spatial evolution of

Emanuele Tumbiolo, Lorenzo Maccone, Chiara Macchiavello, Matteo G. A. Paris

Quantum thermometry aims at determining temperature with ultimate precision in the quantum regime. Standard equilibrium approaches, limited by the Quantum Fisher Information given by static energy fluctuations, lose sensitivity outside a fixed temperature window. Non-equilibrium strategies have therefore been recently proposed to overcome these limits, but t

Valentina P. Miranda, Patricia B. Tissera, Emanuel Sillero, Jenny Gonzalez-Jara

Supernova (SN) feedback-driven galactic outflows are a key physical process that contributes to the baryon cycle by regulating the star formation activity, reducing the amount of metals in low-mass galaxies and enriching the circumgalactic (CGM) and intergalactic media (IGM). We aim to understand the chemical loop of sub-Milky Way (MW) galaxies and their nea

Chi Hsuan Wu, Kumar Ashutosh, Kristen Grauman

Egocentric perception on smart glasses could transform how we learn new skills in the physical world, but automatic skill assessment remains a fundamental technical challenge. We introduce SkillSight for power-efficient skill assessment from first-person data. Central to our approach is the hypothesis that skill level is evident not only in how a person perf