April 2026 arXiv papers — page 47

Houman Mehrafarin, Amit Parekh, Ioannis Konstas

Whether intermediate reasoning is computationally useful or merely explanatory depends on whether chain-of-thought (CoT) tokens contain task-relevant information. We present a mechanistic causal analysis of CoT on GSM8K using activation patching: transferring token-level hidden states from a CoT generation to a direct-answer run for the same question, then m

Ruiling Jiang, Yong Zhang, Houbiao Li

Deep generative models and neural operators have demonstrated significant potential for 3D aerodynamic inference. However, they often face inherent challenges in maintaining physical consistency and preserving high-frequency features, primarily due to spectral bias and gradient conflicts within the governing equations. To address these issues, we propose Geo

Xinqun Mei, Jin Yan

In this paper, we study the interior $C^{2}$ estimates for Hessian quotient equations $\frac{\sigma_{3}(D^{2}u)}{\sigma_{l}(D^{2}u)}=1$ for $l=1, 2$, in arbitrary dimensions, under the natural ellipticity and semi-convexity conditions. We further derive analogous results for the corresponding sum Hessian equations. In addition, we establish several rigidity

He Hu, Tengjin Weng, Zebang Cheng, Yu Wang

Recent multimodal large language models (MLLMs) have shown strong capabilities in perception, reasoning, and generation, and are increasingly used in applications such as social robots and human-computer interaction, where understanding human emotions is essential. However, existing benchmarks mainly formulate emotion understanding as a static recognition pr

Edan Gabay, Yael Maoz, Jonathan Stahl, Naama Maoz

We present a comprehensive evaluation of Large Language Models (LLMs) on Computer Science (CS) Data Structure examination questions. Our work introduces a new benchmark dataset comprising exam questions from Tel Aviv University (TAU), curated to assess LLMs' abilities in handling closed and multiple-choice questions. We evaluated the performance of OpenAI's

Kay Joerg Wiese, Alapan Das, Adam Nahum

A process that images or measures bond energies in the critical Ising model can be in distinct measurement ``phases'', depending on the precision of measurement. We study the transition into the strong-measurement phase using replica field theory (an epsilon expansion around six dimensions) and numerical simulations in two dimensions. The results reveal mult

Yangyang Zhao, Linfan Dai, Li Cai, Bowen Xing

Cross-domain task-oriented dialogue requires reasoning over implicit and explicit feasibility constraints while planning long-horizon, multi-turn actions. Large language models (LLMs) can infer such constraints but are unreliable over long horizons, while Reinforcement learning (RL) optimizes long-horizon behavior yet cannot recover constraints from raw dial

Sanghoon Lee, Geon Lee, Hyekang Park, Bumsub Ham

Conventional object detectors typically operate under a closed-set assumption, limiting recognition to a predefined set of base classes seen during training. Open-vocabulary object detection (OVD) addresses this limitation by leveraging vision-language models (VLMs) to generate pseudo labels for novel object classes. However, existing OVD methods suffer from

Xiaomeng Zhang, Jinhan Guo, Yang Guo, Mingde Ding

On 2024 May 6, Active Region 13663 produced an X4.5-class flare associated with a filament eruption that exhibited remarkable rotation and deflection dynamics. This study aims to investigate two key aspects of this event: the formation mechanisms of the complex flare ribbon structures and the physical drivers behind the observed filament deflection. We condu

Jingyu Zhang, Fan Wang, Jacky Keung, Yihan Liao

Deep learning (DL)-based systems can exhibit unexpected behavior when exposed to out-of-distribution (OOD) scenarios, posing serious risks in safety-critical domains such as malware detection and autonomous driving. This underscores the importance of thoroughly testing such systems before deployment. To this end, researchers have proposed a wide range of tes

Taha Hammadia, Lucas Rea, Ahmad Mohammad Saber, Amr Youssef

The deployment of Large Language Models (LLMs) as assistants in electric grid operations promises to streamline compliance and decision-making but exposes new vulnerabilities to prompt-based adversarial attacks. This paper evaluates the risk of jailbreaking LLMs, i.e., circumventing safety alignments to produce outputs violating regulatory standards, assumin

Chun Jie Chong, Muyeed Ahmed, Zhihao, Yao

LLM-generated code is widely used, and the share of committed code produced by LLMs is expected to increase. However, we are not at a point where LLMs can be effective contributors to production code. We present an approach that exposes the shortcomings of LLM generation on such projects, and proposes recommendations; the targets of our study are sizable ope

Rufaidah Alharbi, Mohamed Ben Ayed, Khalil El Mehdi

In this paper, we study the Dirichlet elliptic problem $(\mathcal{P}_\varepsilon)$: $-\Delta u +V\,u = u^{p-\varepsilon}$, $u>0$ in $\Omega$, $u=0$ on $\partial\Omega$, where $\Omega\subset \R^n$ ( $n\geq 3$) is a bounded domain, $V$ is a smooth positive function on $\overline{\Omega}$, $p+1= 2n/(n-2)$ is the critical Sobolev exponent, and $\varepsilon >0$ i

Kexin Chu

Agentic AI systems introduce a security surface that is qualitatively different from that of stateless LLMs. They persist memory, invoke external tools, coordinate with peer agents, and operate across sessions, allowing attacks to emerge not only at the prompt interface but also through architectural state, delegated authority, and long-horizon interactions.

Yida Xue, Ningyu Zhang, Tingwei Wu, Zhe Ma

The vast and underexplored ocean plays a critical role in regulating global climate and supporting marine biodiversity, yet artificial intelligence has so far delivered limited impact in this domain due to a fundamental data bottleneck. Specifically, ocean data are highly fragmented across disparate sources and inherently exhibit multi-modal, high-noise, and

Hugo P. Maia, Guilherme Ferraz de Arruda, Silvio C. Ferreira, Yamir Moreno

The organization of higher-order interactions plays a central role in shaping collective dynamics, yet a general structural principle governing contagion on hypergraphs remains lacking. Here we introduce a nesting coefficient that quantifies how lower-order interactions are embedded within higher-order ones, defining a continuum between simplicial complexes

Chandravardhan Singh Raghaw, Anushka Parwal, Shahid Shafi Dar, Prajakta Darade

Knee osteoarthritis (KOA) is a degenerative joint disease that can lead to chronic pain, reduced mobility, and long-term disability. Automated severity grading from knee radiographs can support early assessment, but current methods heavily depend on large labeled datasets and remain sensitive to class imbalance, noisy samples, and variability in clinical ann

Yu Cong, Kangyi Tian

Motivated by the FPTAS for connectivity interdiction of Huang et al. (IPCO'24), we isolate the part of the argument that does not use cuts. The setting is a minimization problem over a feasible-set family $\mathcal F$ with a linear objective $w(S)=\sum_{e\in S}w(e)$. After dualizing the interdiction budget, deletion can be absorbed into truncated weights $w_

Liaoyaqi Wang, Chunsheng Zuo, William Jurayj, Benjamin Van Durme

Scaling test-time computation with reinforcement learning (RL) has emerged as a reliable path to improve large language models (LLM) reasoning ability. Yet, outcome-based reward often incentivizes models to be overconfident, leading to hallucinations, unreliable confidence-based control, and unnecessary compute allocation. We introduce Reinforcement Learning

Muhammad Umair Basharat, Jawad Hussain, Waqas Khalid, Chiew Foong Kwong

The growing adoption of IoT and cloud computing, combined with rapid advancements in digital technologies, has considerably increased the cyber-attack surface, resulting in increasingly complex and persistent attacks. Traditional security methods, primarily based on perimeter defenses, are insufficient to meet these developing threats, especially within the

You Wu, Peter Beerel

Jump-Oriented Programming (JOP) attacks exploit indirect control transfers to bypass backward-edge defenses, yet existing forward-edge CFI mechanisms lack precise source-domain authorization: type-based CFI admits all same-signature callers, while tag-based hardware CFI is limited by fixed-width register storage that caps the number of simultaneously authori

Daniel Bertschinger, Henry Förster, Fabian Klute, Irene Parada

We study the common intersection of arrangements of double-wedges. We consider arrangements where double-wedges may be either bowties (which do not contain a vertical line) or hourglasses (which contain a vertical line), in contrast to earlier studies that focused on arrangements of only bowties. This generalization changes the setting drastically, in partic

Kirill Krylov, Daniil Rystsov, Konstantin Stepanyantz

For supersymmetric extensions of the Standard Model we construct some expressions that include Yukawa couplings for the third and second generations and receive relatively small quantum corrections. This implies that they slightly depend on scale and are therefore approximate renormalization group invariants. Using these invariants we try to analyse possible

Zhonghao Xia, Michele Reticcioli, Yateng Wang, Yali Yang

High-performance thermoelectric (TE) materials are crucial for efficient waste-heat recovery and solid-state cooling technologies. A persistent challenge in TE materials design arises from the strong interdependence among the electrical conductivity ($\sigma$), Seebeck coefficient ($S$), and lattice thermal conductivity ($\kappa_{\mathrm{L}}$). Layered compo

Stefano Bonzio, Andrea Loi, Giuseppe Zecchini

This paper investigates the interplay between algebraic structure, topology, and differentiability in Clifford semigroups. The study is developed along three main themes. First, in the compact Hausdorff setting, we provide an explicit construction of a compatible metric for the Bowman topology. Second, we address Hilbert-fifth-type questions by establishing

Yahui Li, Yinfeng Yu, Liejun Wang, Shengjie Shen

Emotionally talking head video generation aims to generate expressive portrait videos with accurate lip synchronization and emotional facial expressions. Current methods rely on simple emotional labels, leading to insufficient semantic information. While introducing high-level semantics enhances expressiveness, it easily causes lip-sync degradation. Furtherm

Taihua Xu, Genhao Tian, Jicong Fan, Xibei Yang

Graph Neural Networks (GNNs) have demonstrated impressive performance in learning representations from graph-structured data. However, their message-passing mechanism inherently relies on the assumption of label consistency among connected nodes, limiting their applicability to low-homophily settings. Moreover, since message passing operates as a hierarchica

Irene Aldridge

This paper examines the computational complexity of the \emph{Core Identification Problem} (CIP) in one-sided matching markets governed by the Top Trading Cycles (TTC) algorithm. The central contribution is a formal complexity separation: this paper proves that identifying which agents receive a core allocation is strictly easier than computing the full TTC

Meizhu Liu, Matthew Rowe, Amit Agarwal, Michael Avendi

Audio-text retrieval enables semantic alignment between audio content and natural language queries, supporting applications in multimedia search, accessibility, and surveillance. However, current state-of-the-art approaches struggle with long, noisy, and weakly labeled audio due to their reliance on contrastive learning and large-batch training. We propose a

Małgorzata Nowak-Kępczyk

We study the minimal dimension of maximal commutative subalgebras of the matrix algebra $M_n(k)$ over an algebraically closed field. While examples with dimension strictly smaller than n are known for $n \geq 14$, no such examples are known in smaller dimensions. In this paper, we show that for n = 6 every maximal commutative subalgebra $A\subset M_6(k)$ sat

Zhaoxiang Liu, Zhicheng Ma, Kaikai Zhao, Kai Wang

The Convolutional Neural Networks (CNNs) have been the dominant and effective approach for general computer vision tasks. Recently, Kolmogorov-Arnold neural networks (KANs), based on the Kolmogorov-Arnold representation theorem, have shown potential to replace Multi-Layer Perceptrons (MLPs) in deep learning. KANs, which use learnable nonlinear activations on

Yue Zhang, Kun Xu

The viscous shock tube is a canonical test case for assessing Navier-Stokes (NS) solvers in the continuum-flow regime, widely used to validate numerical accuracy and probe flow physics. It features a rich set of interacting structures-shock and rarefaction waves, contact discontinuities, boundary layers, and their coupling-spanning multiple spatial and tempo

Xinzhu Chen, Wei He, Huichuan Fan, Wenzhe Niu

Group Relative Policy Optimization (GRPO) performs coarse-grained credit assignment in reinforcement learning with verifiable rewards (RLVR) by assigning the same advantage to all tokens in a rollout. Process reward models can provide finer-grained supervision, but they require step-level annotation or additional reward modeling. We show that hidden-state di

Max Tschaikowski, Andrea Vandin

Quantum optimization algorithms promise advantages for difficult problems but are costly to simulate and analyze on classical machines. Recently, constrained quantum optimization has been investigated through the lens of Quantum Zeno dynamics, an approach which constrains the search to a subspace by means of quantum measurements. Exploiting that quantum meas

Marco Robbio, Michael G. Jabbour, Nicolas J. Cerf

It is well known that bosons and fermions exhibit opposite behaviors when experiencing interference, in the sense that bosons have a tendency to bunch whereas fermions have a tendency to antibunch. Recently, this complementarity was mathematically characterized in [arXiv:2312.17709] by means of an identity relating the transition probabilities of both types

Liang Chen

We test a regime-conditional functional-form restriction on aggregate risk-exposure dynamics implied by VaR-constrained intermediary models: exposures contract multiplicatively when capital constraints bind and grow additively (level-independent) when slack. The contraction half follows from binding VaR constraints (Brunnermeier and Pedersen 2009; Adrian and

Jingxuan Kang, Ziqi Zhang, Shaoming Zheng, Shuang Li

Segmentation is central to clinical diagnosis and monitoring, yet the reliability of modern foundation models in medical imaging still depends on the availability of precise prompts. The Segment Anything Model (SAM) offers powerful zero-shot capabilities, although it collapses under the weak, generic, and noisy prompts that dominate real clinical workflows.

Tian Chen, Minyi Huang

This paper investigates a class of linear-quadratic-Gaussian risk-sensitive graphon mean-field games, involving an asymptotically infinite population of heterogeneous agents distributed across an asymptotically infinite network, where each agent aims to minimize an exponential cost functional reflecting its risk sensitivity. Following the Nash certainty equi

Rory Young, Nicolas Pugeault

Deep reinforcement learning policies achieve strong performance in complex continuous control environments with nonlinear contact forces. However, these policies often produce chaotic state dynamics, with trivially small changes to the initial conditions significantly impacting the long-term behaviour of the control system. This high sensitivity to initial c

Yanbo Li, Jiansheng Zhang, Shasha Zhu

In the light of a series of papers on moving vectors, we define and study core abaci of classical affine types for arbitrary charge. This greatly extends the concept of cores with charge zero, and make us being able to parameterize the affine Grassmannian $W^j$ by core abaci of charge $j$ for arbitrary classical affine types. By associating a core abacus $(\

Yuru Zhang, Ming Zhao, Qiang Liu, Ahmed Alkhateeb

Precisely modeling radio propagation in dynamic wireless environments is fundamental to the realization of wireless digital twins. Traditional ray tracing methods rely on accurate 3D models with detailed environment parameters, while recent neural radiance field approaches learn representations tied to specific static scenes, requiring retraining when enviro

Yanpei Gong, Beichen Zhang, Hao Wang, Xuhang Fu

Remote sensing image change captioning (RSICC) aims to describe the difference between two remote sensing images. While recent methods have explored video modeling, they largely overlook the inherent ambiguities in viewpoint, scale, and prior knowledge, lacking effective constraints on the encoder. In this paper, we present STAND, a Semantic Anchoring Constr

Marcel Hedman, Kale-ab Abebe Tessera, Juan Claude Formanek, Anya Sims

Offline multi-agent reinforcement learning (MARL) enables policy learning from fixed datasets, but is prone to coordination failure: agents trained on static, off-policy data converge to suboptimal joint behaviours because they cannot co-adapt as their policies change. We introduce CODA (Coordination via On-Policy Diffusion for Multi-Agent Reinforcement Lear

Thibaud Southiratn, Bonil Koo, Yijingxiu Lu, Sun Kim

Dual-target molecule generation, which focuses on discovering compounds capable of interacting with two target proteins, has garnered significant attention due to its potential for improving therapeutic efficiency, safety and resistance mitigation. Existing approaches face two critical challenges. First, by simplifying the complex dual-target optimization pr

Nicholas Hale, Charis Harley, Prince Nchupang, Jan Nordström

Gauss-Lobatto quadrature nodes and weights are optimal for closed summation-by-parts (SBP) formulations based on polynomial approximation spaces in the sense that for a prescribed function space they yield an SBP operator of minimal dimension. We show that the same principle extends to general (possibly non-polynomial) function spaces: an associated generali

A. Moerman, N. Soshnin, S. A. Brackenhoff, S. O. Dabironezare

Submillimeter (submm) integral field units (IFUs) utilising kinetic inductance detectors (KIDs) are a promising instrument architecture for the study of galaxies, galaxy clusters, and the large-scale structure of the Universe. In order to design successful experiments targeting these science cases, several aspects such as instrument design, observation and c

Shuofeng Huang, Chengjie Yu

In this paper, we obtain a weighted trigonometric summation formula which is an extension of the trigonometric summation formula by Grigor'yan, Lin and Yau \cite{GLY}.

Duaa Alim, Mogtaba Alim, Liam Chalcroft

Vision-language models (VLMs) read an image and produce text in a single forward pass, whereas radiologists typically inspect an image several times and consult the literature before writing a report. We introduce GAZE (Grounded Agentic Zero-shot Evaluation), a framework that lets a medical VLM work in this iterative way by calling viewer-level tools (zoom,

Chu Guo, Wei Wu, Xiansong Xu, Ping-Xing Chen

The Grassmann time-evolving matrix product operator (GTEMPO) method, which represents the Feynman-Vernon influence functional as a temporal matrix product state, has been shown to be a flexible and potentially scalable solution for fermionic quantum impurity problems. In this work, we extend GTEMPO to solve fermionic impurity problems in the Nambu formalism,

Tyler Buchanan, Ali Eidi, Richard P. Dwight

Data-driven correction of turbulence models offers a promising route for improving Reynolds-averaged Navier-Stokes (RANS) predictions, but quantifying uncertainty in such corrections and ensuring generalization across flows remain key challenges. This work presents a Bayesian neural network (BNN) framework for uncertainty-aware correction of RANS models. Two

Can Liu, Sizhe Cheng, Feng Liang, Zhibang Jiang

With the rise of mobile-first consumption, users increasingly engage with data visualizations on mobile devices. However, the vast majority of existing visualizations are originally authored for desktop environments. Due to significant differences in viewport size and interaction paradigms, directly scaling desktop charts often results in illegible text, inf

Hai-Peng Xie, Si-Yi Chen, Ning Li, Wei Chen

We investigate the three-body $DDK$ system with quantum numbers $I(J^P) = \frac{1}{2}(0^-)$ within a coupled-channel framework that incorporates both $DDK$ and $D^{*}D^{*}K$ configurations. The $D^{(*)}D^{(*)}$ interactions are described using the one-boson-exchange model constrained by the heavy-quark symmetry and fitted to the pole positions of $X(3872)$,

Lucas V. Pupim, Mathias S. Scheurer

One of the central goals in the emergent field of altermagnetism is the unambiguous experimental identification and characterization of altermagnetic order across a variety of compounds. This motivates exploring tools that can clearly distinguish altermagnets from antiferromagnets, based on symmetry signatures, and offer access to the dominant orbital charac

Bingfeng Chen, Chenjie Qiu, Yifeng Xie, Boyan Xu

Aspect Sentiment Quad Prediction (ASQP) has seen significant advancements, largely driven by the powerful semantic understanding and generative capabilities of large language models (LLMs). However, while syntactic structure information has been proven effective in previous extractive paradigms, it remains underutilized in the generative paradigm of LLMs due

Adam Herling, Ofer Neufeld

High harmonic generation (HHG) is a widely explored process in solids, where intense lasers drive attosecond-to-femtosecond electron dynamics within bands, causing high-energy emission. While electrons and photons are considered the main players in HHG, solids also host ubiquitous phonons that are typically assumed negligible in HHG due to their longer times

Călin Spiridon

In this note, we study the resonance variety of rank-two vector bundles over an elliptic curve. Our approach is based on analyzing the flattening stratification of the resonance. We also investigate the linear section of the Grassmann variety $\operatorname{Gr}(2,n)$ from which the resonance is constructed through the lens of its corresponding flattening str

Varun Totakura, Ankita Singh, Yushun Dong, Shayok Chakraborty

Active learning algorithms automatically identify the most informative samples from large amounts of unlabeled data and tremendously reduce human annotation effort in inducing a machine learning model. In a conventional active learning setup, the labeling oracles are assumed to be infallible, that is, they always provide correct answers (in terms of class la

Varun Totakura, Shayok Chakraborty

Due to the unprecedented success of deep learning, it has become an integral component in several multimedia computing applications in todays world. Unfortunately, deep learning systems are not perfect and can fail, sometimes abruptly, without prior warning or explanation. While reducing the error rate of deep neural networks has been the primary focus of th

Kostis Trantzas, Besiana Agko, Christos Tranoris, Irene Denazi

6G network complexity necessitates high levels of autonomy, yet current intent-based systems struggle with ambiguous or incomplete human requests. This paper introduces an agent-based, intent-driven end-to-end (E2E) orchestration framework designed for Network-as-a-Service (NaaS) delivery through collaborative intent co-creation. The proposed system leverage

Yiming Chen, Sean Colin-Ellerin, Ohad Mamroud, Kyriakos Papadodimas

We expect black hole microstates to differ in their chaotic properties from states associated with other geometries. For supersymmetric black holes, ordinary level statistics cannot diagnose this distinction, since their energy levels are exactly degenerate. We propose that there is an intrinsic probe of chaos, encoded in the mixing of the microstates under

John M. Campbell

One of the central open problems in both algebraic combinatorics and representation theory is to find a positive combinatorial rule for Kronecker coefficients $ g_{\lambda \, \mu \, \nu}$. A notable advance in this direction is due to Blasiak, who proved a combinatorial interpretation in terms of colored Yamanouchi tableaux for the case whereby one of the in

Christos Tranoris, Besiana Agko, Kostis Trantzas, Irene Denazi

This work proposes an agentic, intent-driven end-to-end (E2E) orchestration framework that integrates intent co-creation with a Test-Driven Quality Assurance paradigm. In this framework, autonomous agents iteratively refine a user's initial intent into a confirmed, auditable specification. Furthermore, the system automatically derives validation tests from t

Meizhu Liu, Nistha Mitra, Paul Li, Amine Abdaoui

In this work, we present Au-M-ol, a novel multimodal architecture that extends Large Language Models (LLMs) with audio processing. It is designed to improve performance on clinically relevant tasks such as Automatic Speech Recognition (ASR). Au-M-ol has three main components: (1) an audio encoder that extracts rich acoustic features from medical speech, (2)

Zhiyuan Zhai, Ming Li, Xin Wang

Current LLM agents operate under an implicit but universal assumption: execution is a transaction -- the user submits a request, the agent works in isolation, and only upon completion does the dialogue resume. This forces users into a binary choice: wait for a potentially incorrect output, or interrupt and lose all progress. We reject this assumption and pro

Long Jing, Zhixiong Yang, Yajun Zhang, Xinlong Feng

Human activity recognition serves as the foundation for various emerging applications. In recent years, researchers have used collaborative sensing of multi-source sensors to capture complex and dynamic human activities. However, multimodal human activity sensing typically encounters highly heterogeneous data across modalities and label scarcity, resulting i

Takumi Otsuka, Kentaroh Toyoda, Alex Leung

AI agents are now running real transactions, workflows, and sub-agent chains across organizational boundaries without continuous human supervision. This creates a problem no current infrastructure is equipped to solve: how do you identify, verify, and hold accountable an entity with no body, no persistent memory, and no legal standing? We define AI Identity

Philip Wilson, Axel Constant, Mahault Albarracin, Nicolás Hinrichs

The proliferation of agentic artificial intelligence has outpaced the conceptual tools needed to characterize agency in computational systems. Prevailing definitions mainly rely on autonomy and goal-directedness. Here, we argue for a minimal notion open to principled inspection given three criteria: intentionality as action grounded in beliefs and desires, r

Yitian Zhou, Chaoning Zhang, Jiaquan Zhang, Zhenzhen Huang

Long-context large language models remain computationally expensive to run and often fail to reliably process very long inputs, which makes context compression an important component of many systems. Existing compression approaches typically rely on trained compressors, dense retrieval-style selection, or heuristic trimming, and they often struggle to jointl

Meizhu Liu, Yassi Abbasi, Matthew Rowe, Michael Avendi

PDF documents contain critical visual elements such as figures, tables, and forms whose accurate extraction is essential for document understanding and multimodal retrieval-augmented generation (RAG). Existing PDF parsers often miss complex visuals, extract non-informative artifacts (e.g., watermarks, logos), produce fragmented elements, and fail to reliably

Bo Yang, Jianke Yang

We report rogue-wave and lump patterns associated with Umemura polynomials, which arise in rational solutions of the third Painlevé equation. We first show that in many integrable equations such as the nonlinear Schrödinger equation and the Boussinesq equation, when internal parameters of their rogue wave solutions are large and of certain form, then their r

Kaiwen Huang, Yi Zhou, Yizhe Zhang, Jingxiong Li

Semi-supervised learning addresses label scarcity and high annotation costs in medical image segmentation by exploiting the latent information in unlabeled data to enhance model performance. Traditional discriminative segmentation relies on segmentation masks, neglecting feature-level distribution constraints. This limits robust semantic representation learn

Leonardo Pacheco

We study a variant of the modal $\mu$-calculus based on the constructive modal logic $\mathsf{CK}$. We define game semantics for the constructive $\mu$-calculus and prove its equivalence to the birelational Kripke semantics. We then use the game semantics to prove the soundness and completeness of a fully-labeled non-wellfounded proof system for it. At last,

Jimin Lee, Huiwon Jang, Myungkyu Koo, Jungwoo Park

Humans understand and interact with the real world by relying on diverse physical feedback beyond visual perception. Motivated by this, recent approaches attempt to incorporate physical sensory signals into Vision-Language-Action models (VLAs). However, they typically focus on a single type of physical signal, failing to capture the heterogeneous and complem

Ruyi Dai, Tingkwong Ng, Hao Chen

Automated white blood cell (WBC) classification is essential for scalable leukaemia screening. However, real-world deployment is challenged by domain shifts caused by staining protocols, scanner characteristics, and inter-laboratory variability, which often degrade model performance. The White Blood Cell Classification Challenge (WBCBench) at ISBI 2026 aims

M. W. AlMasri

We propose a mathematically rigorous unified framework for hybrid quantum mechanics that systematically combines algebraic deformation and spatial non-locality within a single operator formalism. By constructing a self-adjoint hybrid kinetic operator through spectral calculus, we derive exact generalized uncertainty relations that interpolate between $q$-def

Cristian López, Mckenna Partridge, Sebastian De Pascuale, Jeremy Lore

The control of complex dynamical systems remains a fundamental challenge in science and engineering, where strong nonlinearities, the presence of noise, and computational constraints often pose significant obstacles in traditional control approaches. Recent advances in data-driven methods, particularly system identification techniques, have shown a powerful

Sangwook Baek, Vin Van Duong, Karam Park, Pilkyu Park

This paper introduces a novel multi frame super-resolution network (MFSR) for burst hexadeca Bayer pattern Contact Image Sensor (CIS) images, which includes demosaicing, denoising, multi-frame fusion, and super-resolution. Designing a high-quality reconstruction network poses several challenges as follows: 1) Unlike the Bayer color filter array (CFA) pattern

Qinchen Song, Xinliang Liu, Lei Zhang

Conventional numerical solvers for the radiative transfer equation (RTE) exhibit severe sensitivity to medium parameters. To address this, we propose an operator learning framework that approximates the RTE solution map as a function of material properties. The core architecture, MgNet, preserves the solution operator framework established by recursive skele

Heng Li, Xiaotong Lin, Ling-An Zeng, Yulei Kang

Text-to-motion generation aims to generate 3D human motions that are tightly aligned with the input text while remaining physically plausible and rich in fine-grained detail. Although recent approaches can produce complex and natural movements, they usually operate at only one temporal scale, which limits both semantic alignment and temporal coherence. Inspi

Zhenzhen Huang, Chaoning Zhang, Fachrina Dewi Puspitasari, Jiaquan Zhang

Large language models (LLMs) are increasingly utilized in various complex reasoning tasks due to their excellent instruction following capability. However, the model's performance is highly dependent on the open-ended characteristics of the users' input prompt. Natural prompts often do not follow proper syntactic rules, which creates ambiguous queries that y

Namya Malik, Ashish Patwari, Sangeetha N

Two-fold redundant sparse arrays possess inbuilt redundancy to tackle single-element failures. This property enables them to perform accurate direction of arrival (DOA) estimation even during single sensor faults. However, recent literature suggests that some TFRSAs suffer from hidden dependencies whereby a single sensor fault at peculiar positions within th

Shotaro Murayama, Yasufumi Nitta

In this paper, we study the existence of Mabuchi solitons on admissible manifolds as defined by Apostolov--Calderbank--Gauduchon--T\o nnesen-Friedman. We prove that a Fano admissible manifold admits a Mabuchi soliton if and only if the Mabuchi constant is less than 1. We also provide an explicit formula for the Mabuchi constant on Fano admissible manifolds,

Anthony Lee

An approach to construct explicit integral representations for two-layer ReLU networks is presented, which provides relatively simple representations for any multivariate polynomial. Quantitative bounds are provided for a particular, sharpened ReLU integral representation, which involves a harmonic extension and a projection. The bounds demonstrate that func

Xiaodong Tian, Yan Lin, Hanbin Gao, Han Yu

The integration of epitaxial barium titanate (BTO) on silicon represents a highly promising pathway for next-generation, energy-efficient photonic integrated circuits due to BTO's exceptionally high Pockels coefficients. However, the scalable epitaxy of BTO on Si remains hindered by complex stoichiometric control and slow growth rates. In this work, we demon

Amin Ahadi, Saman Sarshar

In this paper, numerical simulations of four-mode continuous-variable cluster states with different topologies in the framework of measurement-based quantum computation are presented. By utilizing the symplectic representation and covariance matrix, the process of generating cluster states with linear, square, and T-shaped topologies has been systematically

Mark Chua, Samuel Ajila

Software engineering (SE) organizations operate in a knowledge-intensive domain where critical assets -- architectural expertise, design rationale, and system intuition -- are overwhelmingly tacit and volatile. The departure of key contributors or the decay of undocumented decisions can severely impair project velocity and software quality. While conventiona

Kiyoshige Garces, Gloria Milena Fernandez-Nieto, Linxuan Zhao, Sachini Samaraweera

Research shows that dialogue, the interactive process through which participants articulate their thinking, plays a central role in constructing shared understanding, coordinating action, and shaping learning outcomes in teams. Analysing dialogue content has been central to advancing team learning theory and informing the design of computer-supported collabo

Giampiero Lovat

We investigate the transient electromagnetic field radiated by a pulsed vertical electric dipole above a lossy half-space and identify its time-domain signatures associated with the Zenneck wave. Starting from the classical Sommerfeld representation, we derive a causal time-domain formulation based on the double-deformation technique, with successive contour

Oleg Kiselev

We construct a local matched-asymptotic description of time-harmonic elastic fields generated by Rayleigh waves near cuspidal elements of a traction-free surface. The free surface is represented locally by a cusp graph with exponent $0<\alpha<1$, or equivalently by a vanishing-width horn $b(s)=B s^m$, $m=1/\alpha>1$. A cuspidal gorge is a zero-opening re-ent

Zhisheng Xiong, Dimitris Boskos, Bo Zeng, Peter Palensky

This paper studies the robust optimal operation of distribution networks (DNs) under renewable generation and load demand uncertainties, seeking an improved trade-off between robustness and economic performance. Building upon information gap decision theory (IGDT), a generalized uncertainty modelling is proposed to enhance the expressiveness of the uncertain

Eduardo Oliveira, Michael Fu, Patanamon Thongtanunam, Sonsoles López-Pernas

Code review is central to software engineering education but hard to scale in capstone projects due to tight deadlines, uneven peer feedback, and limited prior experience. We investigate an LLM-as-reviewer integrated directly into GitHub pull requests (human-in-the-loop) across two cohorts (more than 100 students, 2023--2024). Using a mixed-methods design --

Yifan Han, Jianxiang Liu, Haoyu Zhang, Yuqi Gu

Learning robot manipulation from human videos is appealing due to the scale and diversity of human demonstrations, but transferring such demonstrations to executable robot behavior remains challenging. Prior work either relies on robot data for downstream adaptation or learns affordance representations that remain at the perception level and do not directly

Tran Thanh Lam Nguyen, Edoardo Di Tullio, Barbara Carminati, Elena Ferrari

Mobile apps offer significant benefits, but their privacy protections often remain ineffective and confusing for users. While prior work mainly analyzes app privacy vulnerabilities, few approaches help users understand, set, and enforce their privacy preferences. This paper presents PrivacyAssist, a multi-agent LLM-based platform that detects inconsistencies

Masoumeh Chapariniya, Jean-Marc Odobez, Volker Dellwo, Teodora Vuković

Avatar fingerprinting, i.e., verifying who drives a synthetic talking-head video rather than whether it is real, is a critical safeguard for authorized use of face-reenactment technology. Existing methods rely on a fixed, non-differentiable landmark extraction stage that prevents the fingerprinting model from being optimized end-to-end from raw pixels. We pr

Kaidi Wang, Daniel K. C. So, Zhiguo Ding, George K. Karagiannidis

As a practical physical implementation of pinching-antenna systems, leaky coaxial cable (LCX) enables distributed radiation in more general wireless environments, particularly for lower-frequency applications. In this paper, a leaky-coaxial pinching-antenna system, referred to as the LCX pinching-antenna system, is investigated, and adjustable slot apertures

Jonathan Steinberg, Oren Gal

Safety-oriented instruction-following is supposed to keep LLM-controlled robots safe. We show it also creates an availability attack surface. By injecting short safety-plausible phrases (1-5 tokens) into a robots audio channel, an adversary can trigger the models safety reasoning to halt or disrupt execution without jailbreaking the model or overriding its p

Alexandre Marques, Beatriz Sá, Rui Botelho, Pedro Pinto

The use of Machine Learning (ML) for data-driven decision-making often relies on access to sensitive datasets, which introduces privacy challenges. Traditional encryption methods protect data at rest or in transit but fail to secure it during processing, exposing it to unauthorized access. Homomorphic encryption emerges as a transformative solution, enabling

A. V. Belashov, A. A. Zhikhoreva, Y. M. Beltukov, I. V. Semenova

We present the comparative experimental analysis of frequency dependencies of linear (Lam\'e) and nonlinear (Murnaghan) elastic moduli of polystyrene, PMMA and polycarbonate. The measurement methodology, based on the acousto-elastic effect, provided data on variations of these moduli in block samples of the polymers in the frequency range of 0.45-3 MHz. In a

Akshara Bhat, Abhishek Deshpande, Chittaranjan Hens, Subrata Ghosh

Compartmental epidemic models, grounded in mass-action kinetics, often assume homogeneous mixing. Although this neglects network structure, recent results show that for Poisson random graphs, the classical SIR model, especially the susceptible decay curve, matches the susceptible decay dynamics of its network counterpart. Motivated by this, we investigate wh

Ali Al-Bayaty

This paper introduces a conceptual framework of technology-dependent ternary quantum gates that could be implemented and fabricated into future superconducting and photonic quantum systems for operating 3-valued quantum bits (qutrits). The "technology-dependent" means that such ternary quantum gates are on-purpose designed analogy to the contemporary binary