November 2025 arXiv papers — page 58

Yu Xia, Sungchul Kim, Tong Yu, Ryan A. Rossi

Agentic recommendations cast recommenders as large language model (LLM) agents that can plan, reason, use tools, and interact with users of varying preferences in web applications. However, most existing agentic recommender systems focus on generic single-agent plan-execute workflows or multi-agent task decomposition pipelines. Without recommendation-oriente

Dilli Babu Porlapothula, Pralay Chakrabarty, Ananya Lakshmi Ravi, Kurian Polachan

In this work, we present ioPUF+, which incorporates a novel Physical Unclonable Function (PUF) that generates unique fingerprints for Integrated Circuits (ICs) and the IoT nodes encompassing them. The proposed PUF generates device-specific responses by measuring the pull-up and pull-down resistor values on the I/O pins of the ICs, which naturally vary across

Sultan Alrashed, Chadi Helwe, Francesco Orabona

Although the community has tackled the acquisition of high-quality Arabic pretraining data, we still lack large-scale, multi-turn Arabic datasets that include reasoning and tool calling. Naive translation can work at the pretraining scale, but post-training demands much higher quality, which requires a stricter approach to dataset curation. In this work, we

Mikiya M. Takahashi, Tomohisa Kawashima, Ken Ohsuga

By investigating the time-variable 230 GHz images using ray-tracing general relativistic radiative transfer calculation, we propose a novel method for estimating the spin parameter of the supermassive black hole at the M87 center by utilizing the sudden and short-term increase in emissivity in the innermost region of the accretion disk. It is found that the

Dana Arad, Yonatan Belinkov, Hanjie Chen, Najoung Kim

Mechanistic interpretability (MI) seeks to uncover how language models (LMs) implement specific behaviors, yet measuring progress in MI remains challenging. The recently released Mechanistic Interpretability Benchmark (MIB; Mueller et al., 2025) provides a standardized framework for evaluating circuit and causal variable localization. Building on this founda

Matteo Guenci, Ivan Heibi, Chiara Parravicini, Silvio Peroni

While Crossref makes available more than 1.8 billion bibliographic references from publications for which it provides a DOI, more than 698 million of these references do not specify a DOI, making the creation of a formal citation link from the citing entity and the cited entity problematic. In this article, we propose an analysis of Crossref bibliographic re

Adil Belhaj, Maryem Jemri

Inspired by non-commutative geometry in string theory, we propose extended derivatives in black hole physics by incorporating a real antisymmetric tensor of rank 2 carrying similarities of certain stringy fields. Using gauge theory formulation of gravity via de Sitter group theory, we first find the associated black hole solutions by solving the Einstein fie

Ankit Singh, Vinay Vaibhav, Caterina Czibula, Astrid Macher

Glassy polymers are central to engineering applications, yet their viscoelastic response over broad frequency and temperature ranges remains difficult to characterize. We extend non-affine deformation theory by incorporating a time-dependent memory kernel within the Generalized Langevin Equation for atomistic non-affine motions, yielding frequency-dependent

Mohammad Nour Al Awad, Sergey Ivanov, Olga Tikhonova, Ivan Khodnenko

Large language models (LLMs) can reshape information processing by handling data analysis, visualization, and interpretation in an interactive, context-aware dialogue with users, including voice interaction, while maintaining high performance. In this article, we present Talk2Data, a multimodal LLM-driven conversational agent for intuitive data exploration.

Van Thuy Hoang, O-Joun Lee

Recent pre-training strategies for molecular graphs have attempted to use 2D and 3D molecular views as both inputs and self-supervised signals, primarily aligning graph-level representations. However, existing studies remain limited in addressing two main challenges of multi-view molecular learning: (1) discovering shared information between two views while

Aayush Kumar, Sanket Mhatre

Large Language Models have been widely been adopted by users for writing tasks such as sentence completions. While this can improve writing efficiency, prior research shows that LLM-generated suggestions may exhibit cultural biases which may be difficult for users to detect, especially in educational contexts for non-native English speakers. While such prior

An V. Q. Huynh, Minh B. Nguyen, Nhan X. V. Nguyen, Minh Q. Vu

We prove that for any two definable germs in a polynomially bounded o-minimal structure, there exists a critical threshold $\alpha_0 \in (0,1)$ such that if these germs are bi-$\alpha$-H"older equivalent for some $\alpha \ge \alpha_0$, then they satisfy the following: \begin{itemize}[label=$\circ$] \item The Lipschitz normal embedding (LNE) property is prese

Jean Gutt, Vinicius G. B. Ramos, Shira Tanny

We complete the spectral characterization of Besse and Zoll Reeb flows on the standard contact sphere $S^{2n-1}$ initiated by Ginzburg-G\"urel-Mazzucchelli. Roughly speaking, it states that a Reeb flow on the boundary of any star-shaped domain in $\mathbb{R}^{2n}$ is Besse if and only if it has $n$ coinciding Ekeland-Hofer capacities, and that it is Zoll if

Takaki Kiyozumi, Soshi Yoshida, Yuta Higa, Keisuke Motoda

In the maintenance of optical fiber networks, there is a growing demand for high-precision measurement of optical loss distribution and fault locations over long distances. In this study, we propose an OCDR method incorporating periodic pseudo-random modulation (PPRM), and demonstrate that it enables the acquisition of loss distribution based on Rayleigh sca

Yuxiang Nie, Han Wang, Yongjie Ye, Haiyang Yu

This paper introduces ChineseVideoBench, a pioneering benchmark specifically designed for evaluating Multimodal Large Language Models (MLLMs) in Chinese Video Question Answering. The growing demand for sophisticated video analysis capabilities highlights the critical need for comprehensive, culturally-aware evaluation frameworks. ChineseVideoBench addresses

Xiangzhou Zhu, Patrick Rinke, David A. Egger

The presence of defects strongly influences semiconductor behavior. However, predicting the electronic properties of defective materials at finite temperatures remains computationally expensive even with density functional theory due to the large number of atoms in the simulation cell and the multitude of thermally accessible configurations. Here, we present

Monte MacDiarmid, Benjamin Wright, Jonathan Uesato, Joe Benton

We show that when large language models learn to reward hack on production RL environments, this can result in egregious emergent misalignment. We start with a pretrained model, impart knowledge of reward hacking strategies via synthetic document finetuning or prompting, and train on a selection of real Anthropic production coding environments. Unsurprisingl

Jinhao Li, Sarah M. Erfani, Lei Feng, James Bailey

Aligning large-scale commercial models with user intent is crucial to preventing harmful outputs. Current methods rely on human supervision but become impractical as model complexity increases. When models surpass human knowledge, providing accurate feedback becomes challenging and inefficient. A novel solution proposed recently is using a weaker model to su

Parker Whitfill, Ben Snodin, Joel Becker

METR's time horizon metric has grown exponentially since 2019, along with compute. However, it is unclear whether compute scaling will persist at current rates through 2030, raising the question of how possible compute slowdowns might impact AI agent capability forecasts. Given a model of time horizon as a function of training compute and algorithms, along w

Shuo Li, Lei Shi

In this paper, we introduce Schur elements for supersymmetrizing superalgebras. We show that the cyclotomic Hecke-Clifford algebra $\mathcal{H}^f_{c}(n)$ is supersymmetric if $f=f^{(\mathtt{0})}_{\underline{Q}}$ and, symmetric if $f=f^{(\mathtt{s})}_{\underline{Q}}$ and an invertibility condition holds. In the semisimple case, we compute the Schur elements f

Chinmay Karkar, Paras Chopra

Large Language Models (LLMs) demonstrate partial forecasting competence across social, political, and economic events. Yet, their predictive ability varies sharply with domain structure and prompt framing. We investigate how forecasting performance varies with different model families on real-world questions about events that happened beyond the model cutoff

Heejoon Koo

A decade of rapid advances in artificial intelligence (AI) has opened new opportunities for clinical decision support systems (CDSS), with large language models (LLMs) demonstrating strong reasoning abilities on timely medical tasks. However, clinical texts are often degraded by human errors or failures in automated pipelines, raising concerns about the reli

Teo Banica

This is an introduction to the finite groups, with focus on the groups of permutations and reflections, and more generally, on the finite groups of unitary matrices. We first discuss the basics of group theory, featuring the cyclic, dihedral and symmetric groups, and the structure result for finite abelian groups. Then we study the complex reflection groups,

Adam Chudecki, Michał Dobrski

Algebraically general para-K\"ahler Einstein spaces equipped with 3D algebras of infinitesimal symmetries are considered. It is shown that if the algebra contains 2D trivial subalgebra then vacuum Einstein field equations with cosmological constant can be reduced to a single, first-order differential equation. One of the cases is solved explicitly. Hence, a

Ajay Tak, Mayank Baranwal

Many large-scale constrained optimization problems can be formulated as bilevel distributed optimization tasks over undirected networks, where agents collaborate to minimize a global cost function while adhering to constraints, relying only on local communication and computation. In this work, we propose a distributed stochastic gradient aggregation scheme a

Shu Yang, Fahim Tanfeez Mahmood, C. Ricardo Constante-Amores

Capillary retraction of liquid ligaments is well understood for Newtonian fluids, whereas viscoplastic effects remain comparatively unexplored. Here, we consider Herschel-Bulkley fluids, which incorporate both yield stress and shear-rate-dependent viscosity, thereby introducing a spatially varying effective viscosity that is absent in simpler yield-stress mo

Jitendra Parmar, Praveen Singh Thakur

Open-world machine learning is an emerging technique in artificial intelligence, where conventional machine learning models often follow closed-world assumptions, which can hinder their ability to retain previously learned knowledge for future tasks. However, automated intelligence systems must learn about novel classes and previously known tasks. The propos

Plein Versace

Implicit Neural Representations (INRs) have emerged as a powerful paradigm for representing signals such as images, 3D shapes, signed distance fields, and radiance fields. While significant progress has been made in architecture design (e.g., SIREN, FFC, KAN-based INRs) and optimization strategies (meta-learning, amortization, distillation), existing approac

Peter Siegel, Federico Tombari, Marc Pollefeys, Daniel Barath

We have introduced SegSplat, a novel framework designed to bridge the gap between rapid, feed-forward 3D reconstruction and rich, open-vocabulary semantic understanding. By constructing a compact semantic memory bank from multi-view 2D foundation model features and predicting discrete semantic indices alongside geometric and appearance attributes for each 3D

Chuang Peng, Renshuai Tao, Zhongwei Ren, Xianglong Liu

Automatic X-ray prohibited items detection is vital for security inspection and has been widely studied. Traditional methods rely on visual modality, often struggling with complex threats. While recent studies incorporate language to guide single-view images, human inspectors typically use dual-view images in practice. This raises the question: can the secon

Guijun Liu, Yuwen Cao, Tomoaki Ohtsuki, Jiguang He

Deep autoencoder (DAE) frameworks have demonstrated their effectiveness in reducing channel state information (CSI) feedback overhead in massive multiple-input multiple-output (mMIMO) orthogonal frequency division multiplexing (OFDM) systems. However, existing CSI feedback models struggle to adapt to dynamic environments caused by user mobility, requiring re

Plein Versace

Implicit Neural Representations (INRs) have emerged as a powerful paradigm for representing signals such as images, audio, and 3D scenes. However, existing INR frameworks -- including MLPs with Fourier features, SIREN, and multiresolution hash grids -- implicitly assume a \textit{global and stationary} spectral basis. This assumption is fundamentally misalig

Francçois Gay-Balmaz

We develop a covariant variational framework for relativistic electromagnetic continua (fluids and solid) based on Hamilton's principle formulated directly in the material description. The approach extends the geometric theory of relativistic continua introduced in Part I to include polarization, magnetization, and general elastic-electromagnetic coupling th

Timing Yang, Sucheng Ren, Alan Yuille, Feng Wang

Text-to-video generation has surged in interest since Sora, yet open-source models still face a data bottleneck: there is no large, high-quality, easily obtainable video-text corpus. Existing public datasets typically require manual YouTube crawling, which yields low usable volume due to link rot and access limits, and raises licensing uncertainty. This work

Narinder Kumar Wadhawan

This paper introduces a new numerical method for approximating the Lambert W function in the real domain. The method transforms the function into a simpler form that allows iterative refinement of an initial guess. Two iterative strategies are proposed for positive inputs, and the method is extended to handle negative inputs within a defined range. Unlike st

Timing Yang, Guoyizhe Wei, Alan Yuille, Feng Wang

Mamba has recently garnered attention as an effective backbone for vision tasks. However, its underlying mechanism in visual domains remains poorly understood. In this work, we systematically investigate Mamba's representational properties and make three primary contributions. First, we theoretically analyze Mamba's relationship to Softmax and Linear Attenti

Andreea Elena Drăgnoiu, Andrei Ciobanu, Ruxandra F. Olimid

Self-Sovereign Identity (SSI) grants holders full ownership and control of their digital identities, being the ultimate digital identity model. Operating in a decentralized manner, SSI enables the verification of claims, including privacy-preserving mechanisms. Blockchain, which can be used to implement a Verifiable Data Registry (VDR), is often considered o

Shijian Wang, Runhao Fu, Siyi Zhao, Qingqin Zhan

Text-to-Image (T2I) generation has long been an open problem, with compositional synthesis remaining particularly challenging. This task requires accurate rendering of complex scenes containing multiple objects that exhibit diverse attributes as well as intricate spatial and semantic relationships, demanding both precise object placement and coherent inter-o

Alessandro Giovagnoli

The Quantum Approximate Optimization Algorithm (QAOA) is a promising variational quantum algorithm introduced to tackle classically intractable combinatorial optimization problems. This tutorial offers a comprehensive, first-principles introduction to QAOA and its properties, focusing on its application to Quadratic and Polynomial Unconstrained Binary Optimi

Haohan Wang, Xu Shi, Hengyu Zhang, Yashuai Cao

The channel knowledge map (CKM) enables efficient construction of high-fidelity mapping between spatial environments and channel parameters via electromagnetic information analysis. Nevertheless, existing studies are largely confined to single-antenna systems, failing to offer dedicated guidance for multi-antenna communication scenarios. To address the inher

Joachim Diederich

This paper demonstrates that progressive localization, the gradual increase of attention locality from early distributed layers to late localized layers, represents the optimal architecture for creating interpretable large language models (LLMs) while preserving performance. Through systematic experimentation with GPT-2 fine-tuned on The Psychology of Artifi

Jiaxun Sun, Hengyu Xue, Yuyang Zhao

We derive a computable closed-form upper bound on the Hausdorff distance between a truncated minimal robust positively invariant (mRPI) set and its infinite-horizon limit. The bound depends only on a disturbance-set size measure and an induced-norm contraction factor of the system matrix, and it yields an explicit, fully analytic horizon-selection rule that

Xiyang Wu, Zongxia Li, Jihui Jin, Guangyao Shi

Vision Language Models (VLMs) perform well on standard video tasks but struggle with physics-related reasoning involving motion dynamics and spatial interactions. We present a novel approach to address this gap by translating physical-world context cues into interpretable representations aligned with VLM perception, comprehension, and reasoning. We introduce

Sofiane Bouarroudj, Quentin Ehret, Abdenacer Makhlouf, Nurtas Shyntas

The main goal of this paper is to introduce the notion of restricted Lie-Rinehart superalgebra over a field of characteristic $p>2$, motivated by a generalization of Hochschild's lemma to the super setting. We extend Schauenburg's proof of Hochschild's lemma to Lie-Rinehart superalgebras and we prove a superized version that serves as a foundation for our co

Wen-Jun Tan, Chen-Wei Wang, Shao-Lin Xiong, Shi-Jie Zheng

The Space-based multiband astronomical Variable Objects Monitor (SVOM), detected its first short gamma-ray burst, GRB 240715A, in-flight, which was jointly observed by Fermi. Based on observational data of SVOM/GRM and Fermi/GBM, we perform a comprehensive temporal and spectral analysis for individual pulse in the prompt emission of this burst, and novel cha

Zenghao Chai, Chen Tang, Yongkang Wong, Xulei Yang

3D pose transfer aims to transfer the pose-style of a source mesh to a target character while preserving both the target's geometry and the source's pose characteristic. Existing methods are largely restricted to characters with similar structures and fail to generalize to category-free settings (e.g., transferring a humanoid's pose to a quadruped). The key

Manon Berriche

This thesis addresses two paradoxes: (1) why empirical studies find that fake news represent only a small share of the information consulted and shared on social media despite the absence of editorial control or journalistic norms, and (2) how political polarization has intensified even though users do not appear especially receptive to fake news. To investi

Xiaoming He, Gaofeng Wang, Huajun Cui, Rui Yuan

Autonomous Aerial Vehicle (AAV)-assisted Internet of Things (IoT) represents a collaborative architecture in which AAV allocate resources over 6G links to jointly enhance user-intent interpretation and overall network performance. Owing to this mutual dependence, improvements in intent inference and policy decisions on one component reinforce the efficiency

Zilong Chen, Huan-ang Gao, Delin Qu, Haohan Chi

Existing dynamic scene reconstruction methods based on Gaussian Splatting enable real-time rendering and generate realistic images. However, adjusting the camera's focal length or the distance between Gaussian primitives and the camera to modify rendering resolution often introduces strong artifacts, stemming from the frequency constraints of 4D Gaussians an

Jean-Christophe Novelli, Jean-Yves Thibon

We investigate the geode and some of its generalizations from the point of view on noncommutative symmetric functions.

Dangyang He

Let $M=(0,\infty)_r\times Y$ be a $d$-dimensional ($d\ge 3$) metric cone with metric<br/>$g=dr^2+r^2h$, where $(Y,h)$ is a closed Riemannian manifold. Let<br/>$H=\Delta+V_0/r^2$ be the associated Schrodinger operator, with<br/>$V_0\in C^\infty(Y)$ satisfying the positivity condition<br/>$\Delta_Y+V_0+(d-2)^2/4>0$. First, we complement previous results by pro

Marvin Hofer, Erhard Rahm

Building high-quality knowledge graphs (KGs) from diverse sources requires combining methods for information extraction, data transformation, ontology mapping, entity matching, and data fusion. Numerous methods and tools exist for each of these tasks, but support for combining them into reproducible and effective end-to-end pipelines is still lacking. We pre

Habib Ammari, Jiayu Qiu, Alexander Uhlmann

We consider interface modes in block disordered subwavelength resonator chains in one dimension. Based on the capacitance operator formulation, which provides a first-order approximation of the spectral properties of dimer-type block resonator systems in the subwavelength regime, we show that a two-fold topological characterization of a block disordered reso

Matan Even Tzur, Chen Mor, Noa Yaffe, Michael Birk

Until recently, attosecond optical spectroscopy and quantum optics evolved along non-overlapping directions. In attosecond science, attosecond pulses have been regarded as classical waves, applied to probe electron dynamics on their natural time scale. Here, we transfer fundamental concepts of quantum optics into attosecond physics, enabling control of both

Ayumi Ozawa, Yoji Kawamura

We develop a phase reduction method for reaction-diffusion systems with a discrete delay. On the basis of the recent developments in the phase reduction theory for infinite-dimensional systems, we introduce a bilinear form tailored to spatially extended systems involving a discrete delay. By solving the adjoint equation associated with the bilinear form, we

Alexandros Stergiou

How do video understanding models acquire their answers? Although current Vision Language Models (VLMs) reason over complex scenes with diverse objects, action performances, and scene dynamics, understanding and controlling their internal processes remains an open challenge. Motivated by recent advancements in text-to-video (T2V) generative models, this pape

Jiachen Zhu, Fangjiong Chen, Jie Wu, Ming Xia

This paper proposes a constant false alarm rate (CFAR) target detection algorithm based on the CLEAN concept and truncated statistics to mitigate the non-homogeneity of reference samples caused by sidelobe contamination and other abnormal interferences within the reference window. The proposed algorithm employs truncated statistics to separate target and noi

Mou Yan, Yu-Liang Tao, Yichong Hu, Zhenxing Cui

Topological phases have been extensively studied primarily in crystalline systems with translational symmetry. Recent theoretical studies, however, have demonstrated the existence of topological phases in quasicrystals that are absent in crystals. Despite numerous experimental observations of topological phases in various crystalline systems, observing these

Xiaoli Lang, Boris A. Malomed, Liangwei Dong

We systematically investigate the existence, stability, and propagation dynamics of multipole-mode (necklace-shaped) solitons in the two-dimensional model of an optical medium with the defocusing saturable nonlinearity and an annular potential trough. Various families of stable multipole solitons trapped in the trough, from dipole, quadrupole, and octupole o

Sukila Khatun, Khairul Hasan, Amar Kumar Banerjee

In this paper we study the notion of rough $\mathcal{I}$-statistical convergence of sequences in a partial metric space as an extension work of both the notions of rough statistical and rough ideal convergence. Here we define rough $\mathcal{I}$-statistical limit set and discuss some relevant properties associated with this set.

Muhammad Bilal, Zafar Qazi, Marco Canini

The rise of Generative AI Search is fundamentally transforming how users and intelligent systems interact with the Internet. LLMs increasingly act as intermediaries between humans and web information. Yet the web remains optimized for human browsing rather than AI-driven semantic retrieval, resulting in wasted network bandwidth, lower information quality, an

Sigrid Helene Strand, Thomas Wiedemann, Bram Burczek, Dmitriy Shutin

Search and rescue missions are often critical following sudden natural disasters or in high-risk environmental situations. The most challenging search and rescue missions involve difficult-to-access terrains, such as dense forests with high occlusion. Deploying unmanned aerial vehicles for exploration can significantly enhance search effectiveness, facilitat

Zitong Xu, Dake Shen, Yaosong Du, Kexiang Hao

Recent advances in AIGC (Artificial Intelligence Generated Content) models have enabled significant progress in image and video generation. However, users still struggle to obtain content that aligns with their preferences due to the difficulty of crafting detailed prompts and the lack of mechanisms to retain their preferences. To address these challenges, w

Voalaza Mahavily Romuald Aubert, Benjamin Randrianirina

Using the lattice paths in $\mathbb{N}\times\mathbb{N}$, we derive a general formula for sequences $\big(T(n,k)\big)$ satisfying the recurrence relation of the form: \begin{equation*} T((n,k)=a_{n,k}T(n-1,k)+b_{n,k}T(n-1,k-1). \end{equation*} We apply this result to the case where $a_{n,k}=a_0+a_1k+a_2n$ and $b_{n,k}=b_0+b_1k+b_2n$. This leads to explicit ex

Serkan Akkoyun, Cafer Mert Yeşilkanat, Paul Stevenson

The fission barrier energy is a fundamental property of nuclear structure that governs the stability of nuclei against fission, directly affecting their spontaneous fission half-lives and the formation of superheavy elements. However, because it can only be measured indirectly, it also enables the emergence of alternative, complementary, fast, and accurate p

Minghui Ding

We employ the continuum Schwinger function method to investigate the unpolarized valence-quark transverse-momentum-dependent parton distribution function (TMD) of the pion at the hadron scale. The first seventeen generalized Mellin-transverse moments, constructed from lightlike and transverse vectors, are computed and found to be well described by a factoriz

Qingzheng Gao, Xiao-Jin Liu, Zhi-Qiang You, Zheng Li

We present a comprehensive re-evaluation of pulsar radio spectra using the largest curated dataset of calibrated flux densities to date, comprising 897 pulsars, and employing a robust Bayesian framework for model comparison alongside frequentist methods. Contrary to the established consensus that pulsar spectra are predominantly simple power laws, our analys

Haoyan Fu, Zhida Qin, Shixiao Yang, Haoyao Zhang

Sequential recommendation (SR) is widely deployed in e-commerce platforms, streaming services, etc., revealing significant potential to enhance user experience. However, existing methods often overlook two critical factors: irregular user interests between interactions and highly uneven item distributions over time. The former factor implies that actual user

Wenshuo Gao, Junyi Fan, Jiangyue Zeng, Shuai Yang

Video relighting with background replacement is a challenging task critical for applications in film production and creative media. Existing methods struggle to balance temporal consistency, spatial fidelity, and illumination naturalness. To address these issues, we introduce FlowPortal, a novel training-free flow-based video relighting framework. Our core i

Luca Ornigotti, Darren W. Moore, Radim Filip

Controllable nonlinear quantum interactions are a much sought after target for modern quantum technologies. They are typically difficult and costly to engineer for bespoke purposes. However controllable nonlinearities may have always been in reach via the natural and fundamental forces between quantum particles. The Coulomb interaction between charged partic

Tianyang Xu, Jinjie Gu, Xuefeng Zhu, XiaoJun Wu

With the proliferation of low altitude unmanned aerial vehicles (UAVs), visual multi-object tracking is becoming a critical security technology, demanding significant robustness even in complex environmental conditions. However, tracking UAVs using a single visual modality often fails in challenging scenarios, such as low illumination, cluttered backgrounds,

Dongming Jin, Zhi Jin, Xiaohong Chen, Zheng Fang

In open source software development, the reuse of existing artifacts has been widely adopted to avoid redundant implementation work. Reusable artifacts are considered more efficient and reliable than developing software components from scratch. However, when faced with a large number of reusable artifacts, developers often struggle to find artifacts that can

Jiaming Zhang, Yuyuan Li, Xiaohua Feng, Zhifei Ren

Large language model-based Recommender Systems (LRSs) have demonstrated superior recommendation performance by integrating pre-training with Supervised Fine-Tuning (SFT). However, this approach introduces item-side unfairness. Existing studies primarily attribute this issue to the absence of fairness constraints during SFT and attempt to mitigate unfairness

Weiyi Deng, Chengcheng Han, Zhanhong Lei, Jin Min Yang

In this work we investigate curvature perturbations and non-Gaussianity arising from Higgs modulated reheating in the early Universe. We employ three different methods -- the period-averaging (PA) method, the exact method, and the non-perturbative $\delta N$ formalism -- to compute the power spectrum and bispectrum of curvature perturbations. Our results sho

András Mészáros

For a fixed dimension $k\ge 1$, let us consider the randomly growing simplical complex on the vertex set $\{1,2,\dots,n\}$ defined as follows: We start with the empty complex, and for each $k+1$-element subset $\sigma$ of $\{1,2,\dots,n\}$, we add $\sigma$ and all of its subsets to the complex at some random time $t_\sigma$, where $(t_\sigma)$ are i.i.d. uni

Zi-Han Wang, Peng-Hui Chen, Ya-Ling Zhang, Ming-Hui Huang

The synthesis of superheavy elements strongly relies on the competition of the quasifission and fusion fission dynamics in the fusion-evaporation reactions. The systematics on the formation of superheavy nuclei in the $^{48}$Ca, $^{50}$Ti, $^{51}$V and $^{54}$Cr induced fusion reactions on actinide nuclei $^{232}$Th, $^{231}$Pa, $^{238}$U, $^{237}$Np, $^{242

Kaito Shiku, Kazuya Nishimura, Shinnosuke Matsuo, Yasuhiro Kojima

Spatial transcriptomics (ST) is a novel technology that enables the observation of gene expression at the resolution of individual spots within pathological tissues. ST quantifies the expression of tens of thousands of genes in a tissue section; however, heavy observational noise is often introduced during measurement. In prior studies, to ensure meaningful

Zhe Zhao, Yuheng Yang, Haibin Wen, Xiaojie Qiu

The integration of Large Language Models (LLMs) with Evolutionary Computation (EC) has unlocked new frontiers in scientific discovery but remains shackled by a fundamental constraint: the reliance on an Oracle--an objective, machine-computable fitness function. This paper breaks this barrier by asking: Can evolution thrive in a purely subjective landscape go

James Y. Huang, Wenxuan Zhou, Nan Xu, Fei Wang

The ability of Large Language Models (LLMs) to generate structured outputs that follow arbitrary schemas is crucial to a wide range of downstream tasks that require diverse structured representations of results such as information extraction, table generation, and function calling. While modern LLMs excel in generating unstructured responses in natural langu

Chibuike E. Ugwu, Roschelle Fritz, Diane J. Cook, Janardhan Rao Doppa

Urinary tract infection (UTI) flare-ups pose a significant health risk for older adults with chronic conditions. These infections often go unnoticed until they become severe, making early detection through innovative smart home technologies crucial. Traditional machine learning (ML) approaches relying on simple binary classification for UTI detection offer l

Xuanke Shi, Boxuan Li, Xiaoyang Han, Zhongang Cai

Unified multimodal models that couple visual understanding with image generation have advanced rapidly, yet most systems still focus on visual grounding-aligning language with image regions-while their generative counterpart, linguistic-embedded layout-grounded generation (LELG) for layout-controllable multi-instance generation, remains underexplored and lim

Kiminori Hattori, Shinji Nakata

In this paper, we formulate the real-space Chern number in a supercell framework. In this framework, the overlap matrix between two corners of the Brillouin zone (BZ) is derived from diagonalizing the real-space Hamiltonian with periodic boundary conditions. The path-ordered product of overlap matrices around the BZ boundary forms a Wilson loop, and defines

Sohini Roychowdhury, Adam Holeman, Mohammad Amin, Feng Wei

For online ad-recommendation systems, processing complete user-ad-engagement histories is both computationally intensive and noise-prone. We introduce Dynamix, a scalable, personalized sequence exploration framework that optimizes event history processing using maximum relevance principles and self-supervised learning through Event Based Features (EBFs). Dyn

Xiangwen Cao, Zongyun Chen, Steven J. Miller

We illustrate how to invite and excite students about research by exploring higher-dimensional generalizations of the classical egg drop problem, in which the goal is to locate a critical breaking point using the fewest number of trials. Beginning with the one-dimensional case, we prove that with $k$ eggs and $N$ floors, the minimal number of drops in the wo

Shohei Tanaka, Atsushi Hashimoto, Yoshitaka Ushiku

Scientific posters play a vital role in academic communication by presenting ideas through visual summaries. Analyzing reading order and parent-child relations of posters is essential for building structure-aware interfaces that facilitate clear and accurate understanding of research content. Despite their prevalence in academic communication, posters remain

Svitlana Volkova, Will Dupree, Hsien-Te Kao, Peter Bautista

This paper introduces BRIES, a novel compound AI architecture designed to detect and measure the effectiveness of persuasion attacks across information environments. We present a system with specialized agents: a Twister that generates adversarial content employing targeted persuasion tactics, a Detector that identifies attack types with configurable paramet

Akaki Mamageishvili, Christoph Schlegel, Ko Sunghun, Jinsuk Park

We study the performance of the TimeBoost auction, by comparing cumulative fixed time markout of fast lane trades over the TimeBoost interval to bids for the fast lane. Such comparison allows us to assess how well bids predict future extracted value from the time advantage. The correlation between winning bids and markouts is weak across bidders, suggesting

Hao Liao, Xuanqin Huang, Ping Wang

Characterizing the nonlocal nature of quantum states is a central challenge in the practical application of large-scale quantum computation and simulation. Quantum mutual information (QMI), a fundamental nonlocal measure, plays a key role in quantifying entanglement and has become increasingly important in studying nonequilibrium quantum many-body phenomena,

Mingming Zhao, Xiaokang Wei, Yuanqi Shao, Kaiwen Zhou

Large language models (LLMs) have shown strong potential in automating the design of agentic workflows. However, existing methods still rely heavily on manually predefined operators, limiting generalization and scalability. To address this issue, we propose $A^2Flow$, a fully automated framework for agentic workflow generation based on self-adaptive abstract

Helia Abedini, Saba Rahimi, Reza Vaziri

The accurate identification of brain tumors from magnetic resonance imaging (MRI) is essential for timely diagnosis and effective therapeutic intervention. While deep convolutional neural networks (CNNs), particularly those pre-trained on extensive datasets, have shown considerable promise in medical image analysis, a key question arises when working with li

Sin-Yee Yap, Fuad Noman, Junn Yong Loo, Devon Stoliker

Psychedelics, such as psilocybin, reorganise large-scale brain connectivity, yet how these changes are reflected across electrophysiological (electroencephalogram, EEG) and haemodynamic (functional magnetic resonance imaging, fMRI) networks remains unclear. We present Brain-MGF, a multimodal graph fusion network for joint EEG-fMRI connectivity analysis. For

Syed Mohaiminul Hoque, Naimur Rahman, Md Sakhawat Hossain

This paper introduces the approach of "Gradient Masters" for BLP-2025 Task 1: "Bangla Multitask Hate Speech Identification Shared Task". We present an ensemble-based fine-tuning strategy for addressing subtasks 1A (hate-type classification) and 1B (target group classification) in YouTube comments. We propose a hybrid approach on a Bangla Language Model, whic

Juha Jeon

Deep learning training relies on periodic checkpoints to recover from failures, but unsafe checkpoint installation can leave corrupted files on disk. This paper presents an experimental study of checkpoint installation protocols and integrity validation for AI training on macOS/APFS. We implement three write modes with increasing durability guarantees: unsaf

Martin Hasenbusch

We accurately compute the RG exponents $Y_q$ of large $q$ fields at the $O(2)$ invariant fixed point in three dimensions. We build on an iterative approach that has been previously proposed and is implemented by using the worm algorithm. We simulate an improved XY model, that has next-to-next-to-nearest couplings in addition to nearest ones. In the worm upda

Dhruvil Solanki, Ishmael Apachigawo, Sazzad Khan, Santanu Maity

Alzheimer disease (AD) is considered one of the leading causes of death in the United States, and there is no effective cure for it. Understanding the neuropathological mechanisms underlying AD is essential for identifying early, reliable biomarkers and developing effective therapies. In this paper, we report on a comprehensive multimodal study of AD patholo

Aman Kumar, Ekant Muljibhai Amin, Xian Yeow Lee, Lasitha Vidyaratne

Large Language Models (LLMs) have shown remarkable success in supporting a wide range of knowledge-intensive tasks. In specialized domains, there is growing interest in leveraging LLMs to assist subject matter experts with domain-specific challenges. However, deploying LLMs as SaaS solutions raises data privacy concerns, while many open-source models demand

Hsien-Te Kao, Aleksey Panasyuk, Peter Bautista, William Dupree

Organization's communication is essential for public trust, but the rise of generative AI models has introduced significant challenges by generating persuasive content that can form competing narratives with official messages from government and commercial organizations at speed and scale. This has left agencies in a reactive position, often unaware of how t

Xian Yeow Lee, Lasitha Vidyaratne, Gregory Sin, Ahmed Farahat

Autonomous inspection in hazardous environments requires AI agents that can interpret high-level goals and execute precise control. A key capability for such agents is spatial grounding, for example when a drone must center a detected object in its camera view to enable reliable inspection. While large language models provide a natural interface for specifyi

Allison Brattley, Tomas Opatrny, Kunal K. Das

We present a general algorithm, based on machine learning, which can create optimal unitary operators to implement quantum teleportation in any system with well-defined set of measurements in a relevant entangled basis. We illustrate it with a collective spin model and demonstrate its versatility by applying it to teloportation of single and multiple qubit s

Dongcai Tan, Shunkun Liang, Bin Li, Banglei Guan

Stereo optical measurement techniques, such as digital image correlation (DIC), are widely used in 3D deformation measurement as non-contact, full-field measurement methods, in which stereo calibration is a crucial step. However, current stereo calibration methods lack intuitive optimal pose guidance, leading to inefficiency and suboptimal accuracy in deform