October 2025 arXiv papers — page 66

Xiaolong Wang, Lixiang Ru, Ziyuan Huang, Kaixiang Ji

We propose a novel AutoRegressive Generation-based paradigm for image Segmentation (ARGenSeg), achieving multimodal understanding and pixel-level perception within a unified framework. Prior works integrating image segmentation into multimodal large language models (MLLMs) typically employ either boundary points representation or dedicated segmentation heads

Sandra Kiefer, T. Devini de Mel

The Colour Refinement algorithm is a classical procedure to detect symmetries in graphs, whose most prominent application is in graph-isomorphism tests. The algorithm and its generalisation, the Weisfeiler-Leman algorithm, evaluate local information to compute a colouring for the vertices in an iterative fashion. Different final colours of two vertices certi

Simon Zhang

Voxelized vector field data consists of a vector field over a high dimensional lattice. The lattice consists of integer coordinates called voxels. The voxelized vector field assigns a vector at each voxel. This data type encompasses images, tensors, and voxel data. Assume there is a nice energy function on the vector field. We consider the problem of lossy c

Shiva Sreeram, Alaa Maalouf, Pratyusha Sharma, Daniela Rus

Recently, Sharma et al. suggested a method called Layer-SElective-Rank reduction (LASER) which demonstrated that pruning high-order components of carefully chosen LLM's weight matrices can boost downstream accuracy -- without any gradient-based fine-tuning. Yet LASER's exhaustive, per-matrix search (each requiring full-dataset forward passes) makes it imprac

Wei-Chen Lin, Dong-han Yeom, Dejan Stojkovic

For a unitary description of an evaporating black hole, one usually chooses the time slices that cover only outside of the event horizon, which is mostly problem-free because the event horizon is not encountered. However, is there any justification for avoiding time slices that cover inside the event horizon? To answer the question, we investigate the Wheele

Jose J. Orquin-Marques, Carlos Flores-Garrigos, Alejandro Gomez Cadavid, Anton Simen

We present a quantum-native approach to quantum feature selection (QFS) based on analog quantum simulation with neutral atom arrays, adaptable to a variety of academic and industrial applications. In our method, feature relevance-measured via mutual information with the target-is encoded as local detuning amplitudes, while feature redundancy is embedded thro

Gabriel Grand, Valerio Pepe, Jacob Andreas, Joshua B. Tenenbaum

Many emerging applications of AI--from scientific discovery to medical diagnosis--require agents to seek information strategically: forming hypotheses, asking targeted questions, and making decisions under uncertainty. In high-stakes settings with limited resources, do language models (LMs) behave like rational agents? Drawing on insights from human cognitio

Yair Feldman, Yoav Artzi

Context compression reduces Transformer inference costs by replacing lengthy inputs with shorter pre-computed representations. It carries significant benefits for retrieval-augmented generation (RAG) and has attracted growing research attention. However, progress remains difficult to measure due to inconsistent evaluations and baselines. We design a standard

Chenheng Zhang, Tianqi Du, Jizhe Zhang, Mingqing Xiao

Conventional research on large language models (LLMs) has primarily focused on refining output distributions, while paying less attention to the decoding process that transforms these distributions into final responses. Recent advances, such as scaling the computation of inference time with reward models, have underscored the importance of decoding, but thes

John Sengendo, Fabrizio Granelli

As 5G and future 6G mobile networks become increasingly more sophisticated, the requirements for agility, scalability, resilience, and precision in real-time service provisioning cannot be met using traditional and heuristic-based resource management techniques, just like any advancing technology. With the aim of overcoming such limitations, network operator

Juan Alejandro Pinto Castro, Héctor J. Hortúa, Jorge Enrique García-Farieta, Roger Anderson Hurtado

Deep learning has emerged as a transformative methodology in modern cosmology, providing powerful tools to extract meaningful physical information from complex astronomical datasets. This paper implements a novel Bayesian graph deep learning framework for estimating key cosmological parameters in a primordial magnetic field (PMF) cosmology directly from simu

Lei Cheng, Siyang Cao

This paper presents a Multi-Object Tracking (MOT) framework that fuses radar and camera data to enhance tracking efficiency while minimizing manual interventions. Contrary to many studies that underutilize radar and assign it a supplementary role--despite its capability to provide accurate range/depth information of targets in a world 3D coordinate system--o

E. de la Hoz, A. Mennella, K. Arnold, C. Baccigalupi

Upcoming cosmic microwave background (CMB) experiments aim to detect primordial gravitational waves with unprecedented sensitivity. Effective foreground removal is essential to avoid biases in the measurement of the tensor-to-scalar ratio ($r$) in this high-precision regime. Recent analyses highlight the unexpected complexity of synchrotron emission at low f

Liang Ye, Shengqin Chen, Jiazhu Dai

The rapid progress of graph generation has raised new security concerns, particularly regarding backdoor vulnerabilities. Though prior work has explored backdoor attacks against diffusion models for image or unconditional graph generation, those against conditional graph generation models, especially text-guided graph generation models, remain largely unexam

Zhonghao Tu, Ang Li

Glitches in neutron stars originate from the sudden transfer of angular momentum between superfluid components and the observable crust. By modeling this glitch dynamics--including vortex motion, mutual friction, and angular momentum exchange--one may hope to probe the dense matter equation of state. In this work, we explore, within a highly idealized three-

Louise Wolswijk, Luca Cavicchioli, Giuseppe Vinelli, Mauro Chiarotti

Engineered ultracold atomic systems are a valuable platform for fundamental quantum mechanics studies and the development of quantum technologies. At near zero absolute temperature, atoms exhibit macroscopic phase coherence and collective quantum behavior, enabling their use in precision metrology, quantum simulation, and even information processing. This re

Nathaniel Johnston, Benjamin Lovitz, Vincent Russo, Jamie Sikora

The problem of quantum state classification asks how accurately one can identify an unknown quantum state that is promised to be drawn from a known set of pure states. In this work, we introduce the notion of $k$-learnability, which captures the ability to identify the correct state using at most $k$ guesses, with zero error. We show that deciding whether a

Nicole Hayes, Ekaterina Merkurjev, Guo-Wei Wei

Understanding the flexibility of protein-nucleic acid complexes, often characterized by atomic B-factors, is essential for elucidating their structure, dynamics, and functions, such as reactivity and allosteric pathways. Traditional models such as Gaussian Network Models (GNM) and Elastic Network Models (ENM) often fall short in capturing multiscale interact

Mutian He, Philip N. Garner

Linear-attention models that compress the entire input sequence into a fixed-size recurrent state offer an efficient alternative to Transformers, but their finite memory induces forgetfulness that harms retrieval-intensive tasks. To mitigate the issue, we explore a series of hybrid models that restore direct access to past tokens. We interleave token mixers

Deeksha Adil, Brian Bullins, Aaron Sidford, Chenyi Zhang

We develop optimization methods which offer new trade-offs between the number of gradient and Hessian computations needed to compute the critical point of a non-convex function. We provide a method that for any twice-differentiable $f\colon \mathbb R^d \rightarrow \mathbb R$ with $L_2$-Lipschitz Hessian, input initial point with $\Delta$-bounded sub-optimali

Alessandro Carones

Current and future Cosmic Microwave Background (CMB) experiments aim to achieve high-precision reconstruction of the CMB polarization signal, with the most ambitious objective being the detection of primordial $B$ modes sourced by cosmic inflation. Given the expected low amplitude of the signal, its estimate-parametrized by the tensor-to-scalar ratio $r$-is

Fares Fourati

Recent approaches to evaluating Artificial General Intelligence (AGI) typically summarize a system's capability using the arithmetic mean of its proficiencies across multiple cognitive domains. While simple, this implicitly assumes compensability: exceptional performance in some areas can offset severe deficiencies in others. Genuine general intelligence, ho

Anna Mészáros, Patrik Reizinger, Ferenc Huszár

Chess is a canonical example of a task that requires rigorous reasoning and long-term planning. Modern decision Transformers - trained similarly to LLMs - are able to learn competent gameplay, but it is unclear to what extent they truly capture the rules of chess. To investigate this, we train a 270M parameter chess Transformer and test it on out-of-distribu

Alicia Sagae, Chia-Jung Lee, Sandeep Avula, Brandon Dang

Current methods for evaluating large language models (LLMs) typically focus on high-level tasks such as text generation, without targeting a particular AI application. This approach is not sufficient for evaluating LLMs for Responsible AI dimensions like fairness, since protected attributes that are highly relevant in one application may be less relevant in

Hossein Safi, Mohammad Taghi Dabiri, Mazen Hasna, Iman Tavakkolnia

In this paper, we present a robust entanglement-assisted synchronization framework for indoor optical wireless systems that explicitly captures the coupling between spatial beam geometry and temporal synchronization accuracy. Unlike conventional approaches that treat beam steering and timing estimation independently, a unified spatio temporal model is develo

Wesley J. M. Ridgway, Mohit P. Dalwadi, Philip Pearce, S. Jonathan Chapman

Structured models, such as PDEs structured by age or phenotype, provide a setting to study pattern formation in heterogeneous populations. Classical tools to quantify the emergence of patterns, such as linear and weakly nonlinear analyses, pose significant mathematical challenges for these models due to sharply peaked or singular steady states. Here, we pres

Runzhe Zhan, Zhihong Huang, Xinyi Yang, Lidia S. Chao

Recent advancements in large reasoning models (LRMs) have introduced an intermediate "thinking" process prior to generating final answers, improving their reasoning capabilities on complex downstream tasks. However, the potential of LRMs as evaluators for machine translation (MT) quality remains underexplored. We provides the first systematic analysis of LRM

Rima Chatterjee

We give a complete classification of non-loose Legendrian Hopf links in $L(p,q)$ generalizing a result of the author with Geiges and Onaran. The classification is for non-loose Hopf links for both zero and non-zero Giroux torsion in their complement. We also give an explicit algorithm for the contact surgery diagrams for all these Legendrian representatives

Padmavathi Venkatraman, Sydney Erickson, Phil Marshall, Martin Millon

Strong gravitational lensing of active galactic nuclei (AGN) enables measurements of cosmological parameters through time-delay cosmography (TDC). With data from the upcoming LSST survey, we anticipate using a sample of O(1000) lensed AGN for TDC. To prepare for this dataset and enable this measurement, we construct and analyze a realistic mock sample of 130

Damian Owerko, Anna Scaglione, Alejandro Ribeiro

Training learning parameterizations to solve optimal power flow (OPF) with pointwise constraints is proposed. In this novel training approach, a learning parameterization is substituted directly into an OPF problem with constraints required to hold over all problem instances. This is different from existing supervised learning methods in which constraints ar

Binbin Huang, Haobin Duan, Yiqun Zhao, Zibo Zhao

We introduce Cupid, a generative 3D reconstruction framework that jointly models the full distribution over both canonical objects and camera poses. Our two-stage flow-based model first generates a coarse 3D structure and 2D-3D correspondences to estimate the camera pose robustly. Conditioned on this pose, a refinement stage injects pixel-aligned image featu

Anil K. Pradhan, Sultana N. Nahar

Accurate determination of opacity is critical for understanding radiation transport in both astrophysical and laboratory plasmas. We employ atomic data from R-Matrix calculations to investigate radiative properties in high-energy-density (HED) plasma sources. Specifically, we analyze environments such as the base of the convective zone (BCZ) of the Sun 2 x 1

Wenhao Wang, Kehe Ye, Xinyu Zhou, Tianxing Chen

Large-scale and diverse datasets are vital for training robust robotic manipulation policies, yet existing data collection methods struggle to balance scale, diversity, and quality. Simulation offers scalability but suffers from sim-to-real gaps, while teleoperation yields high-quality demonstrations with limited diversity and high labor cost. We introduce F

Elaine Cozzi, Nicholas Harrison, Zachary Radke

In their seminal work, Bourgain and Li establish strong ill-posedness of the 2D Euler equations for initial velocity in the critical Sobolev space $H^2(\mathbb{R}^2)$. In this work, we extend those results by demonstrating strong ill-posedness in logarithmically regularized spaces which are strictly contained in $H^2(\mathbb{R}^2)$ and which contain $H^s(\ma

Kai-Isaak Ellers, Marios Christodoulou, K. C. Schwab, K. Birgitta Whaley

We propose a laboratory-scale experiment to locally measure the general relativistic frame-dragging effect on Earth using the macroscopic quantum properties of a novel superfluid $^4$He single Josephson junction gyrometer. We derive the frame-dragging and related geodetic and Thomas effects in the superfluid gyrometer and present a procedure for their experi

Huijie Zhang, Aliaksandr Siarohin, Willi Menapace, Michael Vasilkovsky

MeanFlow has recently emerged as a powerful framework for few-step generative modeling trained from scratch, but its success is not yet fully understood. In this work, we show that the MeanFlow objective naturally decomposes into two parts: trajectory flow matching and trajectory consistency. Through gradient analysis, we find that these terms are strongly n

Wenchong Chen, Gennian Ge, Yang Shu, Zhouningxin Wang

Let $f_r(d,s_1,\ldots,s_r)$ denote the least integer $n$ such that every $n$-point set $P\subseteq\mathbb{R}^d$ admits a partition $P=P_1\cup\cdots\cup P_r$ with the property that for any choice of $s_i$-convex sets $C_i\supseteq P_i$ $(i\in[r])$ one necessarily has $\bigcap_{i=1}^r C_i\neq\emptyset$, where an $s_i$-convex set means a union of $s_i$ convex s

Austin Jia, Avaneesh Ramesh, Zain Shamsi, Daniel Zhang

Retrieval-Augmented Generation (RAG) has emerged as the dominant architectural pattern to operationalize Large Language Model (LLM) usage in Cyber Threat Intelligence (CTI) systems. However, this design is susceptible to poisoning attacks, and previously proposed defenses can fail for CTI contexts as cyber threat information is often completely new for emerg

Tomás Cabrera, Antonella Palmese, Maya Fishbach

While the LIGO/Virgo/KAGRA (LVK) gravitational wave (GW) detectors have detected over 300 binary black hole (BBH) mergers to date, the first confirmation of an electromagnetic (EM) counterpart to such an event remains elusive. Previous works have performed searches for counterpart candidates in transient catalogs and have identified active galactic nuclei (A

Pranamya Kulkarni, Puranjay Datta, Burak Varıcı, Emre Acartürk

Causal representation learning (CRL) has emerged as a powerful unsupervised framework that (i) disentangles the latent generative factors underlying high-dimensional data, and (ii) learns the cause-and-effect interactions among the disentangled variables. Despite extensive recent advances in identifiability and some practical progress, a substantial gap rema

Noam Issachar, Guy Yariv, Sagie Benaim, Yossi Adi

Diffusion Transformer models can generate images with remarkable fidelity and detail, yet training them at ultra-high resolutions remains extremely costly due to the self-attention mechanism's quadratic scaling with the number of image tokens. In this paper, we introduce Dynamic Position Extrapolation (DyPE), a novel, training-free method that enables pre-tr

Natalie Behague, Daniel Il'kovič, Richard Montgomery

In 2004, Kim and Vu conjectured that, when $d=\omega(\log n)$, the random $d$-regular graph $G_d(n)$ can be sandwiched with high probability between two random binomial graphs $G(n,p)$ with edge probabilities asymptotically equal to $\frac{d}{n}$. That is, there should exist $p_*=(1-o(1))\frac{d}{n}$, $p^*=(1+o(1))\frac{d}{n}$ and a coupling $(G_*,G,G^*)$ su

Andrew P. Bunger, Yunxing Lu, Ayyaz Mustafa, Michael M. McDowell

Quasi brittle materials such as rock and bone are understood to fail via microcrack coalescence. The accompanying Acoustic Emission (AE) event rate is known to increase as failure progresses. Here we examine the progression of the AE event rate for both rock and bone under conditions where failure progresses under fixed loading. The experiments for rock enta

David Itkin

We study a consumption-investment problem in a multi-asset market where the returns follow a generic rank-based model. Our main result derives an HJB equation with Neumann boundary conditions for the value function and proves a corresponding verification theorem. The control problem is nonstandard due to the discontinuous nature of the coefficients in rank-b

Jan Sobotka, Luca Baroni, Ján Antolík

Decoding visual stimuli from neural population activity is crucial for understanding the brain and for applications in brain-machine interfaces. However, such biological data is often scarce, particularly in primates or humans, where high-throughput recording techniques, such as two-photon imaging, remain challenging or impossible to apply. This, in turn, po

Antônio H. Ribeiro, David Vävinggren, Dave Zachariah, Thomas B. Schön

Adversarial training has emerged as a key technique to enhance model robustness against adversarial input perturbations. Many of the existing methods rely on computationally expensive min-max problems that limit their application in practice. We propose a novel formulation of adversarial training in reproducing kernel Hilbert spaces, shifting from input to f

Belle, Belle II Collaborations, :, M. Abumusabh

Using data samples of 988.4 fb$^{-1}$ and 427.9 fb$^{-1}$ collected with the Belle and Belle II detectors, we present a study of the singly Cabibbo-suppressed decays $\Xi_c^{0} \to \Lambda \eta$, $\Lambda \eta'$, and $\Lambda \pi^0$. We observe the decay $\Xi_c^0 \to \Lambda \eta$ and find evidence for the decay $\Xi_c^0 \to \Lambda \eta'$, with correspondin

Aristomenis Donos, Polydoros Kailidis

We study the nearly critical behaviour of holographic superfluids at finite temperature and chemical potential. Using analytic techniques in the bulk, we derive an effective theory for the long wavelength dynamics of gapless and pseudo-gapped modes, at first subleading order in a derivative expansion and we match the classical limit of our field theory const

Federico Bongiorno

We identify a class of singular algebraic foliations whose leaves through singular points retain regularity. The proof consists in showing existence of residual gerbes for certain formal stacks, which do not enjoy smooth presentations. As applications, we extend a theorem of Cerveau to the case where the ambient scheme is not smooth and we give a proof of th

Julia Wilkins, Jaehun Kim, Matthew E. P. Davies, Juan Pablo Bello

Music representations are the backbone of modern recommendation systems, powering playlist generation, similarity search, and personalized discovery. Yet most embeddings offer little control for adjusting a single musical attribute, e.g., changing only the mood of a track while preserving its genre or instrumentation. In this work, we address the problem of

Anurag Shekhar

Black men face a double barrier to mental health help-seeking: traditional masculinity norms demanding emotional restrictiveness and systemic racism fostering institutional mistrust. While celebrity mental health disclosures show promise for stigma reduction, limited research examines their impact on Black masculine communities through digital platforms. Thi

Alexander Gorsky, Ilya Liubimov

We investigate the phase structure of the deterministic and disordered versions of the Russian Doll Model (RDM), which is a generalization of Richardson model of superconductivity in a finite system with time-reversal symmetry breaking parameter $\theta$. It is one of the simplest examples of the cyclic RG where $\log N$ plays the role of the RG time. The de

Tobias Barker

In \cite{hou}, Hou gave a compelling numerical candidate for a singular solution of the 3D Navier-Stokes equations. We pioneer classifications of potentially singular solutions, motivated by the issue of investigating the viability of numerical candidates.For approximately axisymmetric initial data, we give the first quantitative classification of potentiall

Dylan Cant, Julio Sampietro Christ

This paper proves that certain monotone Lagrangians in the standard symplectic vector space cannot be displaced by a Hamiltonian isotopy which commutes with the antipodal map. The method of proof is to develop a Borel equivariant version of the quantum cohomology of Biran and Cornea, and prove it is sensitive to equivariant displacements. The Floer--Euler cl

Nima Torbati, Anastasia Meshcheryakova, Ramona Woitek, Diana Mechtcheriakova

Automated histopathological image analysis plays a vital role in computer-aided diagnosis of various diseases. Among developed algorithms, deep learning-based approaches have demonstrated excellent performance in multiple tasks, including semantic tissue segmentation in histological images. In this study, we propose a novel approach based on attention-driven

John Sengendo, Fabrizio Granelli

As we evolve towards more heterogeneous and cutting-edge mobile networks, Network Digital Twins (NDTs) are proving to be a promising paradigm in solving challenges faced by network operators, as they give a possibility of replicating the physical network operations and testing scenarios separately without interfering with the live network. However, with mobi

Harbir Antil

We study Maxwell's equations in conducting media with perfectly conducting boundary conditions on Lipschitz domains, allowing rough material coefficients and $L^2$-data. Our first contribution is a direct proof of well-posedness of the first-order weak formulation, including solution existence and uniqueness, an energy identity, and continuous dependence on

Jacob Reinheimer, Yuan Li, Trung Ha, Melanie Habouzit

Supermassive black holes (SMBHs) are known to correlate with many properties of their host galaxies, but we do not fully understand these correlations. The strengths (tightness) of these correlations have also been widely debated. In this work, we explore SMBH-host relations in three state-of-the-art cosmological simulations: Illustris, TNG, and EAGLE. Using

Aristomenis Donos, Polydoros Kailidis

We use standard techniques of hydrodynamics to construct a relativistic effective field theory for the low energy dynamics of nearly critical superfluids. In an appropriate non-relativistic limit, our theory predicts an additional coefficient when compared and contrasted to earlier work of Khalatnikov and Lebedev. In addition, we provide an alternative deriv

Jhon James Hernández-Sarria, João Paulo Silva Dias, Luciano Leonel Mendes, Nicolò Maccaferri

Multipole expansion methods have been primarily used for analyzing the electromagnetic scattering from non-magnetic isotropic dielectric scatterers, and studies about the scattering from magnetic objects seem to be lacking. In this work, we used the multipolar expansion framework for decomposing the electromagnetic scattering by dielectric particles with mag

Manus R. Visser, Zihan Yan

We develop a general framework for electromagnetic potential-charge contributions to the first law of black hole mechanics, applicable to dynamical first-order perturbations of stationary black objects with possibly non-compact bifurcate Killing horizons. Working in the covariant phase space formalism, we derive both comparison and physical process versions

Nesta Benno Joseph, Arka Bandyopadhyay, Ajit C. Balram, Awadhesh Narayan

The observation of non-linear Hall effects in time-reversal invariant systems has established the intriguing role of band topology beyond Berry curvature in determining transport phenomena. Many of these non-linear responses owe their origin to the Berry curvature dipole (BCD), which, like the Berry curvature (monopole), is also an electronic band structure

Annie Marsden, Liam O'Carroll, Aaron Sidford, Chenyi Zhang

We consider the problem of minimizing a $d$-dimensional Lipschitz convex function using a stochastic gradient oracle. We introduce and motivate a setting where the noise of the stochastic gradient is isotropic in that it is bounded in every direction with high probability. We then develop an algorithm for this setting which improves upon prior results by a f

Parinya Chalermsook, Ly Orgo, Minoo Zarsav

Ferrer dimension, along with the order dimension, is a standard dimensional concept for bipartite graphs. In this paper, we prove that a graph is of Ferrer dimension three (equivalent to the intersection bigraph of orthants and points in ${\mathbb R}^3$) if and only if it admits a biadjacency matrix representation that does not contain $\Gamma= \begin{bmatri

Emma Dugan, Xian-Yu Wang, Agustin Heron, Hareesh Gautham Bhaskar

Sub-Saturns have been reported to preferentially occupy near-polar orbits, but this conclusion has so far been based primarily on systems with cool host stars; obliquity measurements for sub-Saturns orbiting hot stars remain scarce. Expanding the census into the hot-star regime is essential to test whether the polar preference persists across the Kraft break

Ronghao Ni, Aidan Z. H. Yang, Min-Chien Hsu, Nuno Sabino

Program analysis tools often produce large volumes of candidate vulnerability reports that require costly manual review, creating a practical challenge: how can security analysts prioritize the reports most likely to be true vulnerabilities? This paper investigates whether machine learning can be applied to prioritizing vulnerabilities reported by program an

Albert Dorador

Radar charts are widely used to visualize multivariate data and compare multiple profiles across features. However, the visual clarity of radar charts can be severely compromised when feature values alternate drastically in magnitude around the circle, causing areas to collapse, which misrepresents relative differences. In the present work we introduce a per

Parinya Chalermsook, Ly Orgo, Minoo Zarsav

This paper considers the \textit{Zarankiewicz problem} in graphs with low-dimensional geometric representation (i.e., low Ferrers dimension). Our first result reveals a separation between bipartite graphs of Ferrers dimension three and four: while $Z(n;k) \leq 9n(k-1)$ for graphs of Ferrers dimension three, $Z(n;k) \in \Omega\left(n k \cdot \frac{\log n}{\lo

Tsai Hor Chan, Feng Wu, Yihang Chen, Guosheng Yin

Developing effective multimodal fusion approaches has become increasingly essential in many real-world scenarios, such as health care and finance. The key challenge is how to preserve the feature expressiveness in each modality while learning cross-modal interactions. Previous approaches primarily focus on the cross-modal alignment, while over-emphasis on th

Maneesh Bilalpur, Megan Hamm, Young Ji Lee, Natasha Norman

Storytelling is a powerful form of communication and may provide insights into factors contributing to gaps in healthcare outcomes. To determine whether Large Language Models (LLMs) can identify potential underlying factors and avenues for intervention, we performed topic-aware hierarchical summarization of narratives from African American (AA) storytellers.

Alyssa Gerhart, Balaji Iyangar

Adversarial attacks pose a severe risk to AI systems used in healthcare, capable of misleading models into dangerous misclassifications that can delay treatments or cause misdiagnoses. These attacks, often imperceptible to human perception, threaten patient safety, particularly in underserved populations. Our study explores these vulnerabilities through empi

Troy Criss, Ahmed Sidi El Valli, Naomi Li, Andrew Haas

We demonstrate a method to generate application-ready truly random bits from a magnetic tunnel junction driven by a Field-Programmable Gate Array (FPGA). We implement a real-time feedback loop that stabilizes the switching probability near 50\% and apply an XOR operation, both on the FPGA, to suppress short-term correlations, together mitigating long-term dr

Abhishek Bairwa, Ananthanarayanan Chockalingam

In this paper, we consider the problem of spread pilot design and effective channel estimation in multiple-input multiple-output Zak-OTFS (MIMO-Zak-OTFS) with superimposed spread pilots, where data and spread pilot signals are superimposed in the same frame. To achieve good estimation performance in a MIMO setting, the spread pilots at different transmit ant

Yujia Zheng, Zhuokai Zhao, Zijian Li, Yaqi Xie

Natural language has long enabled human cooperation, but its lossy, ambiguous, and indirect nature limits the potential of collective intelligence. While machines are not subject to these constraints, most LLM-based multi-agent systems still rely solely on natural language, exchanging tokens or their embeddings. To go beyond language, we introduce a new para

Pier Roberto Pastorino

The larger the Lefschetz defect delta(X) of a smooth complex Fano variety X, the more information we can deduce about the geometry of X. The structure of varieties with delta(X) greater than 2 is known. In this paper, we study the case delta(X)=2. In particular, we focus on Fano varieties with delta(X)=2 arising from the so called Casagrande-Druel constructi

Myrtille O. J. Y Hunault, Timothy G. Burrow, Fabien Besnard, Amélie Juhin

Resonant Inelastic X-ray scattering (RIXS) is a synchrotron-based spectroscopy that has seen growing interest across a range of scientific disciplines beyond fundamental physics. The interpretation of experimental RIXS data requires theoretical calculations based on the Kramers-Heisenberg formula. However, due to the dependence of RIXS on both the incident a

Jonathan Nemirovsky, Lee Peleg, Amit Ben Kish, Yotam Shapira

We investigate quantum circuits built from arbitrary single-qubit operations combined with programmable all-to-all multiqubit entangling gates that are native to, among other systems, trapped-ion quantum computing platforms. We report a constant-cost of no more than four applications of such Clifford entangling multiqubit gates to realize any sequence of Cli

Gozde Oney, Fayçal Adrar, Junhao Cao, Chunyang Zhang

The electrochemical performance of the spinel LiNi0.5Mn1.5O4, a high-voltage positive electrode material for Li-ion batteries, is influenced by the transition metal arrangement in the octahedral network, leading to disordered (Fd m S.G.) and ordered3 (P4332 S.G.) structures. However, widely used techniques lack the spatial resolution necessary to elucidate t

Xi He, Sirui Lu, Bei Zeng

Exact scientific discovery requires more than heuristic search: candidate constructions must be turned into exact objects and checked independently. We address this gap by extending TeXRA with an independent Lean 4 verification layer, turning it into a human-guided multi-agent platform for exact scientific discovery. The platform couples symbolic synthesis,

Xizhi Wu, Madeline S. Kreider, Philip E. Empey, Chenyu Li

Objective: Fluoropyrimidines are widely prescribed for colorectal and breast cancers, but are associated with toxicities such as hand-foot syndrome and cardiotoxicity. Since toxicity documentation is often embedded in clinical notes, we aimed to develop and evaluate natural language processing (NLP) methods to extract treatment and toxicity information. Mate

Jiacheng Chen, Ziyu Jiang, Mingfu Liang, Bingbing Zhuang

This paper proposes AutoScape, a long-horizon driving scene generation framework. At its core is a novel RGB-D diffusion model that iteratively generates sparse, geometrically consistent keyframes, serving as reliable anchors for the scene's appearance and geometry. To maintain long-range geometric consistency, the model 1) jointly handles image and depth in

Jasmine Bayrooti, Sattar Vakili, Amanda Prorok, Carl Henrik Ek

Thompson sampling (TS) is a powerful and widely used strategy for sequential decision-making, with applications ranging from Bayesian optimization to reinforcement learning (RL). Despite its success, the theoretical foundations of TS remain limited, particularly in settings with complex temporal structure such as RL. We address this gap by establishing no-re

Nannan Shi, Chuanyu Qin, Shipeng Song, Man Luo

Large language models (LLMs) have demonstrated strong reasoning capabilities in text-based mathematical problem solving; however, when adapted to visual reasoning tasks, particularly geometric problem solving, their performance substantially declines because geometric problems present unique challenges. Specifically, these challenges stem from two key factor

Milad Nourbakhsh, Kiernan E. Arledge, Vincent R. Whiteside, Jiangang Ma

Surface phonon polaritons (SPhPs) are promising candidates for enhanced light--matter interactions due to their efficient and low-loss light confinement features. In this work, we present unique light-matter interactions in saphhire within its Reststrahlen bands (RBs) across the long-wave infrared (LWIR) spectrum ($\omega = 385$-$1050~\mathrm{cm}^{-1}$). Par

Haiyang Wang

We study the growth of the support size of the capacity-achieving input distribution for the amplitude-constrained additive white Gaussian noise (AWGN) channel. While it is known since Smith (1971) that the optimal input is discrete with finitely many mass points, tight bounds on the number of support points $K_A$ as the amplitude constraint $A$ increases re

Xiaoyuan Wu, Roshni Kaushik, Wenkai Li, Lujo Bauer

Large language models (LLMs) are rapidly being adopted for tasks like drafting emails, summarizing meetings, and answering health questions. In these settings, users may need to share private information (e.g., contact details, health records). To evaluate LLMs' ability to identify and redact such information, prior work introduced real-life, scenario-ba

Carlos Román

We continue our study of the first critical field $H_{c_1}$ for extreme type-II superconductors governed by the three-dimensional magnetic Ginzburg--Landau functional with a pinning term $a_\varepsilon$, as introduced in our previous work [arXiv:2507.10915]. Building upon the lower bound for $H_{c_1}$ and the characterization of the Meissner solution, we now

Miguel Tierz

Within a fixed hyperangular channel $s>0$ of a harmonically trapped system, the $1/R^2$ perturbation is absorbed exactly into a shift of the channel parameter, $s\to s_\eta$, so the single-channel model remains a harmonic oscillator with a shifted inverse-square term: radial gaps stay at $2\hbar\omega$ exactly and no monopole spectral weight appears at forbi

Daniel Sorensen, Bappaditya Dey, Minjin Hwang, Sandip Halder

Semiconductor manufacturing is an extremely complex and precision-driven process, characterized by thousands of interdependent parameters collected across diverse tools and process steps. Multi-variate time-series analysis has emerged as a critical field for real-time monitoring and fault detection in such environments. However, anomaly prediction in semicon

Lucas Kania, Tudor Manole, Larry Wasserman, Sivaraman Balakrishnan

Many scientific applications involve testing theories that are only partially specified. This task often amounts to testing the goodness-of-fit of a candidate distribution while allowing for reasonable deviations from it. The tolerant testing framework provides a systematic way of constructing such tests. Rather than testing the simple null hypothesis that d

Ilya Chevyrev, Massimiliano Gubinelli

We derive a priori estimates for singular differential equations of the form \[ \mathcal{L} \phi = P(\phi,\nabla\phi) + f(\phi,\nabla\phi)\xi \] where $P$ is a polynomial, $f$ is a sufficiently well-behaved function, and $\xi$ is an irregular distribution such that the equation is subcritical. The differential operator $\mathcal L$ is either a derivative in

Patricio A. Gallardo, Yulin Gong, Boryana Hadzhiyska, Yun-Hsin Hsu

The peculiar motions of massive halos probe the distribution of matter in the universe, the gravitational potential, and the history of cosmic structure growth. The kinematic Sunyaev-Zeldovich (kSZ) effect offers a robust observational window into these properties. The pairwise kSZ estimator probes the pairwise momentum of groups of galaxies by cross-correla

Yaoting Zhang, Mikaella Brillantes, Justine Kuczera, Keyvan Ferasat

This study investigates interlayer diffusion dynamics in sodium montmorillonite (Na--MMT), a smectite clay widely used in environmental remediation, pharmaceutical formulations, and advanced materials. Understanding diffusion in Na--MMT is critical, yet current models often rely on fitted parameters rather than directly linking transport to microscopic struc

Fardin Ganjkhanloo, Emmett Springer, Erik H. Hoyer, Daniel L. Young

In this study we aim to better align fall risk prediction from the Johns Hopkins Fall Risk Assessment Tool (JHFRAT) with additional clinically meaningful measures via a data-driven modelling approach. We conducted a retrospective analysis of 54,209 inpatient admissions from three Johns Hopkins Health System hospitals between March 2022 and October 2023. A to

Evan Philip, Julius de Hond, Vytautas Abramavicius, Kaonan Micadei

Solving and optimizing differential equations (DEs) is ubiquitous in both engineering and fundamental science. The promise of quantum architectures to accelerate scientific computing thus naturally involved interest towards how efficiently quantum algorithms can solve DEs. Differentiable quantum circuits (DQC) offer a viable route to compute DE solutions usi

Peter Milonni

The energy shift of an oscillator in blackbody radiation is calculated based simply on the total energy of the interacting field-oscillator system as a function of the refractive index. For high temperatures T the energy and free-energy shifts are found to vary as -T^2 and +T^2, respectively, in agreement with the result originally obtained by Ford, Lewis, a

Şahsene Altınkaya, Sibel Yalçın

For the error functions of the form \begin{equation*} E_{r}\mathfrak{f}(z)=\frac{\sqrt{\pi z}}{2}er\ \mathfrak{f}(\sqrt{z})=z+\Sigma_{n=2}^{\infty} \frac{(-1)^{n-1}}{(2n-1)(n-1)!}z^{n}, \end{equation*}% let $\mathcal{E}S_{\mathcal{H}}(k,\lambda ,\gamma )\,$\ represent the class of harmonic error functions $\mathcal{ERF}=\mathcal{ERH}+\overline{\mathcal{% ERG

Haozhe Shan, Sun Minni, Lea Duncker

The ability to continually learn, retain and deploy skills to accomplish goals is a key feature of intelligent and efficient behavior. However, the neural mechanisms facilitating the continual learning and flexible (re-)composition of skills remain elusive. Here, we study continual learning and the compositional reuse of learned computations in recurrent neu

Samuel Soutullo, Miguel Yermo, David L. Vilariño, Óscar G. Lorenzo

3D LiDAR sensors are essential for autonomous navigation, environmental monitoring, and precision mapping in remote sensing applications. To efficiently process the massive point clouds generated by these sensors, LiDAR data is often projected into 2D range images that organize points by their angular positions and distances. While these range image represen

Xuyang Liu, Xiyan Gui, Yuchao Zhang, Linfeng Zhang

Recent large vision-language models (LVLMs) demonstrate remarkable capabilities in processing extended multi-modal sequences, yet the resulting key-value (KV) cache expansion creates a critical memory bottleneck that fundamentally limits deployment scalability. While existing KV cache compression methods focus on retaining high-importance KV pairs to minimiz