March 2026 arXiv papers — page 12

Hugo Molinares, Guillermo Romero, Victor Montenegro, Vitalie Eremeev

Quantum mechanical oscillators operating at frequencies up to the GHz regime have been predicted to support phonon lasing -- self-sustained coherent vibrational motion emerging when the effective gain exceeds intrinsic losses. Current phonon-laser proposals face two key limitations, namely: they lack scalability and rely on coupling all oscillators to a comm

Katie L. Y. Zhou, Anna T. Bui, Stephen J. Cox

Fluid mixtures that exhibit an azeotrope cannot be purified by simple bulk distillation. Consequently, there is strong motivation to understand the behavior of azeotropic mixtures under confinement. We address this problem using an ML-enhanced classical density functional theory (cDFT) applied to a binary Lennard-Jones mixture that exhibits azeotropic phase

Yujun Yao, Luming Sun, Tao Wu, Ning Jiang

Statistics on tidal disruption events (TDEs) may be contaminated by repeating TDEs (rTDEs), which have been extensively discovered recently. However, the origin of rTDEs remains unclear. In addition, no statistical research on rTDEs with time intervals $>5$ years has been made yet. In this work, we searched for rTDEs with time intervals of 5--19 years using

Sayan Ghosh, Gourab Roy, Ekta Kushwaha, Mohit Kumar

Here we have experimentally demonstrated spin-correlation-driven ferroelectric quantum criticality in a prototype quantum spin-liquid system, Ba3CuSb2O9, a quantum phenomenon rarely observed. The dielectric constant follows a clear T2 scaling, showing that the material behaves as a quantum paraelectric without developing ferroelectric order. Magnetically, th

Jérémy Itier, Gilles Renversez, Frédéric Zolla

We present a perturbative technique for modeling the scattering of light by a nonlinear material. This approach eliminates the need for an iterative algorithm to solve the fully coupled nonlinear problem. We demonstrate its effectiveness in the cases of a nonlinear anisotropic slab and a nonlinear periodic crystal, both illuminated by a plane wave under coni

Rong Jin, Guangyao Wang, Xingsheng Sun

High-fidelity simulations are essential for predicting material behavior under high-velocity impact (HVI), but their accuracy depends on material models and parameters that are often calibrated by manual fitting to multiple costly experiments. In this study, we develop an ensemble-based data assimilation framework for automatic calibration of selected plasti

J. Fournier, P. -O. Downey, O. Gingras, C. -D. Hébert

Asymptotically close to critical end-points of first-order transitions, maxima in thermodynamic quantities occur along a line called the Widom line, a concept first introduced in classical fluids. This concept has been extended to strongly correlated electronic fluids in the context of the Mott transition. Namely, upon increasing interaction strength in the

Till Bargheer, Albert Bekov, Carlos Bercini, Frank Coronado

We construct generating functions of five- and six-point correlators up to two loops at weak 't Hooft coupling in planar N=4 SYM. These generating functions unify the correlators of the lightest scalar operator in the stress-tensor multiplet with those of all higher R-charge single-trace half-BPS scalar operators, thereby extending previous results for f

Anthony Chen, Robert Krasny

This work describes a new version of the Fast Multipole Method for summing pairwise particle interactions that arise from discretizing integral transforms and convolutions on the sphere. The kernel approximations use barycentric Lagrange interpolation on a quadtree composed of cubed sphere grid cells. The scheme is kernel-independent and requires kernel eval

Ali Zahra, Jerome Dubail, Gunter M. Schütz

We study the symmetric Dyson exclusion process (SDEP) - a lattice gas with exclusion and long-range, Coulomb-type interactions that emerge both as the maximal-activity limit of the symmetric exclusion process and as a discrete version of Dyson's Brownian motion on the unitary group. Exploiting an exact ground-state (Doob) transform, we map the stochastic

Matthew Dickson, Yucheng Liu

We study a percolation model on $\mathbb R^d$ called the random connection model. For $d$ large, we use the lace expansion to prove that the critical two-point connection probability decays like $|x|^{-(d-2)}$ as $|x| \to \infty$, with possible anisotropic decay. Our proof also applies to nearest-neighbour Bernoulli percolation on $\mathbb Z^d$ in $d \ge 11$

Vic Vander Linden, Boris De Vos, Kevin Vervoort, Frank Verstraete

The quantum $5$-state Potts model is known to possess a perturbative description using complex conformal field theory (CCFT), the analytic continuation of ``theory space" to a complex plane. To study the corresponding complex fixed point on the lattice, the model must be deformed by an additional non-Hermitian term due to its complex coefficient $λ$. Alt

Yaniv Shulman

The identification of Linear Time-Varying (LTV) systems from input-output data is a fundamental yet challenging ill-posed inverse problem. This work introduces a unified Bayesian framework that models the system's impulse response, $h(t, τ)$, as a stochastic process. We decompose the response into a posterior mean and a random fluctuation term, a formula

D. F. Munoz-Arboleda, M. Stålhammar, C. Morais Smith

We present a non-Hermitian model with gain/loss and non-reciprocal next-nearest-neighbor hopping that emulates black-hole physics. The model describes a one-dimensional lattice with a smooth connection between regions with distinct hopping parameters. By mapping the system to an effective Schwarzschild metric in the Painlevé-Gullstrand coordinates, we find t

Till Bargheer, Carlos Bercini, Andrea Cavaglià, Davide Lai

We construct orthogonality relations in the Separation of Variables framework for the sl(2) sector of planar N=4 supersymmetric Yang-Mills theory. Specifically, we find simple universal measures that make Q-functions of operators with different spins vanish at all orders in perturbation theory, prior to wrapping corrections. To analyze this rank-one sector,

Marcel K. Goh

Recent advances have linked various statements involving sumsets and cardinalities with corresponding statements involving sums of random variables and entropies. In this vein, this paper shows that the quantity $2{\bf H}\{X, Y\} - {\bf H}\{X+Y\}$ is a natural entropic analogue of the additive energy $E(A,B)$ between two sets. We develop some basic theory su

Lucas Riera Abbade, Anna Helena Reali Costa

Reinforcement learning (RL) has shown promise for trading, yet most open-source backtesting environments assume negligible or fixed transaction costs, causing agents to learn trading behaviors that fail under realistic execution. We introduce three Gymnasium-compatible trading environments -- MACE (Market-Adjusted Cost Execution) stock trading, margin tradin

Xingyu Li, Rongguang Wang, Yuying Wang, Mengqing Guo

Large language models (LLMs) remain brittle on multi-hop question answering (MHQA), where answering requires combining evidence across documents through retrieval and reasoning. Iterative retrieval systems can fail by locking onto an early low-recall trajectory and amplifying downstream errors, while planning-only approaches may produce static query sets tha

Yubo Cai, Yujun Huang, Meshal Alharbi, Gioele Zardini

Designing complex engineered systems requires managing tightly coupled trade-offs between subsystem capabilities and resource requirements. Monotone co-design provides a compositional language for such problems, but its generality does not by itself reveal which problem classes admit exact and scalable computation. This paper isolates such a class by introdu

Felipe Albino dos Santos, Mikhail Neklyudov, Vyacheslav Futorny

Let $\mathfrak{g}$ be a finite-dimensional complex simple Lie algebra and $r,m\ge 2$. The universal central extension of the superelliptic current algebra $\mathfrak{g}\otimes A$ is $\widehat{\mathfrak{g}\otimes A}\cong\mathfrak{g}\otimes A \oplus(\Omega^1_A/dA)$, where $A=\mathbb{C}[t,t^{-1},u]/\langle u^m-(1-2ct^r+t^{2r})\rangle$. We compute the recursion

S. Hales Swift, Ihab F. El-Kady

We introduce a method for measuring the velocity of turbulent fluid flow passing through a pipe using piezoelectric tiles without penetrating the pipe, and without having previously designed the pipe to easily allow monitoring. To measure the flow, the vibrations induced on the pipe by the fluctuating pressure loading induced by the turbulent flow are measur

Rhea Chowers, Oshri Naparstek, Udi Barzelay, Yair Weiss

Many modern multi-modal models (e.g. CLIP) seek an embedding space in which the two modalities are aligned. Somewhat surprisingly, almost all existing models show a strong modality gap: the distribution of images is well-separated from the distribution of texts in the shared embedding space. Despite a series of recent papers on this topic, it is still not cl

Caio Vicentino

We present PolarQuant, a post-training weight quantization method for large language models (LLMs) that exploits the distributional structure of neural network weights to achieve near-lossless compression. PolarQuant operates in three stages: (1) block-wise normalization to the unit hypersphere, (2) Walsh-Hadamard rotation to transform coordinates into appro

Aizirek Turdubaeva, Uichin Lee

Large language models (LLMs) are increasingly used in cross-cultural systems to understand and adapt to human emotions, which are shaped by cultural norms of expression and interpretation. However, prior work on emotion attribution has focused mainly on interpretation, overlooking the cultural background of emotion generators. This assumption of universality

Rachid Benbrik, Mbark Berrouj, Mohammed Boukidi, Mohamed Ech-chaouy

We investigate the discovery prospects of a singly produced vector-like bottom quark in the Type-II Two-Higgs-Doublet Model extended by an $SU(2)_L$ vector-like $(T,B)$ doublet. We focus on the non-standard decay chain $B \to \phi b$, followed by $\phi \to t\bar{t}$, where $\phi = H$ or $A$, leading to a final state with one charged lepton, missing transvers

Daniel J. Singer, Luca Garzino Demo

The way we're thinking about generative AI right now is fundamentally individual. We see this not just in how users interact with models but also in how models are built, how they're benchmarked, and how commercial and research strategies using AI are defined. We argue that we should abandon this approach if we're hoping for AI to support groundbreaking inno

Loris Di Cairano

We introduce a universal criterion for criticality in mean-field rotor Hamiltonians based on the geometric structure of the constant-energy shell. Rather than characterizing the onset of a phase transition through the conventional thermodynamic singularities alone, we show that the relevant information is already encoded in the way the geometry of the shell

Andrei Zabolotskii

In a cluster algebra, a subset of initial cluster variables can be specialised in such a way that all elements of the resulting algebra become polynomial in the remaining variables.

Matthew Loftus

Point-cloud persistent homology (PH) -- computing alpha or Rips complexes on spin-position point clouds -- has been widely applied to detect phase transitions in classical spin models since Donato et al. (2016), with subsequent studies attributing the detection to the topological content of the persistence diagram. We ask a simple question that has not been

Luyang Zhang, Cathy Jiao, Beibei Li, Chenyan Xiong

Generative Engines (GEs) such as ChatGPT and Google's AI Overviews are rapidly reshaping search economics by delivering synthesized responses that allow users to bypass third-party websites, cutting those sites' advertising revenue. Yet this shift also leaves GEs facing their own monetization problem: whether to insert ads into synthesized responses or keep

Sandro Ambuehl, Rahul Bhui, Heidi C. Thysen

A burgeoning literature in economics studies how people form beliefs about the causal structures linking economic variables, and what happens when those beliefs are mistaken. We survey this research and connect it to a rich literature in cognitive science. After providing an accessible introduction to causal Directed Acyclic Graphs, the dominant modeling app

Zichao Wei

Integer multiplication has long been considered a hard problem for neural networks, with the difficulty widely attributed to the O(n) long-range dependency induced by carry chains. We argue that this diagnosis is wrong: long-range dependency is not an intrinsic property of multiplication, but a mirage produced by the choice of computational spacetime. We for

Tushar Dhananjay Pathak

This paper presents ARCS (Autoregressive Circuit Synthesis), a system for amortized analog circuit generation. ARCS produces complete, SPICE-simulatable designs (topology and component values) in milliseconds rather than the minutes required by search-based methods. A hybrid pipeline combines two learned generators, a graph VAE and a flow-matching model, wit

Pengtao Zhao, Boyang Yang, Bach Le, Feng Liu

Repository-level automated program repair (APR) increasingly treats stronger localization as the main path to better repair. We ask a more targeted question: once localization is strengthened, which post-localization levers still provide recoverable gains, which are bounded within our protocol, and what residual frontier remains? We study this question on SW

Ibrahim Buba Garba, George Volonakis

Vacancy-ordered double perovskites have emerged as lead-free alternatives, offering remarkable stability and compositional tunability for optoelectronic applications. In this study, we provide first-principles insights into their electronic properties, surface stability, and energy level alignment using a non-empirical, dielectric-dependent hybrid functional

David A. Garcia-Wetten, Mitchell J. Walker, Peter G. Lim, André Vallières

Superconducting qubits show great promise for the realization of fault-tolerant quantum computing, but lossy, amorphous dielectrics limit current technology. Identifying highly crystalline and stoichiometric dielectrics with intrinsically low microwave loss is therefore a central materials challenge, yet experimentally validated platforms remain scarce. In t

Andrew Laeuger, Taylor Knapp

The imitative verb "chirp" is thought to originate from 16th-century Middle English. Meanwhile, this same word has been used to describe the gravitational waves (GWs) emitted from the merger of compact objects, such as black holes and neutron stars, since at least the 1990s. Motivated purely by this linguistic overlap, we study whether the chirps of birds ca

Alex Berke, Güliz Seray Tuncay, Michael Specter, Mihai Christodorescu

The major mobile platforms, Android and iOS, have introduced changes that restrict user tracking to improve user privacy, yet apps continue to covertly track users via device fingerprinting. We study the opportunity to improve this dynamic with a case study on mobile fingerprinting that evaluates developers' perceptions of how well platforms protect user pri

KrishnaSaiReddy Patil

LLM-based chatbots in government services face critical security gaps. Multi-turn adversarial attacks achieve over 90% success against current defenses, and single-layer guardrails are bypassed with similar rates. We present CivicShield, a cross-domain defense-in-depth framework for government-facing AI chatbots. Drawing on network security, formal verificat

Bruce G. Elmegreen, Deidre A. Hunter

Unsharp-mask images of HI emission from 36 dwarf irregular (dIrr) galaxies illustrate star formation in dispersed clouds and on the rims of large cavities. The cavities can extend for a radial scalelength and typically have circular or slightly sheared forms. The average surface density of cloud peaks is ~20 Msun/pc2, and, combined with their average FUV sta

Tanzir Pial, Flavio Hafner, Dakota Handzlik, Enamul Hassan

Full nation-scale social networks are now emerging from countries such as the Netherlands and Denmark, but these networks present challenging technical issues in working with large, multiplex, time-dependent networks. We report on our experiences in producing dynamic node embeddings of the population network of the Netherlands. We present (a) a layer-sensiti

Wenxi Tan, Bing Li, Lingzhou Xue

Mediation analysis for complex, non-Euclidean data, such as probability distributions, compositions, images, and networks, presents significant methodological challenges due to the inherent nonlinearity and geometric constraints of such spaces. Existing approaches are often restricted to Euclidean settings or specific data types. We propose Random Object Med

Muxin Pu, Mei Kuan Lim, Chun Yong Chong, Chen Change Loy

Skeleton-based isolated sign language recognition (ISLR) demands fine-grained understanding of articulated motion across multiple spatial scales, from subtle finger movements to global body dynamics. Existing approaches typically rely on deep feed-forward architectures, which increase model capacity but lack mechanisms for recurrent refinement and structured

Satwik Ghanta, Peggy Gregory, Gul Calikli

While Sustainable Software Engineering (SSE) tools are widely studied in academia, their practical feasibility in industrial workflows, particularly in regulated environments, remains poorly understood. This study investigates how software practitioners perceive the feasibility of existing SSE tools and techniques, and examines the technical, organizational,

Annie Lu, Hong Kiat Tan, Alexander Xue, Alice Koniges

The 2023 Lahaina wildfire killed 102 people on a peninsula served by a single two-lane highway, making exit lane capacity the binding constraint on evacuation time. We model the evacuation as a system of hyperbolic scalar conservation laws on a directed graph with game-theoretic junction conditions that maximize total network flux, an evacuation-calibrated p

Haoyu Jiang, Yuexi Wang, Yun Yang

In many statistical problems, the data distribution is specified through a generative process for which the likelihood function is analytically intractable, yet inference on the associated model parameters remains of primary interest. We develop a likelihood-free inference framework that combines score matching with gradient-based optimization and bootstrap

Ian Selvaggi

We give a structure result on the set of locally constant stability conditions, $\operatorname{Stab}(\mathcal{D}/R)$, defined by Halpern-Leistner-Robotis showing that it has the structure of a complex manifold, in total analogy with Bridgeland's work. As a consequence, we show that the property of having relative mass-hom bounds and the existence of good mod

Mitsumasa Kondo

Modern cloud OLTP databases have sought performance primarily through user-space optimization - separating storage and compute layers, or distributing transactions across multiple nodes using consensus algorithms. This paper turns attention to a previously unexplored layer: kernel-space I/O behavior. From an on-premises perspective, where a single server wit

Matt Booth

These are notes for a graduate-level introductory course on singularity categories.

Arjun Sharma, Jonas A. Actor, Peter A. Bosler

We present a data-driven framework that extends the predictive capability of classical lifting-line theory (LLT) to a wider aerodynamic regime by incorporating higher-fidelity aerodynamic data from panel method simulations. A neural network architecture with a convolutional layer followed by fully connected layers is developed, comprising two parallel subnet

Leonardo Colombo, Álvaro Rodríguez Abella, Alexandre Anahory Simoes, Anthony Bloch

Foot slip is a major source of instability in bipedal locomotion on low-friction or uncertain terrain. Standard control approaches typically assume no-slip contact and therefore degrade when slip occurs. We propose a control framework that explicitly incorporates slip into the locomotion model through virtual nonholonomic constraints, which regulate the tang

Harrison E. Cook, Wladimir Lyra, Mordecai-Mark Mac Low, K. E. Saavik Ford

Supernova (SN) shocks that originate from stars on orbits embedded in dense active galactic nuclei (AGN) accretion disks evolve differently from those that occur in the interstellar medium. We aim to assess how shocks evolve in this dense stratified medium and understand where SNe are muffled and have their kinetic energy absorbed by an AGN disk versus escap

Maurizio Grasselli, Andrea Poiatti

A common paradigm in phase-field models with singular potentials is that global-in-time weak solutions converge to a single equilibrium only after undergoing asymptotic regularization. However, in arXiv:2510.17296 we introduced a novel method to establish the convergence to a single equilibrium for solutions to Cahn--Hilliard equations, and some related coup

Douglas F. Watson, Tiziano Valentinuzzi

We show that single-valuation exponential kernels, under mild regularity assumptions, converge in the continuum limit to a fourth-order operator with heat asymptotics $\Theta(t)\sim t^{-1/4}$ and hence spectral dimension $d_s=\tfrac12$. Independently, a Tauberian analysis implies that any self-adjoint operator with superlinear eigenvalue counting $N(\lambda)

Hirofumi Wakimoto

Standard macroeconomic frameworks have correctly identified Japan's government debt - now exceeding 240% of GDP - as carrying substantial fiscal risk. Yet FRED data from 2013 to 2026 present an empirical record inviting a complementary perspective: debt ratios have stabilized, nominal GDP has exceeded 670 trillion yen (SAAR), and unemployment has remained ne

Peiran Li, Fangzhou Lin, Shuo Xing, Jiashuo Sun

Autonomous scientific discovery is entering a more dangerous regime: once the evaluator is frozen, a sufficiently strong search process can learn to win the exam without learning the mechanism the task was meant to reveal. This is the idea behind our title. To let the abyss stare back is to make evaluation actively push against the candidate through adaptive

Fangting Zhou, Jiaming Wu, Balazs Kulcsar

Public electric vehicle (EV) charging infrastructure has expanded rapidly, yet utilization across charging stations remains uneven and often inefficient. Existing operator-determined pricing schemes offer limited flexibility to coordinate heterogeneous user demand within constrained capacity. This study proposes a user-driven pricing and scheduling framework

C. Fremling, S. Covarrubias, J. Sollerman, K. De

We present HST WFC3/UVIS F275W near-UV imaging of 91stripped-envelope supernovae (SE SNe; Types IIb, Ib, Ic) from Snapshot program SNAP-16657, observed at phases of 270-1845 days (median 952 days) after first optical detection. We detect UV counterparts in 13 SE~SNe, of which 6 are classified as secure and 7 as ambiguous after comparison to nearby H\textsc{i

Iness Halimi, Emmanuel Piffo, Oumnia Boudersa, Yvan Marcel Carre Vilmorin

Clinical trials are characterized by high costs, extended timelines, and substantial operational risk, yet reliable prospective methods for predicting trial success before initiation remain limited. Existing artificial intelligence approaches often focus on isolated metrics or specific development stages and frequently rely on variables unavailable at the tr

Erika Gallo, John Zweck, Yuri Latushkin

We introduce a numerical method to determine the stability of stationary pulse solutions of the complex Ginzburg-Landau equation. The method involves the computation of the point spectrum of the first-order linear differential operator with matrix-valued coefficients on the real line obtained by linearizing the Ginzburg-Landau equation about a stationary pul

Chun Huang

Large Language Models (LLMs) are now widely used in astrophysics, but do they actually make our lives easier, or do they merely invent new physics with enough confidence to hide a minus sign? In a specialized field where checking fluent hallucinations is itself labor-intensive, AI assistance can demand as much work as the task it claims to simplify. To evalu

Bilgehan Sel, Xuanli He, Alwin Peng, Ming Jin

Fine-tuning APIs offered by major AI providers create new attack surfaces where adversaries can bypass safety measures through targeted fine-tuning. We introduce Trojan-Speak, an adversarial fine-tuning method that bypasses Anthropic's Constitutional Classifiers. Our approach uses curriculum learning combined with GRPO-based hybrid reinforcement learning to

Yujin Ham, Junho Kim, Vivek Boominathan, Guha Balakrishnan

Egocentric "walking tour" videos provide a rich source of image data to develop rich and diverse visual models of environments around the world. However, the significant presence of humans in frames of these videos due to crowds and eye-level camera perspectives mitigates their usefulness in environment modeling applications. We focus on addressing this chal

Mario J. Pinheiro

We identify a new class of non-Hermitian causal processes that produce statistically significant temporal correlations invisible to conventional spectral methods. Using a generative model with a strictly causal memory kernel, we demonstrate that time-asymmetric stochastic processes naturally yield sharp transitions at characteristic scales that appear as loc

Kushal Vyas, Alper Kayabasi, Daniel Kim, Vishwanath Saragadam

The approximation and convergence properties of implicit neural representations (INRs) are known to be highly sensitive to parameter initialization strategies. While several data-driven initialization methods demonstrate significant improvements over standard random sampling, the reasons for their success -- specifically, whether they encode classical statis

Abhinna Sundar Samantaray, Finnja Annika Fluhrer, Dhruv Saini, Omkar Charaple

Astrology has long been used to interpret human personality, estimate compatibility, and guide social decision-making. Zodiac-based systems in particular remain culturally influential across much of the world, including in South Asian societies where astrological reasoning can shape marriage matching, naming conventions, ritual timing, and broader life plann

Fehmi Ekin Giritlioglu

We investigate translation actions of countable dense subgroups of non-unimodular locally compact second countable (lcsc) groups. Using left-right actions, we show that the left translation action $\Gamma \curvearrowright G$ given by a countable dense subgroup $\Gamma$ of a locally compact second countable group $G$ can only be strongly ergodic if $G$ is alm

Florian Lehner, Christian Lindorfer, Rögnvaldur G. Möller, Wolfgang Woess

Bass-Serre theory provides a powerful framework for studying group actions on trees. While extremely effective for structural questions in group theory, it is less suited to the systematic construction of group actions with prescribed local behaviour. Motivated by local-to-global constructions such as the Burger-Mozes universal groups and local action diagra

Felipe Dilho Alves, João Carlos Alves Barata

Quantum mechanics has been subject to logical scrutiny since its inception. The behavior of quantum systems, which are fundamentally dissimilar from classical systems, often appears to point to a logical inconsistency in quantum mechanics, allegedly leading to contradictions in the prediction of experimental measurements--though such contradictions have neve

Julius Leino, Jörg Tiedemann

Shared multilingual representations are essential for cross-lingual tasks and knowledge transfer across languages. This study looks at the impact of parallel data, i.e. translated sentences, in pretraining as a signal to trigger representations that are aligned across languages. We train reference models with different proportions of parallel data and show t

Alexandra P. Klipfel, Miguel Vanvlasselaer, Sokratis Trifinopoulos, David I. Kaiser

Exploding primordial black holes can source baryon asymmetry soon after the electroweak phase transition, as high-energy Hawking radiation drives ultrarelativistic shocks in the surrounding plasma. The shocks and their trailing rarefaction waves delineate two bubble-like walls around a shell of superheated fluid, in which electroweak symmetry is restored. Th

Diego C. Lerma-Torres

Large language models lack persistent, structured memory for long-term interaction and context-sensitive retrieval. Expanding context windows does not solve this: recent evidence shows that context length alone degrades reasoning by up to 85% - even with perfect retrieval. We propose a bio-inspired memory framework grounded in complementary learning systems

Felix Duelmer, Jakob Klaushofer, Magdalena Wysocki, Nassir Navab

Novel view synthesis (NVS) in ultrasound has gained attention as a technique for generating anatomically plausible views beyond the acquired frames, offering new capabilities for training clinicians or data augmentation. However, current methods struggle with complex tissue and view-dependent acoustic effects. Physics-based NVS aims to address these limitati

David Tobar, Rory Smith, Antonio Montero-Dorta, Katarina Kraljic

We investigate the evolution of dark matter halo spin alignments with respect to cosmic filaments, exploring how halo mass, proximity to filaments, and major mergers influence their orientation over time. We perform a suite of dark matter-only zoom-in N-body simulations centered on ten filaments extracted from a cosmological box using the 1DREAM structure fi

Deepak Akkil, Mowafak Allaham, Amal Raj, Tamer Abuelsaad

Reliable evaluation of AI agents operating in complex, real-world environments requires methodologies that are robust, transparent, and contextually aligned with the tasks agents are intended to perform. This study identifies persistent shortcomings in existing AI agent evaluation practices that are particularly acute in web agent evaluation, as exemplified

Johnny Guzmán, Anil N. Hirani, Bingyan Liu, Pratyush Potu

Smooth Poincare operators are a tool used to show the vanishing of smooth de Rham cohomology on contractible manifolds and have found use in the analysis of finite element methods based on the Finite Element Exterior Calculus (FEEC). We construct analogous discrete Poincare operators acting on cochains and Whitney forms. We provide explicit, constructive rea

Newton Solórzano, Dik D. Lujerio Garcia, Víctor León, Alexis Rodríguez Carranza

In this paper, we study weakly orthogonally invariant Finsler metrics and derive explicit expressions for their Berwald and Landsberg curvatures. We then obtain the system of partial differential equations characterizing generalized Finsler warped product metrics on $I \times \mathbb{R}^n$, which form a subclass of weakly orthogonally invariant Finsler metri

Matteo Di Carlo, Simone Bacchio, Erik Bäske, Alessandro De Santis

We present a strategy to extend lattice calculations of inclusive hadronic $\tau$ decays from isosymmetric QCD to QCD+QED. The inclusive decay rate can be related to suitable Euclidean correlation functions, allowing for a first-principles evaluation of electromagnetic and isospin-breaking effects. Within the RM123 framework, radiative corrections are decomp

Sergio Urrea, Adrian Basarab, Hervé Liebgott, Henry Arguello

Ultrasound imaging faces a trade-off between image quality and hardware complexity caused by dense transducers. Sparse arrays are one popular solution to mitigate this challenge. This work proposes an end-to-end optimization framework that jointly learns sparse array configuration and image reconstruction. The framework integrates a differentiable Image Form

J. Peltonen, E. Rosolowsky, A Ginsburg, R. Indebetouw

The most direct method of measuring the star formation rate is with young stellar objects (YSOs), but this requires high-resolution observations and high-quality models. Using the latest YSO radiation transfer and stellar evolution models, we have developed a population synthesis code that generates model YSO populations that can be observed by JWST. We comb

C. Gartland, B. Randrianantoanina, N. L. Randrianarivony

We prove that for all metric spaces $X$ the following properties of the lamplighter space $\mathsf{La}(X)$ are equivalent: (1) every snowflake of $\mathsf{La}(X)$ admits a biLipschitz embedding into a finite product of $\mathbb{R}$-trees, (2) every snowflake of $\mathsf{La}(X)$ admits a biLipschitz embedding into a Hilbert space, (3) $\mathsf{La}(X)$ has fin

Siva Kumar Sastry Hari, Vignesh Balaji, Sana Damani, Qijing Huang

Optimizing GPU kernels with LLM agents is an iterative process over a large design space. Every candidate must be generated, compiled, validated, and profiled, so fewer trials will save both runtime and cost. We make two key observations. First, the abstraction level that agents operate at is important. If it is too low, the LLM wastes reasoning on low-impac

Konstantinos Georgiou, Maofeng Tang, Hairong Qi

Masked image modeling (MIM) methods typically operate in either raw pixel space (reconstructing masked patches) or latent feature space (aligning with a pre-trained teacher). We present MEDiC (Multi-objective Exploration of Distillation from CLIP), a framework that combines both spaces in a single pipeline through three complementary objectives: patch-level

Oussama Bensaid, Anthony Genevois, Romain Tessera

In this article, we determine, given a finite graph $\Gamma$ and an integer $n \geq 1$, when a right-angled Artin group $A(\Gamma)$ virtually splits over an abelian subgroup of rank $n$. More precisely, we show that the following assertions are equivalent: (1) $A(\Gamma)$ admits $\mathbb{Z}^n$ as a codimension-one subgroup, (2) $A(\Gamma)$ virtually splits o

Chikako Yasui, Natsuko Izumi, Masao Saito, Ryan M. Lau

This study presents the first high-resolution, high-sensitivity mid-infrared (MIR) investigation of protoplanetary disks in a low-metallicity environment, using JWST/NIRCam and MIRI observations of Digel Cloud 2, a star-forming region in the outer Galaxy ($D \simeq 8$ kpc, ${\rm [M/H]} \simeq -0.7$ dex). It hosts two very young ($\sim$0.1 Myr) embedded clust

P. S. Avdienko, L. Hanschke, Q. Buchinger, N. Akhlaq

We present the observation of electrically tunable quantum coupling of orbital states in individual InAs/InGaAs quantum dot molecules emitting in the telecom O-band (~1300 nm). By tuning the static electric field along the growth axis of the QD-molecule, we observe pronounced anticrossings between excitonic transitions and determine the dependence of the int

Zih-Sing Fu, Peter Zhi Xuan Li, Sertac Karaman, Vivienne Sze

High-fidelity 3D occupancy mapping is essential for many edge-based applications (such as AR/VR and autonomous navigation) but is limited by power constraints. We present Gleanmer, a system on chip (SoC) with an accelerator for GMMap, a 3D occupancy map using Gaussians. Through algorithm-hardware co-optimizations for direct computation and efficient reuse of

L. F. Almeida, A. N. Cesário, P. A. Sousa, P. Rocha-Rodrigues

The $n = 1$ Ruddlesden-Popper titanates, NaRTiO$_{4}$ (R = rare-earth), exhibit a structural behaviour where non-centrosymmetry is driven by cooperative oxygen octahedral rotations (OORs) rather than conventional second-order Jahn-Teller distortions. In this work, we present an \textit{ab-initio} investigation of the structural, electronic and hyperfine prop

Lucas Gautheron, Nori Jacoby, Peter Harrison

Adaptive experiments automatically optimize their design throughout the data collection process, which can bring substantial benefits compared to conventional experimental settings. Potential applications include, among others: computerized adaptive testing (for selecting informative tasks in ability measurements), adaptive treatment assignment (when searchi

César E. Echevarría, José F. Nieves, Francisco Orbe, Sarira Sahu

In a recent work, we considered the propagation of fermions in the background of a scalar Bose-Einstein (BE) condensate. Using some illustrative Yukawa-type coupling models between the fermions and the scalar fields, we determined the dispersion relations of the fermions and the scalar modes in various models. To complement that work, here we consider the co

Ali Tfaily, Youssef Diouane, Nathalie Bartoli, Michael Kokkolaras

The use of transfer learning within Bayesian optimization addresses the disadvantages of the so-called \textit{cold start} problem by using source data to aid in the optimization of a target problem. We present a method that leverages an ensemble of surrogate models using transfer learning and integrates it in a constrained Bayesian optimization framework. W

Yicheng Cai, Mitchell John DeStefano, Guodong Dong, Pulkit Handa

As Large Language Models (LLMs) and multi-agent AI systems are demonstrating increasing potential in cybersecurity operations, organizations, policymakers, model providers, and researchers in the AI and cybersecurity communities are interested in quantifying the capabilities of such AI systems to achieve more autonomous SOCs (security operation centers) and

Youngjoong Kwon, Yao He, Heejung Choi, Chen Geng

We present a feed-forward human performance capture method that renders novel views of a performer from a monocular RGB stream. A key challenge in this setting is the lack of sufficient observations, especially for unseen regions. Assuming the subject moves continuously over time, we take advantage of the fact that more body parts become observable by mainta

Francesco Serra Cassano, Kilian Zambanini

We establish two characterizations of real-valued Sobolev and BV functions on Carnot groups. The first is obtained via a nonlocal approximation of the distributional horizontal gradient, while the second is based on an $L^p$ Taylor approximation, in the spirit of the results by Bourgain, Brezis and Mironescu.

Akshaya Srinivasan, Xiaoyin Cheng, Jianming Yi, Alexander Geng

Hybrid quantum-classical machine learning offers a promising direction for advancing automated quality control in industrial settings. In this study, we investigate two hybrid quantum-classical approaches for classifying defects in aluminium TIG welding images and benchmarking their performance against a conventional deep learning model. A convolutional neur

Srivaths Ranganathan, Nikhil Khani, Shawn Andrews, Chieh Lo

Knowledge Distillation (KD) has been widely used to improve the quality of latency sensitive models serving live traffic. However, applying KD in production recommender systems with low traffic is challenging: the limited amount of data restricts the teacher model size, and the cost of training a large dedicated teacher may not be justified. Cross-domain KD

Christian Parkinson, Adan Baca, Huy Nguyen

We present a method for collisionless multi-agent path planning using the Hamilton-Jacobi-Bellman equation. Because the method is rooted in optimal control theory and partial differential equations, it avoids the need for hierarchical planners and is black-box free. Our model can account for heterogeneous agents and realistic, high-dimensional dynamics. We d

Elham Rostami, Taous-Meriem Laleg-Kirati, Hamidou Tembine

We study training-free flow matching between two Gaussian mixture models (GMMs) using explicit velocity fields that transport one mixture into the other over time. Our baseline approach constructs component-wise Gaussian paths with affine velocity fields satisfying the continuity equation, which yields to a closed-form surrogate for the pairwise kinetic tran

Victoria Dochkina

How much autonomy can multi-agent LLM systems sustain -- and what enables it? We present a 25,000-task computational experiment spanning 8 models, 4--256 agents, and 8 coordination protocols ranging from externally imposed hierarchy to emergent self-organization. We observe that autonomous behavior already emerges in current LLM agents: given minimal structu

Daniel Hill, Martin Slawski

Motivated by the prevalence of environments in which data is abundant while resources for storage and/or transmission might be scarce, we study linear regression when predictors, their squares, and responses are subject to single-bit dithered quantization. An estimator relying on plug-in estimation of the quadratic and linear terms in the quadratic program f