April 2026 arXiv papers — page 30

Samuel Pérez-Ayala

We derive sharp bounds for three types of eigenvalue problems. First, we derive an upper bound for the first $p$-Dirichlet eigenvalue on conformally compact (CC) spaces. As a consequence, we show that for a class of CC submanifolds of asymptotically hyperbolic spaces, the asymptotic sectional curvatures, the meeting angle at infinity, and the vanishing of th

Yongzhong Xu

We show that replacing the rolling SVD of AdamW updates with a rolling SVD of loss gradients changes the diagnostic by 1-2 orders of magnitude. Performing SVD on the loss gradient instead of the AdamW update increases the measured perturbative coupling between SED directions and Linear Centroid Hypothesis (LCH) features from $ \bar{R}_k \approx 3 $--$9\times

Han-Qiu Zhang, Shuang-Shuo Chu, Jian-Song Zhang, Wen-Xue Zhong

We present an alternative scheme to achieve nonreciprocal unconventional magnon blockade (NUMB) in a hybrid system formed by two microwave cavities and one yttrium iron garnet (YIG) sphere, where the pump and signal cavities interact nonlinearly with each other and the signal cavity is coupled to the YIG sphere. It is found that the nonlinear coupling occurs

Zhihao Xie, Adam Miranowicz, Zhenhua Li, Tao Li

We propose a parallel protocol for implementing distributed nonlocal quantum gates between spatially separated stationary qubits encoded in dual-species quantum emitters (i.e., color-center and superconducting qubits). By utilizing entangled photon pairs with distinct frequencies as a quantum data bus, our approach connects spatially separated devices withou

Prabhjot Singh, Manmeet Singh

Operational phase unwrapping is the primary computational bottleneck in InSAR-based volcanic and seismic monitoring. We challenge the industry trend of adopting high-complexity computer vision architectures, such as attention mechanisms, without validating their suitability for physics-constrained geophysical regression. We present the first large-scale arch

Aditya Basarkar, Emmett B. Kendall, David Randahl, Jonathan P. Williams

Whether or not a country is at war, or experiencing escalating or deescalating levels of conflict, has massive ramifications on a country's national and foreign policy. Given a country's history of conflict, or lack thereof, future predictions about the war-status of a country are valuable information. In this paper, we present the use of conformal predictio

Liping Tao, Chee Wei Tan

Spectrum cartography reconstructs spatial radio fields from sparse and heterogeneous wireless measurements, underpinning many sensing and optimization tasks in wireless networks. Attention mechanisms have recently enabled adaptive measurement aggregation via attention kernel-based formulations. However, the resulting exponential kernels exhibit severe spectr

Lei Wang

We solve the time-dependent Schr\"odinger equation by learning the score function, the gradient of the log-probability density, on Bohmian trajectories. In Bohm's formulation of quantum mechanics, particles follow deterministic paths under the classical potential supplemented by a quantum potential depending on the score function of the evolving density. The

James Pustejovsky, Nikhil Krishnaswamy

We propose Frictive Policy Optimization (FPO), a framework for learning language model policies that regulate not only what to say, but when and how to intervene in order to manage epistemic and normative risk. Unlike standard alignment methods that optimize surface-level preference or task utility, FPO treats clarification, verification, challenge, redirect

Amir Saeidi, Venkatesh Mishra, Souradeep Mukhopadhyay, Gaowen Liu

Large Language Models are being increasingly deployed as the decision-making core of autonomous agents capable of effecting change in external environments. Yet, in conversational benchmarks, which simulate real-world customer-centric issue resolution scenarios, these agents frequently fail due to the cascading effects of incorrect decision-making. These cha

Julian Heeck, Brandon B. Le

We present a minimal basis for non-derivative baryon-number-violating operators in the Standard Model Effective Field Theory up to mass dimension 11, as well as for the $(\Delta B,\Delta L) = (2,2)$ and $(2,-2)$ operators at dimension 12. Compared to existing results, our bases generally contain fewer terms and simpler contractions, although we also highligh

Ji-eun Kim, Volker Dellwo

Korean aegyo is a socially recognized childlike speaking style used predominantly in romantic interactions among adults. This study examined vowel space modification in aegyo by analyzing formant frequencies from twelve Seoul Korean speakers who produced identical scripts in aegyo and non-aegyo styles. Results show that aegyo speech features a significant in

Guangzeng Han, Xiaolei Huang

Instruction-tuning datasets often contain substantial redundancy and low-quality samples, necessitating effective data selection methods. We propose an instruction data selection framework based on weighted in-context influence (wICI), which measures how effectively each candidate example reduces instruction-following difficulty for semantically related peer

Sicheng Dai, Kai Chen, Hongwang Xiao, Shan Yu

Recent self-supervised pre-training methods for electroencephalogram (EEG) have shown promising results. However, the pre-trained models typically require full fine-tuning on each downstream task individually to achieve good performance. In practical applications involving multiple tasks, utilizing a separate model for each task is not ideal regarding comput

Liwen Wu, Haolin Lu, Bing Xu, Miloš Hašan

High-fidelity 3D assets exhibit intriguing global illumination effects like subsurface scattering, glossy interreflections, and fine-scale fiber scatterings, which often involve long scattering paths that are expensive to simulate. We introduce 8D neural assets (8DNA) to pre-bake these light transport effects into neural representations. Unlike prior methods

Hanyi Wang, Han Fang, Zheng Wang, Shilin Wang

Recent advances in diffusion models have enabled high-quality image generation, leading to increasing demand for post-generation editing that modifies local regions while preserving global structure. Achieving such flexible and precise editing requires a high-quality starting point, a latent representation that provides both the freedom needed for diverse mo

Shiwu Zhang, Zheng Cai, Fabrizio Arrigoni Battaia, Sebastiano Cantalupo

We present a comprehensive analysis of circumgalactic Lya nebulae around 59 unobscured and 26 obscured quasars at z=2-3, observed with the Keck Cosmic Web Imager (KCWI), to constrain the nature of active galactic nuclei (AGN) at cosmic noon. We find that Lya nebulae around unobscured quasars are significantly less symmetric having a symmetry parameter of a_w

Daniel D. Rivera, Gustavo M. Dalpian, John P. Perdew

Strongly correlated systems have long been a central and highly non-trivial topic in condensed matter physics. At the non-interacting level, strong correlation can be associated with powerful (near) degeneracies between occupied and unoccupied states, which leads to a high density of states near the Fermi level in metallic configurations. Such regimes are co

John M Donohue, Michael J Grabowecky, George Nichols, Martin Laforest

To grow the quantum information science and technology workforce, opportunities for students to gain experiential learning and build a sense of belonging in the broader community are essential. The Undergraduate School on Experimental Quantum Information Processing (USEQIP) is a two-week summer school for undergraduate students that has been held since 2009

Jiatong Ma, Longteng Guo, Yuchen Liu, Zijia Zhao

We present M$^3$-VQA, a novel knowledge-based Visual Question Answering (VQA) benchmark, to enhance the evaluation of multimodal large language models (MLLMs) in fine-grained multimodal entity understanding and complex multi-hop reasoning. Unlike existing VQA datasets that focus on coarse-grained categories and simple reasoning over single entities, M$^3$-VQ

Nozomi Nakatsuyama

We investigated the evolute of a space curve with singular points. As smooth curves with singular points, we apply the theory of framed curves. However, the involute corresponding to the evolute in the sense of the locus of the centre of osculating spheres has not been defined as far as we know. In this paper, we directly define the evolutes and involutes of

Suparno Roy Chowdhury, Manan Roy Choudhury, Tejas Anvekar, Muhammad Ali Khan

We study clinical trial table reasoning, where answers are not directly stored in visible cells but must be reasoned from semantic understanding through normalization, classification, extraction, or lightweight domain reasoning. Motivated by the observation that current LLM approaches often suffer from "bad reasoning" under implicit planning assumptions, we

Vinith M. Suriyakumar, Ayush Sekhari, Lena Stempfle, Robertson Wang

Auditing the fine-tunes of open-weight generative models for harmful specialization has become a new governance challenge for model hosting platforms. The standard toolkit, generative evaluation via curated prompts or red-teaming, does not scale to platform-level auditing and breaks down entirely for domains like CSAM where generation is legally constrained.

Wei Huang, Xiaofang Cai, Qiaozhen Guo, Xiaosong Wu

The Management Information Systems (MIS) discipline has long grappled with how to theorize the complex, mutually constitutive relationships among people, information technology, and organizational structures. Decades of research have produced influential but fragmented theoretical streams from socio-technical systems theory to technology acceptance models, f

Kaiyue Lu, Qiaoxuan Zhang, Yukun Lu

Vehicle platooning has attracted increasing attention as a promising approach to improve traffic efficiency, energy consumption, and roadway safety through coordinated multi-vehicle operation. A key challenge in platooning lies in maintaining stable and accurate path tracking during dynamic maneuvers such as lane changes, where lateral deviations and heading

James Kennington, Joshua Black, Zach Yarbrough, Yun-Jing Huang

Low-latency gravitational-wave search pipelines provide early-warning alerts for multimessenger astrophysical transients. Current pipelines whiten the data stream using acausal, linear-phase filters, which require a look-ahead buffer that introduces several seconds of algorithmic latency. Eliminating this latency requires causal, minimum-phase whitening filt

Zachary Montague, Rhea M Grover, Andrew Baumgartner, Assya Trofimov

T-cells are central to SARS-CoV-2 clearance and immunological memory, yet their contribution to the persistence of post-acute sequelae of COVID-19 (PASC) remains poorly understood. The immunological features that distinguish individuals who develop PASC from those who recover fully are unresolved, in part due to the phenotypic heterogeneity of the condition

Zhongbin Wang, Wenlong Tang, Simin Pang, Hongxing Zhu

The active manipulation of phonon transport remains a central challenge in phononics and spin caloritronics due to the charge-neutral nature of heat carriers. Spin-phonon coupling (SPC) offers a promising route for the dynamic control of heat carriers, yet its progress has been limited due to the lack of a unified framework and suitable material platforms. H

Nassar Doudin, Jian Jiang, Chun Tang, Xiao Cheng Zeng

Hydrogen-bond networks govern molecular structure and function across chemistry, biology and materials science, yet their deterministic control at the atomic scale remains a central challenge (1-9).Here, we directly visualize how an external electric field enables reversible control of a hydrogen-bond network in monolayer ice on graphite through interfacial

Wenping Jiang, Michael Doublier, Russell Korsch, Andy Clark

We have used new magnetotelluric data collected in the Curnamona Province and the adjacent part of the Delamerian Orogen margin to image electrical conductivity structures and to inform the understanding of the crustal architecture within the regional geological context. The preferred 3D resistivity model confirms, and resolves in greater detail, crustal-sca

Yong Sun, Shijia Chen, Minqing He, Sizhong Wu

Fluid models offer crucial computational efficiency for plasma simulations, yet accurately capturing kinetic effects like Landau damping remains a fundamental challenge. While conventional closures (e.g., Hammett-Perkins and Hunana) are widely used, their fidelity relative to exact kinetic response degrades significantly depending on the perturbation wave nu

Chenhui Zhu, Fei Wang, Weimin Han

This paper presents a numerical approach to the stochastic obstacle problem using the stochastic Galerkin (SG) method. Due to the low regularity of the solution, linear finite elements are employed in both the physical and random variable spaces. Properties of random fields and variational inequalities of the first kind are employed to establish the well-pos

Wenshuo Wang

This position paper argues that knowledge distillation must account for what it loses: student models should be judged not only by retained task scores, but by whether they preserve the teacher capabilities that make those scores reliable. This matters because distillation is increasingly used to turn large teacher models into deployable students, yet headli

Lijia Lv, Xuehai Tang, Jie Wen, Jizhong Han

Agent Skills package SKILL.md files, scripts, reference documents, and repository context into reusable capability units, turning pre-load auditing from single-prompt filtering into cross-file security review. Existing guardrails often flag risk but recover malicious intent inconsistently under semantics-preserving rewrites. This paper formulates pre-load au

Christopher D. Long

We develop a terminal-defect method for the double Dixie cup problem and use it to prove the finite-variance extremality conjecture of Doumas and Papanicolaou. For every \(m\ge1\) and \(N\ge2\), among all positive coupon probability vectors \(p=(p_1,\ldots,p_N)\), the variance of the time \(T_m(N)\) to collect \(m\) complete sets is uniquely minimized at the

Y. J. Wang, C. K. Tam, Z. G. Xiao, W. G. Lynch

We report proton-proton correlation function measurements in central $^{132}$Sn+$^{124}$Sn and $^{108}$Sn+$^{112}$Sn collisions at 270 MeV/nucleon. The proton emitting source sizes are extracted for the systems by using femtoscopic imaging technique. The fast dynamic core radius for the neutron-rich system is found to be $2.22 \pm 0.13\ \text{(stat.)} \pm 0.

Sayan Ghosh, Amitav Sahu, Stephanie Gonzalez-Migoni, Thomas L. C. Jansen

Action-detected two-dimensional electronic spectroscopy (A-2DES) could potentially be a versatile chemical tool with applicability across a range of photophysical observables such as photocurrent, photoionization, or fluorescence. However, a prominent absence of excited state energy/charge transfer dynamics signals in archetypal photosynthetic proteins has s

Zeyad Abdelrazek, Young Lee

This paper presents a system combining symbolic execution (KLEE) with a 4-agent multi-LLM architecture for detecting memory vulnerabilities in Rust unsafe code. A central challenge we address is the incomplete-code problem: CVE database entries provide only isolated code snippets that lack struct definitions, imports, and Cargo manifests, causing all existin

J. L. Jin, Q. Zhao, P. J. Li, M. Kimura

To investigate the influence of neutron-proton asymmetry on the formation of asymmetric clusters, we perform a systematic comparative study of $^{3}$H and $\alpha$ cluster preformation in the Boron isotopic chain ($^{11-14}$B). Within the framework of Antisymmetrized Molecular Dynamics (AMD), we compute the nuclear wave functions and subsequently extract the

P. Beretta, A. Codello

We show that it is possible to use dimensional regularization (DR) beyond the usual $\varepsilon$-expansion in the context of renormalization group (RG) calculations in Critical Phenomena. Based on this fact, we propose a new functional RG scheme - Functional Dimensional Regularization (FDR) - and apply it to a scalar theory in three dimensions. We compute t

Ravikumar Balakrishnan, Sanket Mendapara

Typographic prompt injection exploits vision language models' (VLMs) ability to read text rendered in images, posing a growing threat as VLMs power autonomous agents. Prior work typically focus on maximizing attack success rate (ASR) but does not explain \emph{why} certain renderings bypass safety alignment. We make two contributions. First, an empirical stu

ZhiWei Su, Ding Wang, Yuan Guo, Yang Qiao

The second-order harmonic (2f) component generated by twin-rotary compressor is a dominant low-frequency noise source of variable refrigerant flow (VRF) outdoor units, yet its amplitude fluctuates strongly with environmental thermal load and valve opening, making it difficult to assess accurately using conventional mechanism-based models. This paper proposes

Peng Liao, Shangsong Liang, Lin Chen, Peijia Zheng

Inertial Measurement Unit (IMU)-based Human Activity Recognition (HAR) aims to interpret and classify user behaviors from temporal motion signals. Recently, deep learning frameworks have advanced this task by learning and extracting discriminative spatiotemporal representations, significantly improving recognition performance. However, IMU-based HAR still fa

Cristián Erices, Stella Kiorpelidi

Although asymptotically flat black holes generically lack thermodynamic phase transitions, we show that curvature-induced scalarization of electrically charged black holes in Einstein-Maxwell- Scalar-Gauss-Bonnet theory provides a natural setting for nontrivial thermodynamic behavior, without invoking external confining mechanisms or an extended thermodynami

Ronnie de Souza Santos

Role models are widely discussed in educational research as influential in students identity development and sense of belonging, yet less attention has been given to how role model visibility can be systematically embedded within everyday engineering instruction. This paper presents an analytic autoethnographic account of integrating historically grounded ro

Yifan Song, Sina Ilkhani, Leah Webb, Derrick Lin

Electronic band structures and the Fermi energy provide essential information for understanding the electronic properties of solids. In semiconductors, the Fermi energy is determined by the donor and acceptor concentrations. For diamond, the relationship between the Fermi energy and the donor-acceptor concentrations is highly nonlinear; therefore, experiment

Ashish Mehta, Jared Moore, Jacy Reese Anthis, William Agnew

There is growing concern that AI chatbots might fuel delusional beliefs in users. Some have suggested that humans and chatbots mutually reinforce false beliefs over time, but quantitative evidence is lacking. Using a unique dataset of chat logs from individuals who exhibited delusional thinking, we developed a latent state model that captures accumulating an

Andreas Dechant, Kiyoshi Kanazawa

We develop the thermodynamics of non-Markovian generalized Langevin equations by embedding them in a high-dimensional Markovian representation involving auxiliary degrees of freedom. If the memory is linear and satisfies detailed balance with the noise, we provide an explicit construction of the embedding for non-Markovian dynamics with many degrees of freed

Sayam Sethi, Sahil Khan, Aditi Awasthi, Abhinav Anand

Near-term FTQC system designs are constrained by limited error budgets and largely sequential execution of non-Clifford gates. As a result, reducing the number of the most-error prone instructions becomes critical for successful program execution. In this work, we study the extractor architecture, a recently proposed FTQC design that enables universal quantu

Eiyu S. Gushiken, Mizuki Tani, Hiroki Katow, Kenichi L. Ishikawa

We theoretically show that energy absorption in crystalline silicon can be controlled by two-color femtosecond double-pulse irradiation, in which two temporally separated pulses with different wavelengths interact sequentially with the system. Using time-dependent density functional theory, we systematically examine the wavelength and intensity dependence of

Ruijie Yao, Chenhang Li, Danyang Zhuo, Tingjun Chen

Wearable Human Activity Recognition (HAR) still lacks a representation that is both explicit and adaptable. Handcrafted time-series features (TSFs) capture meaningful motion statistics and remain competitive on standard benchmarks, but they are usually used as fixed preprocessing outputs. Deep models learn adaptable representations directly from raw signals,

Riju Basak, Surjeet Singh Choudhary, Daniel Spector

In this article, we study the fractional spherical maximal function and its lacunary counterpart. We study the necessary and sufficient conditions for $L^p-L^q$ boundedness of both maximal functions. In particular, we prove the restricted weak type estimate for both full and lacunary fractional spherical maximal functions at the boundary of the maximal $L^p-

Paulo Areyuna C, Giovanna Cottin, Bastián Díaz Sáez, Zeren Simon Wang

We study an extension of the singlet-scalar Higgs portal featuring a dark vector $V_\mu$ and a real scalar $\phi$. The vector is a dark matter (DM) candidate, while $\phi$ is long-lived and decays via higher-dimensional operators. We explore the DM production via freeze-in at low and high reheating temperatures. At colliders, the decay $\phi\to Z+V$ yields d

Yisu Yan, Jifeng Guo

Weak constitutive fluctuations in dispersive subsurface media can induce distributed clutter that reshapes the observation structure of ground-penetrating radar (GPR). This paper analyzes this effect for single-snapshot frequency-diverse array multiple-input multiple-output GPR. Focusing on medium-induced clutter, rather than on general target--clutter joint

Abigail O'Neill, Alan Zhu, Mihran Miroyan, Narges Norouzi

Language Model (LM)-based agents remain largely untested in mixed-motive settings where agents must leverage short-term cooperation for long-term competitive goals (e.g., multi-party politics). We introduce Cooperate to Compete (C2C), a multi-agent environment where players can engage in private negotiations while competing to be the first to achieve their s

Yasumasa Matsuda, Michel F. C. Haddad

We propose a density-valued vector autoregressive model with latent factors for multivariate time series of density functions. Motivated by weekly regional distributions of SARS-CoV-2 cycle threshold (Ct) values in Brazil, we study their distributional dynamics across regions. The Ct value is the number of amplification cycles required for the viral signal t

Nishant Shukla

Coherent quantum rollout for sequential decision problems requires a unitary simulator: randomness must live in explicit quantum registers, and basis-state selectors must be mapped to actions reversibly. With branch-dependent valid actions, this mapping is totalized coherent rank-select over an entangled $N$-bit validity mask: return the position of the $r$-

Ryan Dickmann, Roberta Shapiro

The fine curve graph of a surface is a graph whose vertices are essential simple closed curves and whose edges connect disjoint curves. Following a rich history of hyperbolicity of various graphs associated to surfaces, the fine curve graph was shown to be hyperbolic by Bowden-Hensel-Webb, while the curve graph, obtained from the fine curve graph by collapsi

Sanmay Das, Fang-Yi Yu, Yuang Zhang

Fraud can pose a challenge in many resource allocation domains, including social service delivery and credit provision. For example, agents may misreport private information in order to gain benefits or access to credit. To mitigate this, a principal can design strategic audits to verify claims and penalize misreporting. In this paper, we introduce a general

Alexander Blasberg, Vasilis Kypriotis, Dimitrios Skarlatos

Rapid advances in Large Language Models (LLMs) create new opportunities by enabling efficient exploration of broad, complex design spaces. This is particularly valuable in computer architecture, where performance depends on microarchitectural designs and policies drawn from vast combinatorial spaces. We introduce Agentic Architect, an agentic AI framework fo

Anna Kuznetsova

Discrete mathematics and probability theory contain foundational material for computer scientists. Despite their importance, instructors often worry that students will find these courses to be too abstract and seemingly disconnected from their future careers. For this research project, we introduced homework questions throughout our introductory theory cours

Francisco Bersetche, Enrique Otarola, Daniel Quero

We analyze a nonlocal coupled system arising as the Euler--Lagrange equations of an energy functional involving regional fractional Laplacians of orders $s_1$ and $s_2$ ($ 0 < s_1,s_2 < 1$), each acting on a separate disjoint domain and coupled through a nonlocal interaction term depending on a kernel $J$. Under suitable assumptions on the domains and the ke

Sean Nian, Jiahao Fang, Qilong Feng, Zhiyu Wu

KV cache restoration has emerged as a dominant bottleneck in serving long-context LLM workloads, including multi-turn conversations, retrieval-augmented generation, and agentic pipelines. Existing approaches treat restoration as a per-request tradeoff between recomputation and I/O transfer, recomputing KV states from scratch or offloading them from external

Peter J. Forrester

The Riesz gas in one-dimension consists of particles interacting via a pair potential, ${\rm sgn}(s) |x - x'|^{-s}$, $s \ne 0$ and $-\log | x - x'|$ for $s=0$. In the infinite density limit, with the particle support the interval $[-1,1]$, we apply a functional derivative method due to Beenakker to compute the covariance of two smooth linear statistics for t

Keenan Powell, Peihong Yu, Pratap Tokekar

Many Multi-Agent Reinforcement Learning (MARL) agents fail to adapt properly to cooperating with agents trained with the same objectives but different seeds, algorithms, or other training differences. This is the problem of Zero-Shot Coordination (ZSC), which focuses on training agents to cooperate well with unknown agents. ZSC has been studied for a variety

Sarvesh Joshi, Jingye Tan, Craig M. Hamel, Stavros Gaitanaros

Architected metamaterials like foams and lattices exhibit complex responses governed by microstructural instabilities, localization, and phase-transition-like phenomena. Their behavior is further affected by heterogeneities inherent in their microstructure often caused through manufacturing processes. In this study we extend a gradient-enhanced, nonlocal con

José Gaitán, Carlos Garzón, José Madrid

We study the problem of finding the "smoothest'' local average of a function $f \in \ell^2(\mathbb{Z})$ when we consider its convolution with suitable kernels $u$. The measurement of smoothness is as follows: Given a positive integer $k$, we aim to minimize the constant \begin{equation*} \sup_{0 \neq f \in \ell^2(\mathbb{Z})} \frac{\|\nabla^{k}(u\ast f)\|_{\

Gábor Damásdi, Laurentiu Ploscaru

Let $X=\{x_1,\dots,x_n\}\subset \mathbb{R}^2$ be a finite set of points of diameter at most $1$. It is natural to expect that if many pairs $(x_i,x_j)$ lie at distance close to $1$ from each other, then some clustering phenomenon must occur, implying that a significant number of these pairs are also very close to each other. %For $0<\varepsilon<1$, we call a

Hassan Karim, Sai Sitharaman, Deepti Gupta, Danda B. Rawat

Agentic AI is crossing trust boundaries faster than current risk models can represent. Existing approaches provide one of two partial views. They either describe failure mechanisms without producing a transferable residual-risk estimate, or they produce a risk estimate while treating the internal failure path as a black box. We couple those two views by prop

Lars Malmqvist

Achieving reliable control of Large Language Models (LLMs) requires a precise, scalable understanding of how they interpret linguistic cues. We introduce a rigorous framework using Shapley values to quantify the steering effect of individual adjectives on model performance, moving beyond anecdotal heuristics to principled attribution. Applying this method to

Ardit Krasniqi, Luan Vejsiu, Elira Dervishi

Test-time scaling (TTS) has emerged as a powerful paradigm for improving the reasoning performance of large language models (LLMs) by investing additional compute at inference time. A central component of TTS is the \emph{verifier}, which selects or scores candidate solutions to guide the search process. While prior work has explored the benefit of verificat

Jinseok Chung, Minkyoung Song, Hyunji Jung, Namhoon Lee

In-Context Learning (ICL) allows LLMs to adapt to new tasks from a few demonstrations, but its reliability remains a concern: predictions are highly sensitive to both prompt design and the model&#39;s ability to understand the context, obscuring whether failures arise from data properties or model limitations. Uncertainty decomposition-separating aleatoric f

Yiming Ni, Zhi-Qi Cheng, Jiayu Li, Wei Cheng

Sign languages are expressive visual languages used by Deaf and Hard-of-Hearing (DHH) communities. Despite substantial progress in sign-language recognition, translation, and production, advances remain constrained by fragmented datasets, inconsistent annotations, and limited linguistic coverage. Existing benchmarks often fail to reflect real-world communica

Chirag Shinde

We present Density Field State Space Models (DF-SSM), a framework for compressing SSMs to a 1-bit scaffold with int8 low-rank correction. Applied to Mamba-2 1.3B, we achieve a 278 MB model (9.7x smaller than the 2.7 GB FP16 teacher) that runs at 21.4x faster inference on GPU (batch=1, relative to the mamba-ssm reference implementation) while maintaining down

Nick Moës, Toni Lorente, Amin Oueslati, Jonathan Smith

Frontier artificial intelligence is reshaping all aspects of society, from economic output or military capability to democratic institutions. The EU is entering this transformation from a position of structural dependence: frontier models originate almost exclusively from the United States or China, the US holds approximately sixteen times the EU&#39;s AI su

Zilong Liu, Krzysztof Janowicz, Mina Karimi

(Gen)AI diversity is not merely an ethical issue. From the perspective of geographic information science (GIScience), it could be interpreted as a function of uncertainty and as a form of cognitive bias, embedded in AI outputs. Recent work has sought to develop information-theoretic diversity measures and apply them to evaluate AI-chatbot outputs in a geogra

Zilong Liu, Krzysztof Janowicz, Gengchen Mai, Song Gao

Among the many challenges hindering the responsible development and deployment of AI, arguably none has faced more intense scrutiny than bias in its various forms. This underscores the widespread concerns across AI researchers that model outputs, e.g., from generative AI, may encode structural distributional imbalances (stemming from training data or model d

Claudio Novelli, Javier Argota Sanchez-Vaquerizo, Jennifer Cyr, Giuliano Formisano

Political parties strive to understand their electorates, and focus groups are a vital tool in these efforts. AI-enhanced simulation technologies (AESTs) enable synthetic focus groups in a fraction of the time (and cost), raising the question of when and how such simulated evidence can be used in campaign research. This paper develops a decision matrix to he

Qingwen Pu, Kun Xie, Chenyu Yan

Coastal flooding increasingly threatens transportation infrastructure, yet the analytical tools needed for resilience management remain difficult for many non-specialist practitioners to use. This study presents TransResAI, a compound AI system that supports analysis of flood-aware transportation resilience via natural-language interactions. The system integ

Pasquale Ardimento, Mario Luca Bernardi, Marta Cimitile, Samuele Latorre

This study analyzes COVID-19 care pathways using the COVID Data for Shared Learning dataset. We build a transparent, reproducible pipeline that transforms heterogeneous clinical tables into a process-mining-ready event log and applies discovery, declarative conformance checking, and outcome analysis. The reconstructed pathways highlight the monitoring backbo

Angxuan Chen, Jiyou Jia

As Generative AI (GenAI) becomes integral to education, fostering GenAI literacy is critical. However, current assessments largely rely on self-reported scales, lacking insights into how literacy manifests in actual learning processes. This study leverages Learning Analytics (LA) to bridge this gap. We collected interaction logs from 162 university students

Divyanshu Kumar Singh, Dipto Das, Deepika Rama Subramanian, Koustuv Saha

Text-to-Image (T2I) models have shown promising utility across various domains. However, such models are also amplifying harmful societal biases in their outputs. In the context of South Asia, recent work has shown caste biases and stereotypes are being perpetuated through Generative AI (GenAI) systems. While this research offers extremely relevant insight i

Zhenxiao Fu, Lei Jiang, Fan Chen

Quantum computing remains in the Noisy Intermediate-Scale Quantum (NISQ) era, where the performance is highly constrained to noise. Addressing the limitation often requires hardware-facing capabilities beyond gate-sequence circuit specification, including mid-circuit measurement and classical feedback for quantum error correction (QEC), precise timing contro

Maissa Abir Smaili, Eren Sadikoglu, Ransalu Senanayake

Alzheimer&#39;s disease is a neurodegenerative disorder marked by progressive declines in memory and language that reduce independence in daily life, motivating socially assistive robotic support. This paper presents MEMOR-E, a mobile quadruped robot with an interactive tablet interface that assists patients and caregivers through medication reminders, routi

Sebastien Kawada

How do multi-turn reasoning systems fail? The expected answer is logical contradiction, in which the system&#39;s maintained state becomes unsatisfiable. We show that the dominant mode is instead satisfiable drift, where the internal state stays consistent while the returned answer silently violates prior commitments. We build DRIFT-Bench (Decomposing Reason

Xirui Liu, Sihang Zhou, Yanning Hou, Rong Zhou

Web agents require both high-level reasoning (for task decomposition) and low-level interactions (for page elements manipulation) to conduct different tasks. However, these knowledge types differ fundamentally: reasoning knowledge (e.g., booking a flight requires first searching for routes) is abstract and transferable across websites, while interaction know

Long Zhang, Zi-bo Qin, Wei-neng Chen

Large language models (LLMs) increasingly fuse heterogeneous inputs in ubiquitous systems. Yet, how LLMs implicitly allocate authority when sensor measurements and user claims conflict remains unexamined, raising critical reliability concerns for deployments where physical sensing must retain priority. Unlike explicit traditional fusion, LLMs bury authority

Hamidou Tembine

This article develops an intergenerational mean-field-type game (MFTG) to model Mali&#39;s and neighbouring countries multi-actor conflict ecosystem, which includes formal state forces, traditional hunters, nonstate militias, jihadists, criminal networks, civil societies, and international proxies. Each decision-maker (agent, a group of agents or representat

Manos Kamarianakis, Pandelis Sofianos, George Papagiannakis

What symbolic format should an LLM emit for reliable 3D scene editing from natural language, and does algebraic structure help beyond compact syntax? We evaluate Conformal Geometric Algebra (CGA) as a compact symbolic interface against a verbose Euclidean 4$\times$4 matrix baseline and a non-CGA Compact SE3 control in a natural-language 3D editing pipeline w

Nanxu Gong, Zixin Chen, Haotian Li, Zishu Zhao

Improving the Theory of Mind (ToM) capability of Large Language Models (LLMs) is crucial for effective social interactions between these AI models and humans. However, the existing benchmarks often measure ToM capability improvement through story-reading, multiple-choice questions from a third-person perspective, while ignoring the first-person, dynamic, and

Athanasios Papanikolaou, Athanasios Tziouvaras, Pavlos Stoikos, Apostolos Xenakis

Early detection of plant diseases is critical for improving crop productivity, while it also facilitates the foundations of precision agriculture. Recent advances in distributed deep learning have enabled plant disease classification models to be trained across geographically distributed agricultural sensing infrastructures. However, deploying such systems i

Paolo Proia, Mauro Sbragaglia, Giacomo Falcucci

We leverage 3D numerical simulations to study phase change materials (PCMs) cells under the effect of buoyancy forces. The solid PCM is heated from a source boundary, triggering melting. The source features multiple solid fins that protrude into the PCM cell; the impact of the fins and their number is investigated by designing and testing equivalent (in term

Jeehyun Yang, Danica J. Adams, Renyu Hu, Yuk L. Yung

We propose a simple and efficient pathway for the formation of precursors to core nucleobases in DNA and RNA using a suite of computational chemistry methods. Benzene, which is thermochemically stable in N2- or CO2-dominated atmospheres, could have formed via upper-atmospheric photochemistry or surface lightning and accumulated on the early Earth or Mars. Ho

Juan Pablo Arbelaez

We investigate the absorption cross sections of electromagnetic perturbations propagating on a four-dimensional brane in the background of higher-dimensional regular black holes arising in quasi-topological gravity. Employing a WKB-based approach for the computation of greybody factors, we analyze the impact of higher-curvature corrections and spacetime dime

Skye Strain, Nicoló Cibrario, Michela Negro, Eric Burns

Gamma-ray bursts (GRBs) are the most energetic bursts of light in our universe, and rapid progenitor association of these events can lead to targeted and optimized follow-up observations, ultimately providing better insights about the physics involved. In this note, we investigate a semi-supervised machine learning algorithm, that utilizes label propagation,

Jelena Ritter, Wuji Wang, Dominika Wylezalek, Carlos De Breuck

High-redshift radio galaxies (HzRGs) are among the most massive galaxies in the Universe and sites of extreme active galactic nuclei (AGN) feedback processes, powering energetic radio jets. They are typically embedded in giant Ly$α$ halos that are known to extend over $100\,\text{kpc}$ into the circumgalactic medium (CGM). In this paper, we target the Ly$α$

Jaël Champagne Gareau, Daniel Lemire

Converting binary integers to variable-length decimal strings is a fundamental operation in computing. Conventional fast approaches rely on recursive division and small lookup tables. We propose a SIMD-based algorithm that leverages integer multiply-add instructions available on recent AMD and Intel processors. Our method eliminates lookup tables entirely an

Eugenio Carretta

The bifurcated age-metallicity relation of globular clusters (GCs) in the Milky Way (MW) shows that GCs are either originated in situ or accreted into the Galaxy from former satellites of the MW. The effects of the Galactic tidal field can leave signatures on the dynamical evolution and structural properties of GCs. We present a homogeneous census of dynamic

Andrea Boccia, Marco Chianese

We study the formation of asteroid-mass Primordial Black Holes (PBHs) as a dark matter candidate in supersymmetric extensions of the Standard Model. We show that the presence of heavy particles predicted in the Minimal Supersymmetric Standard Model (MSSM) can lead to a transient softening of the equation of state of the Universe during their non-relativistic

Minh-Khoa Le-Phan, Minh-Hoang Le, Trong-Le Do, Minh-Triet Tran

Current deepfake detection models achieve state-of-the-art performance on pristine academic datasets but suffer severe spatial attention drift under real-world compound degradations, such as blurring and severe lossy compression. To address this vulnerability, we propose a foundation-driven forensic framework that integrates an extreme compound degradation e