November 2025 arXiv papers — page 156

Patrick Gage Kelley, Steven Rousso-Schindler, Renee Shelby, Kurt Thomas

Generative AI (GenAI) is a powerful technology poised to reshape Trust & Safety. While misuse by attackers is a growing concern, its defensive capacity remains underexplored. This paper examines these effects through a qualitative study with 43 Trust & Safety experts across five domains: child safety, election integrity, hate and harassment, scams, and viole

Nicolò Forcillo, Jun Kitagawa, Russell W. Schwab

Here, we study Hamilton-Jacobi-Bellman equations on graphs. These are meant to be the analog of any of the following types of equations in the continuum setting of partial differential and nonlocal integro-differential equations: Hamilton-Jacobi (typically first order and local), Hamilton-Jacobi-Bellmann-Isaacs (first, second, or fractional order), and ellip

Drew Flieder

Many formal languages of contemporary mathematical music theory -- particularly those employing category theory -- are powerful but cumbersome: ideas that are conceptually simple frequently require expression through elaborate categorical constructions such as functor categories. This paper proposes a remedy in the form of a type-theoretic symbolic language

Lauren Benjamin Mushro

This paper examines gender stratification in the Latin American data annotation gig economy, with a particular focus on the "triple burden" shouldered by women: unpaid care responsibilities, economic precarity, and the volatility of platform-mediated labor. Data annotation, once lauded as a democratizing force within the global gig economy, has evolved into

Yicheng Zhan, Fahim Ahmed, Amy Burrell, Matthew J. Tonkin

Effective crime linkage analysis is crucial for identifying serial offenders and enhancing public safety. To address limitations of traditional crime linkage methods in handling high-dimensional, sparse, and heterogeneous data, we propose a Siamese Autoencoder framework that learns meaningful latent representations and uncovers correlations in complex crime

Mingrui Wang, Prakash Chakraborty

This paper develops fluid limits for nonstationary many-server loss systems with general service-time distributions. For the zero-buffer $M_t/G/n/n$ queuing model, we prove a functional strong law of large numbers for the fraction of busy servers and characterize the limit by a nonlinear Volterra integral equation with discontinuous coefficients induced by i

Pengfei Hu, Ming Fan, Xiaoxue Han, Chang Lu

Reservoir inflow prediction is crucial for water resource management, yet existing approaches mainly focus on single-reservoir models that ignore spatial dependencies among interconnected reservoirs. We introduce AdaTrip as an adaptive, time-varying graph learning framework for multi-reservoir inflow forecasting. AdaTrip constructs dynamic graphs where reser

K. Setoodehnia, J. H. Kelley

In this document, experimental nuclear structure data are evaluated for 8C. The details of each reaction populating 8C levels are compiled and evaluated. The combined results provide a set of adopted values that include level energies, spins and parities, level widths, decay types and branching ratios, and other nuclear properties. This work supersedes the e

Pujian Mao, Baijun Zeng

In this paper, we present the complete transformations of a generic metric from (generalized) harmonic to Newman-Unti coordinates up to the second post-Minkowskian order $(G^2)$. This allows us to determine the asymptotic shear, the Bondi mass aspect, and the angular-momentum aspect at both orders.

Liam Schmidt

The measurement of gamma-rays from decaying nuclei allow for the investigation into nuclear structure. True coincidence summing occurs when two gamma-rays from a single decay get detected in a single detector and their energies sum together to give false peaks in the energy spectrum. Corrections to this summing effect are crucial for an accurate determinatio

Tyler Slater

Context: Large Language Models (LLMs) rely on static, pre-deployment safety mechanisms that cannot adapt to adversarial threats discovered after release. Objective: To design a software architecture enabling LLM-based systems to autonomously detect safety failures and synthesize defense policies at runtime, without retraining or manual intervention. Method:

James K. Leung, Gemma E. Anderson, Alexander J. van der Horst, Claire Morley

We introduce the PanRadio Gamma-ray Burst (GRB) program carried out on the Australia Telescope Compact Array: a systematic, multi-year, radio survey of all southern $\textit{Swift}$ GRB events, comprehensively following the multi-frequency evolution of their afterglows from within an hour to years post-burst. We present the results of the 400-day observing c

Rukaiya Khatoon, Markus Boettcher, Joshua Robinson

We present a multi-wavelength study of a blazar PKS 0446+11, motivated by its spatial association with the neutrino event IC240105A detected by the IceCube Neutrino Observatory on 2024 January 5. The source is located 0.4 degrees from the best-fit neutrino direction and satisfies selection criteria for VLBI-selected, radio-bright AGN that have been identifie

Rafael Da C. Pereira, Túlio Vales, Cássio H. Vieira Morais

In this paper, we study the dynamics of set-valued maps whose graphs are open and such that the image of each point is an open and connected set. Building upon the work of P. Duarte and M. Torres, who introduced and analyzed the combinatorial structure of the final recurrent set, we incorporate the notions of transitivity and mixing, thereby bringing their f

Ratna Kandala, Katie Hoemann

Understanding emotional nuances in everyday language is crucial for computational linguistics and emotion research. While traditional lexicon-based tools like LIWC and Pattern have served as foundational instruments, Large Language Models (LLMs) promise enhanced context understanding. We evaluated three Dutch-specific LLMs (ChocoLlama-8B-Instruct, Reynaerde-

L. Hallacy, D. Hallett, A. Fenzl, N. J. Martin

We present a scalable method for electrically tuning multiple spatially separated quantum dots embedded in photonic crystal waveguides. Ion implantation into the top p-doped layer of a p-i-n diode creates high-resistivity tracks, providing electrical isolation between adjacent regions. Unlike physical etching, this method preserves the guided-mode profile of

Edward Bierstone, Dima Grigoriev, Pierre D. Milman, Jarosław Włodarczyk

Consider a projective variety $X \subset \mathbb{P}^n$ (over an algebraically closed field of characteristic zero), together with a (reduced) simple normal crossings divisor $E \subset \mathbb{P}^n$, where the degrees of both $X$ and $E$ are at most $d$. We show there is a pair $(n',d')$ which can be explicitly computed in terms of $(n,d)$, such that $(X,E)$

Jack Guo, Thomas Roy, Nitish Govindarajan, Joel B. Varley

Tandem catalysis involves two or more catalysts arranged in proximity within a single reaction vessel, with the aim of synergistically aligning the catalysts' reaction pathways to maximize overall system performance. This study presents a proof of concept showing the integration of continuum transport modeling with design optimization in a simplified two-dim

Ruihan Wu, Erchi Wang, Zhiyuan Zhang, Yu-Xiang Wang

Retrieval-augmented generation (RAG) enhances large language models (LLMs) by retrieving documents from an external corpus at inference time. When this corpus contains sensitive information, however, unprotected RAG systems are at risk of leaking private information. Prior work has introduced differential privacy (DP) guarantees for RAG, but only in single-q

Robin Delabays, Yuanzhao Zhang, Florian Dörfler, Giulia De Pasquale

Controllability determines whether a system's state can be guided toward any desired configuration, making it a fundamental prerequisite for designing effective control strategies. In the context of networked systems, controllability is a well-established concept. However, many real-world systems, from biological collectives to engineered infrastructures, ex

Matheus E. Pereira, Alexandre G. M. Schmidt

We present, for the first time, exact solutions for the Schr\"{o}dinger equation in Moon and Spencer's toroidal coordinates, and in the electromagnetic toroidal--poloidal coordinate systems. Curiously, both systems present a fractional angular momentum, because of the torus's hole. We achieve these novel solutions using the irregular $\mathcal{R}$-separation

Henry Adams, Florian Frick, Michael Harrison, Nikola Sadovek

Given a compact space $X$ that does not admit an embedding (an injective continuous function) into $\mathbb{R}^d$, we study the ''degree'' of discontinuity that any injective function $X \to \mathbb{R}^d$ must have. To this end, we define a scale invariant modulus of discontinuity and obtain general lower bounds, thus obtaining quantified nonembeddability re

K. Setoodehnia, J. H. Kelley, C. G. Sheu

In this document, experimental nuclear structure data are evaluated for 19Ne. 19Ne was first identified by (1939Fo01), see (2012Th01). The details of each reaction and decay experiment populating 19Ne levels are compiled and evaluated. The combined results provide a set of adopted values that include level energies, spins and parities, level half-lives, {\ga

Chadani Acharya

Course syllabi are often the first and sometimes only structured artifact that explains how a class will run: deadlines, grading rules, safety procedures, and how to request disability accommodations. For blind and low-vision (BLV) students who use screen readers, independent access depends on whether the syllabus is machine readable and navigable. We audite

Arya Bangun, Maximilian Töllner, Xuan Zhao, Christian Kübel

We introduce FlowTIE, a neural-network-based framework for phase reconstruction from 4D-Scanning Transmission Electron Microscopy (STEM) data, which integrates the Transport of Intensity Equation (TIE) with a flow-based representation of the phase gradient. This formulation allows the model to bridge data-driven learning with physics-based priors, improving

Robert Aloisi, Andrew Vanderburg, Melinda Soares-Furtado, Phillip Cargile

Observations over the past few decades have found that planets are common around nearby stars in our Galaxy, but little is known about planets that formed outside the Milky Way. We describe the design and early implementation of a survey to test whether planets also exist orbiting the remnant stars of ancient dwarf galaxies that merged with the Milky Way, an

Reymond Akpanya

In this paper, we study edge-transitive surfaces, i.e. triangulated 2-dimensional manifolds whose automorphism groups act transitively on the edges of these triangulated surfaces. We show that there exist four types of edge-transitive surfaces, splitting up further into a total of five sub-types. We exploit our theoretical results to compute a census of edge

Nadja Aldarondo Quiñones, Sydney Jenkins, Andrew Vanderburg, Melinda Soares-Furtado

Accurate stellar dating provides crucial information about the formation and development of planetary systems. Existing age-dating techniques are limited in terms of both the spectral type and age range they can accurately probe, and many are unreliable for stars older than 1 Gyr. Recent studies have suggested that a star's X-ray luminosity correlates strong

Yue Jin, Giovanni Montana

Offline multi-agent reinforcement learning (MARL) is severely hampered by the challenge of evaluating out-of-distribution (OOD) joint actions. Our core finding is that when the behavior policy is factorized - a common scenario where agents act fully or partially independently during data collection - a strategy of partial action replacement (PAR) can signifi

Niklas Ippisch, Markus Herklotz, Anna-Carolina Haensch, Carsten Schwemmer

Large language models (LLMs) have been proposed as scalable tools to address the gap between the importance of individualized written feedback and the practical challenges of providing it at scale. However, concerns persist regarding the accuracy, depth, and pedagogical value of their feedback responses. The present study investigates the extent to which LLM

Kartik Singh

Postnikov gave a parametrization for the totally non-negative Grassmannian using the matroid decomposition and associating a network with \reflectbox{L}-diagrams. Talaska and Williams extend this result to the entire Grassmannian by using the Deodhar decomposition instead of the matroid decomposition, and the networks this time are associated with the genera

Felipe Saenz, Joel Torres del Valle

This paper develops the basic theory of formal schemes over fields in the supersymmetric setting. We introduce the notion of a formal superscheme and investigate some of its fundamental properties. Particular emphasis is placed on the study of morphisms between formal superschemes, for which we establish a faithfully flat descent theorem and a fiber dimensio

Tcharlies Schmitz

This paper introduces the Harmonic Token Projection (HTP), a reversible and deterministic framework for generating text embeddings without training, vocabularies, or stochastic parameters. Unlike neural embeddings that rely on statistical co-occurrence or optimization, HTP encodes each token analytically as a harmonic trajectory derived from its Unicode inte

Ola Kenji Forslund, Chin Shen Ong, Moritz M. Hirschmann, Nicolas Gauthier

A major challenge in condensed matter physics is integrating topological phenomena with correlated electron physics to leverage both types of states for next-generation quantum devices. Metal-insulator transitions (MITs) are central to bridging these two domains while simultaneously serving as 'on-off' switches for electronic states. Here, we demonstrate how

K. Setoodehnia, J. H. Kelley, C. G. Sheu

In this document, experimental nuclear structure data are evaluated for 18Ne for ENSDF. 18Ne was first identified by (1954Go17), see (2012Th01). The details of each reaction and decay experiment populating 18Ne levels are compiled and evaluated. The combined results provide a set of adopted values that include level energies, spins and parities, level half-l

William Adkisson, Omer Ben Neria

We study the strength of well-founded ultrafilters on ordinals above choiceless large cardinals and their associated Prikry forcings. Gabriel Goldberg showed that all but boundedly many regular cardinals above a rank Berkeley cardinal carry well-founded uniform ultrafilters. We prove several bounds on the large cardinal strength that is witnessed by such ult

Matus Papajcik, Jiri Podolsky

We present a new effective method of algebraic classification of 2+1 geometries. Our approach simply classifies spacetimes using five real scalars, defined as specific projections of the Cotton tensor onto a suitable null basis. The algebraic type of a spacetime is determined by gradual vanishing of these scalars. We derive the Bel-Debever criteria, together

Nikhil Bachhawat, Vladimir N. Litvinenko, Jean C. Brutus, Luca Cultrera

SRF CW accelerator constructed for Coherent electron Cooling (CeC) Proof-of-principle (POP) experiment at Brookhaven National Laboratory has frequently demonstrated record parameters using 1.5 nC 350 ps long electron bunches, typically compressed to FWHM of 30 ps using ballistic compression. We report experimental demonstration of CW electron beam with param

Luca Macesanu, Boueny Folefack, Samik Singh, Ruchira Ray

Multimodal audiovisual perception can enable new avenues for robotic manipulation, from better material classification to the imitation of demonstrations for which only audio signals are available (e.g., playing a tune by ear). However, to unlock such multimodal potential, robots need to learn the correlations between an object's visual appearance and the so

S Pal, L. M. Woods, C. Persson, I. Brevik

Controlling nanoscale interactions to suppress aggregation from short-range attractive forces is a key problem in nanoengineering. Here, we demonstrate a route to modulate Casmir-Lifshitz interactions between anisotropic nanoparticles with the magnetic fluids. By semi-classical quantum electrodynamics, we study ground state dispersion forces for cylindrical

Szilárd Szalay

We construct an important missing piece in the entanglement theory of pure three-qubit states, which is a polynomial measure of W-entanglement, working in parallel to the three-tangle, which is a polynomial measure of GHZ-entanglement, and to the bipartite concurrence, which is a polynomial measure of bipartite entanglement. We also show that these entanglem

Benjamin Rodenberg, Benjamin Uekermann

Many multiphysics simulations involve processes evolving on disparate time scales, posing a challenge for efficient coupling. A naive approach that synchronizes all processes using the smallest time scale wastes computational resources on slower processes and typically achieves only linear convergence in time. Waveform iteration is a promising numerical tech

Martin Miglioli

The present work develops a framework to derive piecewise polynomial measures arising from invariant measures on adjoint orbits in the context of compact and semisimple Lie groups. These measures are computed from orbital integrals via transformations on spaces of polynomials endowed with the apolar inner product. In the case of the unitary group, we obtain

Jiajie Luo, Gregory Henselman-Petrusek

We study pointwise free and finitely-generated persistence modules over a principal ideal domain, indexed by a (possibly infinite) totally-ordered poset category. We show that such persistence modules admit interval decompositions if and only if every structure map has free cokernel. We also show that, in torsion-free settings, the integer persistent homolog

Danko R. Jocić, Mihailo Krstić, Milan Lazarević, Stevan Milašinović

Let $q, r, s \geqslant 1$ satisfy $\frac{1}{2q} + \frac{1}{2r} = \frac{1}{s}$ and $X \in \mathcal{C}_s(\mathcal{H})$. If $(\lambda_n)_{n=1}^{\infty}, (w_n)_{n=1}^{\infty}$ are sequences in $(0,+\infty)$ and $(\lambda_n^{(1-q)/(2q)} A_n)_{n=1}^{\infty}$, $(\lambda_n^{1/(2q)} A_n^*)_{n=1}^{\infty}, (w_n^{-1/(2r)} B_n)_{n=1}^{\infty}$ and $(w_n^{(r-1)/(2r)} B_n

Samuel W. Flint, Jigyasa Chauhan, Niloofar Mansoor, Bonita Sharif

Modern programming languages, such as Python, support language features from several paradigms, such as object-oriented, procedural, and functional. Research has shown that code written in some paradigms can be harder to comprehend, but to date, no research has looked at which paradigm-specific language features impact comprehension. To this end, this study

Dale R. Worley

The celebrated Robinson-Schensted algorithm and each of its variants that have attracted substantial attention can be constructed using Fomin's "growth diagram" construction from a modular lattice that is also a weighted-differential poset. We classify all such lattices that meet certain criteria; the main criterion is that the lattice is distributive. Intui

Sabarinath Ramachandran, Venkatesh Khammammetti, Prasanthi Maddala, Narayan Mandayam

Millimeter-wave (mmWave) communication offers vast bandwidth for next-generation wireless systems but faces severe path loss, Doppler effects, and hardware impairments. Orthogonal Time Frequency Space (OTFS) modulation has emerged as a robust waveform for high-mobility and doubly dispersive channels, outperforming OFDM under strong Doppler. However, the most

Nikos I. Karachalios, Dionyssios Mantzavinos, Jeffrey Oregero

The distance between the solutions to the integrable Korteweg-de Vries (KdV) equation and a broad class of non-integrable generalized KdV (gKdV) equations is estimated in appropriate Sobolev spaces. This family of equations includes, as special cases, the standard gKdV equation with power nonlinearities as well as weakly nonlinear perturbations of the KdV eq

Siiri Kivimäki

We show that the existence of a universal countably chromatic graph of size $\aleph_1$ together with the failure of continuum hypothesis is consistent. The proof is a forcing iteration of strongly proper ccc posets. The construction works for any uncountable successor cardinal $\kappa^+$, where $\kappa$ is regular.

Rui Han, Wilhelm Schlag

In this paper, we prove H\"older continuity of the integrated density of states for discrete quasiperiodic Jacobi $d\times d$ block matrices with Diophantine frequencies. The H\"older exponent is shown to be any $\beta$ such that $0<\beta<1/(2\kappa^d)$, where $\kappa^d$ is the acceleration, i.e., the slope of the sum of the top $d$ Lyapunov exponents in the

Tianxiong Han, Tyler J. Slade, Liqin Ke, Qing-Ping Ding

Kagome metals can host unconventional electronic phenomena that emerge from their frustrated lattice geometry and associated band topology. Correlated electronic orders, such as charge-density waves and superconductivity, are observed to intertwine with subtle time-reversal symmetry breaking whose microscopic origin is not currently understood. Here, we prov

Dong Xie, Chao Gao, John Lafferty

We study confidence interval construction for linear regression under Huber's contamination model, where an unknown fraction of noise variables is arbitrarily corrupted. While robust point estimation in this setting is well understood, statistical inference remains challenging, especially because the contamination proportion is not identifiable from the data

Halyun Jeong, Palle E. T. Jorgensen, Hyun-Kyoung Kwon, Myung-Sin Song

We present a variety of projection-based linear regression algorithms with a focus on modern machine-learning models and their algorithmic performance. We study the role of the relaxation parameter in generalized Kaczmarz algorithms and establish a priori regret bounds with explicit $\lambda$-dependence to quantify how much an algorithm's performance deviate

Georgiy Shakirov, Albert Arakelov

A common practice in heterogeneous graph neural networks (HGNNs) is to condition parameters on node/edge types, assuming types reflect semantic roles. However, this can cause overreliance on surface-level labels and impede cross-type knowledge transfer. We explore integrating Mixture-of-Experts (MoE) into HGNNs--a direction underexplored despite MoE's succes

J. P. Pridham

We extend the author's and CPTVV's correspondence between shifted symplectic and Poisson structures to establish a correspondence between exact shifted symplectic structures and non-degenerate shifted Poisson structures with formal derivation, a concept generalising constructions by De Wilde and Lecomte. Our formulation is sufficiently general to encompass d

Louigi Addario-Berry, Emma Hogan, Lukas Michel, Alex Scott

It is well-known that any finite triangulation possesses a unique maximal Schnyder wood. We introduce Schnyder woods of infinite triangulations, and prove there exists a unique maximal Schnyder wood of any infinite triangulation with finite boundary, and of the uniform infinite half-planar triangulation. Furthermore, the maximal Schnyder wood of the uniform

Swaroop Panda

Personal heart rate data from wearable devices contains rich information, yet current visualizations primarily focus on simple metrics, leaving complex temporal patterns largely unexplored. We present a speculative exploration of personal heart rate visualization possibilities through five prototype approaches derived from established visualization literatur

The CHROMA Collaboration, R. Kord Valeshabadi, S. Rezaie, K. Azizi

We perform an in-depth comparative analysis of unintegrated parton distribution function (UPDF) models for isolated prompt diphoton production in proton-proton collisions at $\sqrt{s}=13$~TeV within the $k_t$-factorization framework. Predictions are obtained with three UPDF approaches: Parton Branching (PB), NLO-MRW, and Modified KMRW (MKMRW). Tree-level $q

Guanlin Zhu, Zechun Deng, Jiaxin Shen, Junchi Yang

Abandoned oil and gas wells pose significant environmental risks due to the potential leakage of hydrocarbons, brine and chemical pollutants. Detecting such leaks remains extremely challenging due to the weak acoustic emission and high ambient noise in the deep sea. This paper reviews the application of passive sonar systems combined with artificial intellig

Eloisa Detomi, Robert M. Guralnick, Marta Morigi, Pavel Shumyatsky

Given two subgroups $H,K$ of a finite group $G$, the probability that a pair of random elements from $H$ and $K$ commutes is denoted by $Pr(H,K)$. Suppose that a finite group $G$ admits a group of coprime automorphisms $A$ and let $\epsilon>0$. We show that, if for any distinct primes $p,q\in\pi(G)$ there is an $A$-invariant Sylow $p$-subgroup $P$ and an $A$

J. A. Alzate-Trujillo, A. del Pino, C. López-Sanjuan, A. Hidalgo

Wide-field, multi-filter photometric surveys enable the reconstruction of the Milky Way's star formation history (SFH) on Galactic scales and offer new insights into disc assembly. The twelve-filter system of the Javalambre Photometric Local Universe Survey (J-PLUS) is particularly suitable, as its colours trace stellar chemical abundances and help alleviate

Özay Ezerceli, Gizem Gümüşçekiçci, Tuğba Erkoç, Berke Özenç

In this work, we introduce TurkEmbed4Retrieval, a retrieval specialized variant of the TurkEmbed model originally designed for Natural Language Inference (NLI) and Semantic Textual Similarity (STS) tasks. By fine-tuning the base model on the MS MARCO TR dataset using advanced training techniques, including Matryoshka representation learning and a tailored mu

Leonardo Abbrescia, Pieter Blue, Jan Sbierski, Jared Speck

We study the Cauchy problem for classical and weak shock-forming solutions to a model quasilinear wave equation in $1+1$ dimensions arising from a convenient choice of $C^{\infty}$ initial data, which allows us to solve the equation using elementary arguments. The simplicity of our model allows us to succinctly illustrate various phenomena of geometric and a

Amr Akmal Abouelmagd, Amr Hilal

The emergence of crowdsourced data has significantly reshaped social science, enabling extensive exploration of collective human actions, viewpoints, and societal dynamics. However, ensuring safe, fair, and reliable participation remains a persistent challenge. Traditional polling methods have seen a notable decline in engagement over recent decades, raising

Wei Hou Tan, River Robles, Juan Hernandez, Emilio Alessandro Nanni

One pathway to producing high brightness electron beams is to use a radio-frequency (rf) driven high field photogun to rapidly accelerate photoemitted electrons to the relativistic regime and preserve the brightness. However, the highest attainable field is limited by rf breakdowns of materials used in a photogun. Shortening rf pulse duration feeding into a

Eric R. Van Clepper, Felipe Alarcón, Edwin Bergin, Fred J. Ciesla

Giant planet atmospheres are thought to reflect the gas phase composition of the disk when and where they formed. However, these atmospheres may also be polluted via solid accretion or ice sublimation in the disk. Here, we propose a novel mechanism for enriching the atmospheres of these giant planets with volatiles via pebble drift, fragmentation, and ice su

Lisa Mandal, Md. Ali Zinna

We prove that every local complete intersection curve in $Spec(A)$, where $A$ is a commutative Noetherian ring of dimension three, is a set-theoretic complete intersection. An analogous result is established for local complete intersection surfaces when $A$ is a four-dimensional affine algebra over the algebraic closure of a finite field of $p$ elements. Fur

Soumyabrata Bose, Jay Bartroff

This paper develops a framework for incorporating prior information into sequential multiple testing procedures while maintaining asymptotic optimality. We define a weighted log-likelihood ratio (WLLR) as an additive modification of the standard LLR and use it to construct two new sequential tests: the Weighted Gap and Weighted Gap-Intersection procedures. W

Shreyas Rajesh, Pavan Holur, Chenda Duan, David Chong

Large Language Models (LLMs) face fundamental challenges in long-context reasoning: many documents exceed their finite context windows, while performance on texts that do fit degrades with sequence length, necessitating their augmentation with external memory frameworks. Current solutions, which have evolved from retrieval using semantic embeddings to more s

Samuel Audia, Dinesh Manocha, Matthias Zwicker

We introduce a Monte Carlo integration-based Shooting and Bouncing Ray (SBR) algorithm for electromagnetic scattering, specifically targeting complex dielectric materials. Unlike traditional deterministic SBR methods, our approach is the first to reformulate the SBR integral equations using Monte Carlo techniques and advanced variance reduction strategies ad

Tamas Almos Vami

Searching for dark matter (DM) at colliders is one of the biggest challenges in high-energy physics today. Significant efforts have been made to detect DM within the mass range of 1-10,000 GeV at the Large Hadron Collider and other experiments. However, the lower mass range of 0.001-1 GeV remains largely unexplored, despite strong theoretical motivation from

Raffi Khatchadourian, Rolando Franco

Financial institutions deploy Large Language Models (LLMs) for reconciliations, regulatory reporting, and client communications, but nondeterministic outputs (output drift) undermine auditability and trust. We quantify drift across five model architectures (7B-120B parameters) on regulated financial tasks, revealing a stark inverse relationship: smaller mode

Wuyang Zhang, Chenkai Zhang, Zhen Luo, Jianming Ma

Large language models (LLMs) have transformed software development by enabling automated code generation, yet they frequently suffer from systematic errors that limit practical deployment. We identify two critical failure modes: \textit{logical hallucination} (incorrect control/data-flow reasoning) and \textit{schematic hallucination} (type mismatches, signa

Wadim Strielkowski

This paper focuses on assessing the potential of micro-credentials in digitalized higher education in the era of artificial intelligence. Micro-credentials are commonly described as mini qualifications or digital badges that certify an individual competency in a specific skill or cluster of skills. Being different from traditional academic credentials in bot

Rafael D. Benguria, Juan Manuel González-Brantes

We prove an upper bound on the excess charge for non-relativistic atomic systems, independent of the particle statistics. This result improves Lieb's bound of 1984 for any $Z \ge 12$.

Supriti Vijay, Aman Priyanshu, Anu Vellore, Baturay Saglam

Effective information retrieval requires reasoning over partial evidence and refining strategies as information emerges. Yet current approaches fall short: neural retrievers lack reasoning capabilities, large language models (LLMs) provide semantic depth but at prohibitive cost, and query rewriting or decomposition limits improvement to static transformation

Daniel F. Calero-Osorio, Paul W. Ayers

We show how to add the effects of residual electron correlation to a reference seniority-zero wavefunction by making a unitary transformation of the true electronic Hamiltonian into seniority-zero form. The transformation is treated via the Baker Campbell Hausdorff (BCH) expansion and the seniority-zero structure of the reference is exploited to evaluate the

Vishesh Sharma, Gaurav Gautam, Poonam Yadav, Chin-Wei Wang

We present a comprehensive study of the magnetic structure evolution across the spin reorientation transition in orthorhombic (Pnma) TmCrO3. Magnetic susceptibility reveals canted antiferromagnetic (CAFM) ordering at T_N = 125 K, two compensation points (T_comp1 and T_comp2), followed by magnetization reversal with a magnetic susceptibility minimum between T

Avishai Dekel, Dhruba Dutta Chowdhury, Sharon Lapiner, Zhiyuan Yao

We address the origin of the Little Red Dots (LRDs) seen by JWST at cosmic morning ($z \!=\! 4 \!-\! 8$) as compact stellar systems with over-massive black holes (BHs). We propose that LRDs form naturally after feedback-free starbursts (FFB) in thousands of star clusters and following wet compaction. Analytically, we show how the clusters enable efficient dr

Yining Lu, Wenyi Tang, Max Johnson, Taeho Jung

Existing retrieval-augmented generation (RAG) systems typically use a centralized architecture, causing a high cost of data collection, integration, and management, as well as privacy concerns. There is a great need for a decentralized RAG system that enables foundation models to utilize information directly from data owners who maintain full control over th

James M. Miley, Grant M. Kennedy, Alvaro Ribas, Enrique Macias

Intermediate-mass stars (1-3 Msun) host some of the brightest and best-studied debris discs. This stellar class is also the most frequent host of molecular gas in systems older than typical protoplanetary disc lifetimes, and the most likely to host detected giant planets in radial-velocity surveys. The debris discs of intermediate-mass stars have therefore b

Anurag Singh, Nitu Kumari, Arun Kumar

In this study, we investigate the prey predator dynamics of the elk wolf system in northern Yellowstone National Park, USA, using a data driven modeling approach. We used yearly population data for elk and wolves from 1995 to 2022 to construct a mathematical model using a sparse regression modeling framework. To the best of our knowledge, no previous work ha

Vasilis Bountris, Lauritz Thamsen, Ulf Leser

Provenance plays a crucial role in scientific workflow execution, for instance by providing data for failure analysis, real-time monitoring, or statistics on resource utilization for right-sizing allocations. The workflows themselves, however, become increasingly complex in terms of involved components. Furthermore, they are executed on distributed cluster i

Kamand Kalashi, Babak Teimourpour

The rapid expansion of online fashion platforms has created an increasing demand for intelligent recommender systems capable of understanding both visual and textual cues. This paper proposes a hybrid multimodal deep learning framework for fashion recommendation that jointly addresses two key tasks: outfit compatibility prediction and complementary item retr

Sean P. Fillingham, Andrew Gordon, Peter Lai, Xavier Poncini

Mechanistic interpretability aims to decompose neural networks into interpretable features and map their connecting circuits. The standard approach trains sparse autoencoders (SAEs) on each layer's activations. However, SAEs trained in isolation don't encourage sparse cross-layer connections, inflating extracted circuits where upstream features needlessly af

Chen Jin, Ankush Agarwal

Diffusion Probabilistic Model (DDPM) for generating one-day-ahead arbitrage-free implied volatility surfaces. To capture the path-dependent nature of volatility dynamics, we condition our model on a set of market variables, including exponentially weighted moving averages (EWMAs) of historical vol-surfaces, returns and squared returns of the underlying asset

Sam Bastida, Nick Brettell, Rutger Campbell, George Drummond

Mayhew et al.\ (2021) posed the problem of showing that the minor-closed class of spikes and their minors has a finite set of excluded minors and describing all of them. In this paper, we resolve this problem.

David Simmons-Duffin, Yixin Xu

We explore a "genus-2" crossing equation obeyed by CFTs in general dimensions $d\geq 2$. This crossing equation relates two different decompositions of the "genus-2 partition function" -- namely the partition function on the connected sum $M_2=(S^1\times S^{d-1})\sharp (S^1\times S^{d-1})$. The "sunrise" channel decomposition expresses $M_2$ as a pair of thr

Vincent Hsiao, Mark Roberts, Leslie Smith

Large language models (LLMs) often struggle when performing agentic tasks without substantial tool support, prom-pt engineering, or fine tuning. Despite research showing that domain-dependent, procedural knowledge can dramatically increase planning efficiency, little work evaluates its potential for improving LLM performance on agentic tasks that may require

Ujunwa Mgboh, Rafi Sultan, Dongxiao Zhu, Joshua Kim

Accurate fluence map prediction is essential in intensity-modulated radiation therapy (IMRT) to maximize tumor coverage while minimizing dose to healthy tissues. Conventional optimization is time-consuming and dependent on planner expertise. This study presents a deep learning framework that accelerates fluence map generation while maintaining clinical quali

Adamu Issifu, Prashant Thakur, Davood Rafiei Karkevandi, Franciele M. da Silva

Neutron stars (NSs) provide a unique laboratory to probe dark matter (DM) through its gravitational imprint on stellar evolution. We use a two-fluid framework with non-annihilating, asymmetric DM, both fermionic and bosonic, that interacts with ordinary matter (OM) solely through gravity. Within this framework, we track protoneutron stars (PNSs) across their

Christopher Roth, Ao Chen, Anirvan Sengupta, Antoine Georges

Whether the ground state of the square lattice Hubbard model exhibits superconductivity remains a major open question, central to understanding high temperature cuprate superconductors and ultra-cold fermions in optical lattices. Numerical studies have found evidence for stripe-ordered states and superconductivity at strong coupling but the phase diagram rem

Calin Galeriu

We derive the formula for the self-inductance of an ideal solenoid by calculating the total kinetic energy associated with the drift velocity of the conduction electrons.

Nuno Soares, António Grilo

This thesis presents the development of ARGUS, a framework for mission planning for Unmanned Ground Vehicles (UGVs) in tactical environments. The system is designed to translate battlefield complexity and the commander's intent into executable action plans. To this end, ARGUS employs a processing pipeline that takes as input geospatial terrain data, military

Trishit Mondal, Ricardo Vinuesa, Ameya D. Jagtap

Active flow control of compressible transonic shock-boundary layer interactions over a two-dimensional RAE2822 airfoil at Re = 50,000 is investigated using deep reinforcement learning (DRL). The flow field exhibits highly unsteady dynamics, including complex shock-boundary layer interactions, shock oscillations, and the generation of Kutta waves from the tra

Thinh Huu Nguyen, Kirk S. S. Barrow, Susie Byrom, Varun Satish

Accurately modeling molecular hydrogen ($\text{H}_{2}$) is an important task in cosmological simulations because it regulates star formation. One fundamental property of $\text{H}_{2}$ is the ability to self-shield, a phenomenon in which the $\text{H}_{2}$ in the outer layer of a molecular cloud absorbs the photodissociating Lyman-Werner UV radiation and shi

Giovanni Tedeschi-Prades, Til Birnstiel, Klaus Dolag, Barbara Ercolano

The integration of aerodynamic drag is a fundamental step in simulating dust dynamics in hydrodynamical simulations. We propose a novel integration scheme, designed to be compatible with Strang splitting techniques, which allows for the straightforward integration of external forces and hydrodynamic fluxes in general-purpose hydrodynamic simulation codes. Mo

C. Lazzoni, R. Bendahan-West, S. Marino, K. D. Lawson

We present the first observations of the HD92945 debris disk obtained with JWST, targeting this nearby K0V star located at 21.54 pc from the Sun. High-contrast coronagraphic imaging was performed using JWST/NIRCam in the F200W and F444W filters. After subtracting the disk contribution through forward modeling by means of synthetic PSFs and MCMC optimizations

Saya Hashemian, Azam Asilian Bidgoli

In computational pathology, the gigapixel scale of Whole-Slide Images (WSIs) necessitates their division into thousands of smaller patches. Analyzing these high-dimensional patch embeddings is computationally expensive and risks diluting key diagnostic signals with many uninformative patches. Existing patch selection methods often rely on random sampling or