December 2025 arXiv papers — page 59

Kaican Li, Lewei Yao, Jiannan Wu, Tiezheng Yu

The ability for AI agents to "think with images" requires a sophisticated blend of reasoning and perception. However, current open multimodal agents still largely fall short on the reasoning aspect crucial for real-world tasks like analyzing documents with dense charts/diagrams and navigating maps. To address this gap, we introduce O3-Bench, a new benchmark

Justine Fasquel, Shigenori Nakatsuka

In this paper, we establish a general criterion for good pairs, namely pairs consisting of a nilpotent orbit and an even good grading in a simple Lie algebra, which guarantees the existence of a quantum Hamiltonian reduction between associated affine W-algebras. In particular, we show that for type A, any two affine W-algebras associated with two adjacent ni

Dian K. Palagachev, Lubomira G. Softova

A non-homogeneous conormal derivative problem is considered for quasilinear divergence form elliptic equations modeled on the $m$-Laplacian operator. The nonlinear terms are given by Carath\'eodory functions and satisfy controlled growth structure conditions with respect to the solution and its gradient, while their $x$-behaviour is controlled in terms of su

Tianrui Zhu, Shiyi Zhang, Zhirui Sun, Jingqi Tian

Frame-level autoregressive (frame-AR) models have achieved significant progress, enabling real-time video generation comparable to bidirectional diffusion models and serving as a foundation for interactive world models and game engines. However, current approaches in long video generation typically rely on window attention, which naively discards historical

Tim Stoetzel, Stefan Floerchinger

Renormalization group flow equations of the fluid dynamical shear viscosity transport coefficient of a relativistic real scalar field are derived. The flowing effective action contains branch cut contributions to the self energy and interaction vertex in the symmetric phase. We demonstrate how the flow equation method can systematically extend the perturbati

Y. N. Lima, C. Jahnke, M. Munhoz, F. S. Navarra

With the growth of statistics in the future experiments at the LHC, the number of events with charm production will increase substantially. It may become possible to measure the multiplicity distribution of charm particles. Using PYTHIA-8, we generated charm multiplicity distributions in $pp$ collisions for different pseudorapidity ranges ($|\eta| < 0.5, 1.0

James E. Gallagher, Edward J. Oughton

Landmines remain a persistent humanitarian threat, with an estimated 110 million mines deployed across 60 countries, claiming approximately 26,000 casualties annually. Current detection methods are hazardous, inefficient, and prohibitively expensive. We present the Adaptive Multispectral Landmine Identification Dataset (AMLID), the first open-source dataset

Zichuan Lin, Xiaokai Huang, Jiate Liu, Yuxuan Han

The estimation of individual treatment effects (ITE) focuses on predicting the outcome changes that result from a change in treatment. A fundamental challenge in observational data is that while we need to infer outcome differences under alternative treatments, we can only observe each individual's outcome under a single treatment. Existing approaches addres

Daniel Pfrommer, Zehao Dou, Christopher Scarvelis, Max Simchowitz

We study the inductive biases of diffusion models with a conditioning-variable, which have seen widespread application as both text-conditioned generative image models and observation-conditioned continuous control policies. We observe that when these models are queried conditionally, their generations consistently deviate from the idealized "denoising" proc

Jinqiu Chen, Huyan Xu

Predicting breast cancer recurrence risk is a critical clinical challenge. This study investigates the potential of computational pathology to stratify patients using deep learning on routine Hematoxylin and Eosin (H&E) stained whole-slide images (WSIs). We developed and compared three Multiple Instance Learning (MIL) frameworks -- CLAM-SB, ABMIL, and ConvNe

Junjun Pan, Yixin Liu, Rui Miao, Kaize Ding

Large language model (LLM)-based multi-agent systems (MAS) have shown strong capabilities in solving complex tasks. As MAS become increasingly autonomous in various safety-critical tasks, detecting malicious agents has become a critical security concern. Although existing graph anomaly detection (GAD)-based defenses can identify anomalous agents, they mainly

Chainarong Amornbunchornvej

Concept learning becomes possible only when existing representations fail to account for experience. Most models of learning and inference, however, presuppose a fixed representational basis within which belief updating occurs. In this paper, I address a prior question: under what structural conditions can the representational basis itself expand in a princi

Anatoly Golberg, Vladimir Gutlyanskiĭ, Vladimir Ryazanov, Toshiyuki Sugawa

We explore the phenomenon of cavitation in higher-dimensional elasticity, defining it as the mapping of a punctured ball onto a non-degenerate ring domain. Crucially, for the class of locally quasiconformal mappings (or more general mappings) defined on the punctured ball $0<|x|<1$ in $\mathbb R^n$ that we examine, cavitation is equivalent to a failure of co

Zhiquan Tan, Yinrong Hong

Large language models (LLMs) trained via KL-regularized reinforcement learning demonstrate strong instruction following, self-correction, and reasoning abilities. Yet their theoretical underpinnings remain limited. We exploit the closed-form energy-based model (EBM) structure of the optimal KL-regularized policy to provide a unified variational analysis of L

Robert C. Viesca, Dmitry I. Garagash

Models of faults incorporating slip rate- and state-dependent friction have reproduced phenomena from spontaneous slow, aseismic slip to earthquake-generating dynamic rupture. Numerical explorations of model parameter space regularly show sudden transitions in behavior. However these boundaries are poorly constrained analytically, with commonly used scalings

Bhagyarathi Sahoo, Captain R. Singh, Raghunath Sahoo

The observation of spin alignment of quarkonia in ultra-relativistic heavy-ion collisions provides deep insight into the possible formation of the quark-gluon plasma (QGP). The present study investigates the spin alignment of quarkonia induced by dissociation mechanisms arising from medium effects imposed on quarkonia. We implement an effective Hamiltonian w

Chia-Jui Chou, Hans B. Lao, Yi Yang

We present a graphical method for proving holographic entanglement entropy inequalities (HEIs) in general multipartite systems. By introducing a geometric representation of the entanglement structure, we develop a systematic approach that enables one to visualize and verify the validity of HEIs for any number of subsystems $n$. Several theorems are establish

Habib Ullah, Kun Li, Haoyu Lu, Youjin Deng

The classical $J_1$-$J_2$-$J_3$ Ising model on the honeycomb lattice is important for understanding frustrated magnetic phenomena in materials such as FePS$_3$ and Ba$_2$CoTeO$_6$, where diverse phases (e.g., striped, zigzag, armchair) and magnetization plateaus have been experimentally observed. To explain the experimental results, previous mean-field studi

Bibek Poudel, Weizi Li

Designing efficient transit route networks is an NP-hard problem with exponentially large solution spaces that traditionally relies on manual planning processes. We present an end-to-end reinforcement learning (RL) framework based on graph attention networks for sequential transit network construction. To address the long-horizon credit assignment challenge,

Valentin Lallemant

Sandpiles form one of the largest class of models displaying a critical stationary state. Despite a few decades of research, a comprehensive and systematic rigorous characterisation of their spatial and, even more, time dependent properties has remained elusive. Among the obstacles, we can mention their out of equilibrium and non-linear dynamics features whi

Han Yu, Hao Zou, Xingxuan Zhang, Zhengyi Wang

Although neural networks achieve promising performance in many tasks, they may still fail when encountering some examples and bring about risks to applications. To discover risky samples, previous literature attempts to search for patterns of risky samples within existing datasets or inject perturbation into them. Yet in this way the diversity of risky sampl

Yunji Ha

The magnetic charges of monopoles arising in ultraviolet completions of the Standard Model are constrained by the global structure of the gauge group. After electroweak symmetry breaking, a subset of the ultraviolet monopoles carrying magnetic charges 3$g_D$ and $6g_D$ can survive as isolated, colour-neutral states in the infrared. We show that this selectio

Feng Yu, MD Saifur Rahman Mazumder, Ying Su, Oscar Contreras Velasco

Effective feature selection is essential for high-dimensional data analysis and machine learning. Unsupervised feature selection (UFS) aims to simultaneously cluster data and identify the most discriminative features. Most existing UFS methods linearly project features into a pseudo-label space for clustering, but they suffer from two critical limitations: (

Linwei Qiu, Gongzhe Li, Xiaozhe Zhang, Qilin Sun

Image correction and rectangling are valuable tasks in practical photography systems such as smartphones. Recent remarkable advancements in deep learning have undeniably brought about substantial performance improvements in these fields. Nevertheless, existing methods mainly rely on task-specific architectures. This significantly restricts their generalizati

M. S. Kurilov, P. M. Ostrovsky

Reflection of particles from a disordered or chaotic medium is characterized by a scattering matrix that can be represented as a superposition of resonances. Each resonance corresponds to an eigenstate inside the medium and has a width related to the decay time of this eigenstate. We develop a general approach to study the distribution function of these reso

Chenyu Wang, Wanjian Yin, Ke Zhou

Water's ability to self-dissociate into H$_3$O$^+$ and OH$^-$ ions is central to acid-base chemistry and bioenergetics. Recent experimental advances have enabled the confinement of water down to the nanometre scale, even to the single-molecule limit, yet how this process is altered at the extreme nanoconfinement remains unclear. Using \emph{ab-initio} calcul

Kaiyi Chi, Yinghui He, Qianqian Yang, Yuanchao Shu

Deep joint source-channel coding (DeepJSCC) has emerged as a promising paradigm for efficient and robust information transmission. However, its intrinsic characteristics also pose new security challenges, notably an increased vulnerability to eavesdropping attacks. Existing studies on defending against eavesdropping attacks in DeepJSCC, while demonstrating c

Adam Hahn, Rubin Krief, Daniel Rebori-Carretero, Rami Puzis

The increasing cyber threats to critical infrastructure highlight the importance of private companies and government agencies in detecting and sharing information about threat activities. Although the need for improved threat information sharing is widely recognized, various technical and organizational challenges persist, hindering effective collaboration.

Adrien Fradin, Abdurakhmon Sadiev, Laurent Condat, Peter Richtárik

We study stochastic nonconvex optimization under heavy-tailed noise. In this setting, the stochastic gradients only have bounded $p$-th central moment ($p$-BCM) for some $p \in (1,2]$. Building on the foundational work of Arjevani et al. (2022) in stochastic optimization, we establish tight sample complexity lower bounds for all first-order methods under \em

Maozeng Zhang, Ke Shi, Huijun Li, Tongshu Chen

Stroke-induced motor impairment often results in substantial loss of upper-limb function, creating a strong demand for rehabilitation robots that enable safe and transparent physical human-robot interaction (pHRI). Variable stiffness actuators are well suited for such applications. However, in most existing designs, stiffness is coupled with the deflection a

Yue Zhang, Yaru Fu, Pei Liu, Yalin Liu

Pinching antenna systems (PASS) have emerged as a technology that enables the large-scale movement of antenna elements, offering significant potential for performance gains in next-generation wireless networks. This paper investigates the problem of maximizing the average per-user data rate by optimizing the antenna placement of a multi-waveguide PASS, subje

Rajnish Kumar, Suriya Prakash Muthukrishnan, Lalan Kumar, Sitikantha Roy

Accurate EMG-driven musculoskeletal (MSK) modeling is critical for biomechanics, rehabilitation, and assistive technology. However, most models calibrate parameters under a single load, ignoring the fact that tasks with similar kinematics may differ in mechanical demand. This study introduces a load-aware calibration framework to improve joint torque predict

Zhifei Li, Lifan Chen, Jiali Yi, Xiaoju Hou

Knowledge Tracing (KT) aims to dynamically model a student's mastery of knowledge concepts based on their historical learning interactions. Most current methods rely on single-point estimates, which cannot distinguish true ability from outburst or carelessness, creating ambiguity in judging mastery. To address this issue, we propose a Knowledge Mastery-State

Hai-Feng Wang, Xiao Han, Giovanni Carraro, Martin Lopez-Corredoira

We analyze the structure and chemo-dynamical properties of the Galactic bulge using ab-type RR Lyrae stars (RRabs) from OGLE-IV and giant stars from APOGEE and Gaia. Orbital integration of 1,879 RRab variables reveals three populations: central bulge, inner bulge, and halo/disk contaminants. Inner bulge RRabs display bar-like kinematics, whereas central bulg

Gopalkrishna Prabhu, Uddeepta Deka, Sumanta Chakraborty, Shasvath J. Kapadia

Propagating gravitational waves (GWs) can encounter a massive object (lens) whose gravitational radius is comparable to the wavelength of the GWs (wave-optics regime). The resulting `microlensed' signal contains imprints about the properties of the lens. In this work, we compute the GW waveforms microlensed by a rotating compact object in weak-field gravity.

Zhanxun Liu, Yifan Duan, Mengmeng Wang, Pengchao Feng

We present X-Talk, an open-source framework that champions a decoupled, modular design for LLM-driven speech-to-speech (S2S) systems. While the dominant trend favors end-to-end (E2E) modeling to optimize information flow, these "omni-models" often struggle to balance the competing objectives of complex speech tasks within a single network. X-Talk challenges

Sara van de Geer, Sylvain Sardy, Maximę van Cutsem

We study the high-dimensional linear model with noise distribution known up to a scale parameter. With an $\ell_1$-penalty on the regression coefficients, we show that a transformation of the log-likelihood allows for a choice of the tuning parameter not depending on the scale parameter. This transformation is a generalization of the square root Lasso for qu

Mariagrazia Bianchi, Nicola Sambonet

Following Gluck and Wolf we complete the It\^o--Michler's Theorem for the projective representations of a $p$-solvable or $\pi$-separable group, and then we relate the projective irreducible modules of such a group with those of its Sylow $p$-subroups and Hall $\pi$-subgroups.

Cailin Lei, Haiyang Wu, Yuxiong Ji, Xiaoyu Cai

Enhancing the performance of trajectory planners for lane - changing vehicles is one of the key challenges in autonomous driving within human - machine mixed traffic. Most existing studies have not incorporated human drivers' prior knowledge when designing trajectory planning models. To address this issue, this study proposes a novel trajectory planning fram

Felisia Angela Chiarello, Alexander Keimer, Lukas Pflug

In this contribution, we study scalar nonlocal conservation laws with the $p$-norm. Here, 'nonlocal' means that the velocity of the conservation law depends on an integral term in space. Typically, the nonlocal term consists of integrating the solution in $L^{1}$, whereas here we will study the case when the solution is integrated in the $L^{p}$-norm. We con

Li Li, Xukai Yan, Zhibo Yang

We study the rigidity problem for $(-\alpha)$-homogeneous solutions to the two-dimensional incompressible stationary Euler equations in sector-type domains $\Omega_{a, b, \theta_0}:= \{(r,\theta): a<r<b, \ 0<\theta<\theta_0\}$, where $\alpha\in\mathbb{R}$, $0\leqslant a < b \leqslant +\infty$ and $0< \theta_0 \leqslant 2\pi$. For each type of domains, depend

Pengchao Feng, Yao Xiao, Ziyang Ma, Zhikang Niu

Recent advances in text-to-speech (TTS) have yielded remarkable improvements in naturalness and intelligibility. Building on these achievements, research has increasingly shifted toward enhancing the expressiveness of generated speech, such as dialectal and emotional TTS. However, cross-style synthesis combining both dialect and emotion remains challenging a

Mehrdad Salimnejad, Marios Kountouris, Nikolaos Pappas

We investigate real-time tracking of two correlated stochastic processes over a shared wireless channel. The joint evolution of the processes is modeled as a two-dimensional discrete-time Markov chain. Each process is observed by a dedicated sampler and independently reconstructed at a remote monitor according to a task-specific objective. Although both proc

Giuseppe Cosma Brusca

We analyze a family of non-local integral functionals of convolution-type depending on two small positive parameters $\varepsilon,\delta$: the first rules the length-scale of the non-local interactions and produces a `localization' effect as it tends to $0$, the second is the scale of oscillation of a finely inhomogeneous periodic structure in the domain. We

Lukas M Payne, Furqan Alabdullah, Paola Borri, Wolfgang Langbein

Quantitative polarisation-resolved optical extinction microscopy of individual plasmonic nanoparticles has recently been introduced as a powerful tool to characterise the nanoparticle's morphology with a precision comparable to electron microscopy, while using a simple optical microscope [Nanoscale 12, 16215 (2020)]. Here we provide a step change by adding m

Emiko Hiyama, Atsushi Hosaka, Makoto Oka, Georg Wolschin

A potential quark model is used to search for a $P_{c\bar{c}s}^0=(c\bar{c}uds)^0$, $J^P=1/2^-$ pentaquark state that has recently been observed experimentally by the LHCb collaboration at 4338.2 MeV, with a width of 7.0 MeV and high statistical significance $>15\sigma$. Our model Hamiltonian reproduces the masses of the low-lying charmed and strange hadrons.

Hai-Feng Wang, Guan-Yu Wang, Giovanni Carraro, Yuan-Sen Ting

We present GS3 Hunter (Galactic-Seismology Substructures and Streams Hunter), a novel deep-learning method that combines Siamese Neural Networks and K-means clustering to identify substructures and streams in stellar kinematic data. Applied to Gaia EDR3 and GALAH DR3, it recovers known groups (e.g., Thamnos, Helmi, GSE, Sequoia) and, with DESI dataset, revea

Jongmin Park, Minh-Quan Viet Bui, Juan Luis Gonzalez Bello, Jaeho Moon

Feed-forward 3D Gaussian Splatting (3DGS) enables efficient one-pass scene reconstruction, providing 3D representations for novel view synthesis without per-scene optimization. However, existing methods typically predict pixel-aligned primitives per-view, producing an excessive number of primitives in dense-view settings and offering no explicit control over

V. V. Flambaum, H. B. Tran Tan

Charged spin 1 (vector) particles behave very differently from electrons or scalars in a Coulomb field. For an infinitely heavy point-like nucleus their bound state wave functions fall to the centre, and embedding the system in a renormalisable electroweak-type theory does not remedy this short-distance pathology. We therefore solve the pure Coulomb problem

Lijuan Liu, Hanzhao Yang

Significant photospheric magnetic field changes during major solar eruptions -- interpreted as coronal feedback from eruptions to the photosphere -- are well-observed. However, analogous short-time scale field changes preceding eruptions are rarely reported. In this study, we present the first detailed analysis of a pre-flare decrease in the photospheric hor

Xiangrui Cai, Shaocheng Ma, Lei Cao, Jie Li

Electroencephalography (EEG) signal decoding is a key technology that translates brain activity into executable commands, laying the foundation for direct brain-machine interfacing and intelligent interaction. To address the inherent spatiotemporal heterogeneity of EEG signals, this paper proposes a multi-branch parallel architecture, where each temporal sca

Michał Kijaczko

We find sharp constants in fractional Hardy inequalities for weighted Triebel--Lizorkin seminorms on the whole space and half-spaces. Our results generalize recently obtained weighted fractional Hardy inequalities for Gagliardo seminorms, but are new even for the unweighted case.

Yosuke Taniuchi, Chie Hieida, Atsushi Noritake, Kazushi Ikeda

Social comparison$\unicode{x2014}$the process of evaluating one's rewards relative to others$\unicode{x2014}$is an essential feature of social emotions such as envy and plays a fundamental role in primate social cognition. However, it remains unknown how information about others' rewards affects one's own reward valuation. This study examines whether monkeys

Yuan Ji-Bing, Song Ya-Ju, Tang Shi-Qing, Wang Xin-Wen

We investigate the quantum sensing performance of a dephasing qubit as a probe in Ohmic environments, characterized by the coupling strength $\eta$, the Ohmicity parameter $s$, and the cutoff frequency $\omega_c$ to be estimated. The performance is quantified by the dimensionless quantum signal-to-noise ratio $\mathcal{Q}$. We show that the evolution of $\ma

Maarten Solleveld

Consider a reductive group G over a non-archimedean local field. The Galois group Gal(C/Q) acts naturally on the category of smooth complex G-representations. We prove that this action stabilizes the class of standard modules. This generalizes and relies on an analogous result about essentially square-integrable representations. Other important objects in th

Huimin Wu, Kwang-Ting Cheng, Stephen Lin, Zhirong Wu

This paper presents an investigation of vision transformer learning for multi-view geometry tasks, such as optical flow estimation, by fine-tuning video foundation models. Unlike previous methods that involve custom architectural designs and task-specific pretraining, our research finds that general-purpose models pretrained on videos can be readily transfer

Sebastian Sun

Recent advances in retrieval-augmented generation (RAG) have shown promise in enhancing recommendation systems with external knowledge. However, existing RAG-based recommenders face two critical challenges: (1) vulnerability to distribution shifts across different environments (e.g., time periods, user segments), leading to performance degradation in out-of-

Yuchen Li, Handing Wang, Bing Xue, Mengjie Zhang

In the high-cost simulation-driven design domain, translating ambiguous design requirements into a mathematical optimization formulation is a bottleneck for optimizing product performance. This process is time-consuming and heavily reliant on expert knowledge. While large language models (LLMs) offer potential for automating this task, existing approaches ei

Rohan Sarkar, Sharmistha Maji, Tuhin Subhra Pal, Achal Dharmalal Rajratna

Animals that display behavioural flexibility and adaptability thrive in urban environments, due to their ability to exploit novel anthropogenic resources. Since humans are an important component of such urban environments, animals that apply heterospecific learning in their decision-making are more likely to succeed as urban adapters. Free-ranging dogs, that

D. Ortega, E. Korutcheva

Segregation is a growing concern around the world. One of its main manifestations is the creation of ghettos, whose inhabitants have difficult access to well-paid jobs, which are often located far from their homes. In order to study this phenomenon, we propose an extension of Schelling's model of segregation to take into account the existence of economic exc

Maxime Kayser, Maksim Gridnev, Wanting Wang, Max Bain

We present brat (brain report alignment transformer), a multi-view representation learning framework for brain magnetic resonance imaging (MRI) trained on MRIs paired with clinical reports. Brain MRIs present unique challenges due to the presence of numerous, highly varied, and often subtle abnormalities that are localized to a few slices within a 3D volume.

Qiang Guo, Rubo Zhang, Bingbing Zhang, Junjie Liu

Crowd counting typically relies on labor-intensive point-level annotations and computationally intensive backbones, restricting its scalability and deployment in resource-constrained environments. To address these challenges, this paper proposes the TCFormer, a tiny, ultra-lightweight, weakly-supervised transformer-based crowd counting framework with only 5

Daniel Czarnowske, Amrei Stammann

Inference for fixed effects estimators is often unreliable due to Nickell- and incidental parameter biases. While these issues are well understood for classical two-dimensional panels, little is known about three-dimensional panel structures (e.g., sender x receiver x time). We develop inferential theory for a broad class of linear and nonlinear fixed effect

David Berghaus, Andriy Bondarenko, Danylo Radchenko, Kristian Seip

The basis functions of the Fourier interpolation formula of Radchenko and Viazovska, constructed by means of weakly holomorphic modular forms for the Hecke theta group, are entire functions of order $2$ having interesting time-frequency properties. We give precise size estimates and study the distribution of zeros of these functions. We give in particular as

Tamim Aziz, Sanjoy Ghosal

If $\omega_t > \beta$ for every $t \in \mathbb{N}$ and for some $\beta > 0$, then the sequence $\{\omega_t\}_{t \in \mathbb{N}}$ represents a weighted sequence of real numbers. In this article, we primarily introduce the concepts of rough weighted ideal limit set and rough weighted ideal cluster points set associated with sequences in normed spaces. Building

Longhuan Xu, Feng Yin, Cunjian Chen

Text-to-image diffusion inference typically follows synchronized schedules, where the numerical integrator advances the latent state to the same timestep at which the denoiser is conditioned. We propose an asynchronous inference mechanism that decouples these two, allowing the denoiser to be conditioned at a different, learned timestep while keeping image up

Wentao Liu, Yuhao Hu, Ruiting Zhou, Baochun Li

Mixture-of-Experts (MoE) has become a dominant architecture in large language models (LLMs) due to its ability to scale model capacity via sparse expert activation. Meanwhile, serverless computing, with its elasticity and pay-per-use billing, is well-suited for deploying MoEs with bursty workloads. However, the large number of experts in MoE models incurs hi

Ning Lyu, Junjie Jiang, Lu Chang, Chihui Shao

This paper proposes a structure-aware driven scheduling graph modeling method to improve the accuracy and representation capability of anomaly identification in scheduling behaviors of complex systems. The method first designs a structure-guided scheduling graph construction mechanism that integrates task execution stages, resource node states, and schedulin

Ugo Moschella

We construct a manifestly covariant and coordinate-free plane-wave representation of scalar two-point functions in $d$-dimensional anti-de Sitter spacetime. The construction is based on a new class of holomorphic plane waves defined globally on the universal covering of the AdS via chiral cones in the complex null cone. Imposing AdS invariance, locality, pos

Yiming Sun, Mi Zhang, Feifei Li, Geng Hong

Despite Video Large Language Models having rapidly advanced in recent years, perceptual hallucinations pose a substantial safety risk, which severely restricts their real-world applicability. While several methods for hallucination mitigation have been proposed, they often compromise the model's capacity for video understanding and reasoning. In this work, w

Xue Yang, Michael Schukat, Junlin Lu, Patrick Mannion

Reinforcement learning (RL) excels in various applications but struggles in dynamic environments where the underlying Markov decision process evolves. Continual reinforcement learning (CRL) enables RL agents to continually learn and adapt to new tasks, but balancing stability (preserving prior knowledge) and plasticity (acquiring new knowledge) remains chall

Jones David, Shreya Ghosh

LLM-based tutors are typically single-turn assistants that lack persistent representations of learner knowledge, making it difficult to provide principled, transparent, and long-term pedagogical support. We introduce IntelliCode, a multi-agent LLM tutoring system built around a centralized, versioned learner state that integrates mastery estimates, misconcep

Dibyendu Biswas

We study a new perspective on a certain Pieri rules for classical groups. Furthermore, we extend a fundamental theorem of Kostant concerning tensor products for classical groups. We show that a certain form of the Pieri rule is equivalent to the converse of this extended version of Kostant's theorem. In addition, we show an equivalence between the Pieri rule

Pietro Menotti

We give a recursive method to compute the classical conformal blocks in Liouville field theory. The values of the expansion coefficients are given by an algebraic scheme which works to all orders. The algebraic expression of the intervening matrices are explicitly given. With regard to the problem of the convergence of the series we rigorously prove that it

Dmitry Bennett, Fernand Gobet

A key issue in cognitive science concerns the fundamental psychological processes that underlie the formation and retrieval of multiple types of concepts in short-term and long-term memory (STM and LTM, respectively). We propose that chunking mechanisms play an essential role and show how the CogAct computational model grounds concept learning in fundamental

Panagiotis Charalambous, Laura Donnay, Alexandru Lupsasca

The Laplace equation on Euclidean flat space admits a discrete radial inversion symmetry. In 1983, Couch and Torrence (CT) found -- surprisingly -- that the massless wave equation continues to display this symmetry on the background of an extremal (and asymptotically flat) black hole, where the inversion interchanges horizon and infinity while preserving the

Shoeb Athar, Jeremy Guazzagaloppa, Fabrice Boyrie, Cedric Huillet

Thermoelectric generators (TEGs) based on commercially used thermal super-insulating materials can facilitate sustainable and large-scale ambient waste heat recovery while bequeathing an added economic and environmental value to thermal insulations in industry. This requires the optimization of the thermoelectric (TE) properties through electrical functional

Chihiro Noguchi, Takaki Yamamoto

End-to-end (E2E) autonomous driving models that take only camera images as input and directly predict a future trajectory are appealing for their computational efficiency and potential for improved generalization via unified optimization; however, persistent failure modes remain due to reliance on imitation learning (IL). While online reinforcement learning

Hafiz Saif Ur Rehman, Ling Liu, Kaleem Ullah Qasim

Financial time series forecasting is fundamentally an information fusion challenge, yet most existing models rely on static architectures that struggle to integrate heterogeneous knowledge sources or adjust to rapid regime shifts. Conventional approaches, relying exclusively on historical price sequences, often neglect the semantic drivers of volatility such

Pengxiang Ouyang, Qing Ma, Zheng Wang, Cong Bai

Remote sensing (RS) image-text retrieval faces significant challenges in real-world datasets due to the presence of Pseudo-Matched Pairs (PMPs), semantically mismatched or weakly aligned image-text pairs, which hinder the learning of reliable cross-modal alignments. To address this issue, we propose a novel retrieval framework that leverages Cross-Modal Gate

Qinyi Liu, Lin Li, Valdemar Švábenský, Conrad Borchers

The expansion of large-scale online education platforms has made vast amounts of student interaction data available for knowledge tracing (KT). KT models estimate students' concept mastery from interaction data, but their performance is sensitive to input data quality. Gaming behaviors, such as excessive hint use, may misrepresent students' knowledge and und

Pierre Colombo, Malik Boudiaf, Allyn Sweet, Michael Desa

Before closing venture capital financing rounds, lawyers conduct diligence that includes tying out the capitalization table: verifying that every security (for example, shares, options, warrants) and issuance term (for example, vesting schedules, acceleration triggers, transfer restrictions) is supported by large sets of underlying legal documentation. While

Yingli Cui, Xiangliang Kong, Zhentong Li, Bing Wang

Solar type II radio bursts are commonly attributed to coronal shocks driven by coronal mass ejections (CMEs). However, some metric type II bursts have occasionally been reported to occur in the absence of a CME and to be associated with weak solar activities. This study aims to identify the driver of the coronal shock in this kind of type II event. We invest

Yue Wen, Liang Song, Hesheng Wang

Low-light 3D reconstruction from sparse views remains challenging due to exposure imbalance and degraded color fidelity. While existing methods struggle with view inconsistency and require per-scene training, we propose SplatBright, which is, to our knowledge, the first generalizable 3D Gaussian framework for joint low-light enhancement and reconstruction fr

Rahim Rahmati-asghar

We introduce and systematically study \emph{hierarchical filtrations} of vector bundles on smooth projective varieties. These are filtrations by saturated subsheaves of equal rank whose successive quotients are torsion sheaves supported in codimension one. The associated numerical invariant, called \emph{hierarchical depth}, measures the maximal length of su

Shoeb Athar, Adrien Mecibah, Philippe Jund

Machine Learning (ML) driven discovery of novel and efficient thermoelectric (TE) materials warrants experimental TE datasets of high volume, diversity, and quality. While the largest publicly available dataset, Starrydata2, has a high data volume, it contains inaccurate data due to the inherent limitations of Large Language Model (LLM)-assisted data curatio

Muhammad Usman, Sung-Min Gho

Susceptibility Map Weighted Imaging (SMWI) is an advanced magnetic resonance imaging technique used to detect nigral hyperintensity in Parkinsons disease. However, full resolution SMWI acquisition is limited by long scan times. Efficient reconstruction methods are therefore required to generate high quality SMWI from reduced k space data while preserving dia

Zhiyuan Peng, Zihan Ye, Shreyank N Gowda, Yuping Yan

Zero-shot Learning (ZSL) aims to enable image classifiers to recognize images from unseen classes that were not included during training. Unlike traditional supervised classification, ZSL typically relies on learning a mapping from visual features to predefined, human-understandable class concepts. While ZSL models promise to improve generalization and inter

Kangmin Chen, Xiaoying Qu, Hui Tong, Sibo Wang

Constraining the nuclear matter equation of state (EOS) beyond saturation density is a central goal of nuclear physics and astrophysics. While the relativistic Brueckner-Hartree-Fock (RBHF) theory, an \textit{ab initio,} non-perturbative nuclear many-body theory starting from realistic interactions, accurately describes nuclear matter properties near the sat

Oleg S. Balashov, Andrei V. Faminskii

Results on well-posedness of three inverse problems with integral conditions on a bounded interval for the generalized Korteweg-de Vries equation without any restrictions on the growth rate of nonlinearity are established. Either the right-hand side of equation or the boundary data, or both are chosen as controls. The considered solutions are regular with re

Sungwoo Kang

We establish a general matched filter principle for order flow normalization: optimal normalization must match the scaling behaviour of the signal-generating process. For capacity-constrained institutional investors, market capitalization normalization ($S^{MC}$) is the matched filter; for volume-targeting traders (e.g., VWAP/TWAP algorithms), trading value

Xuyao Deng, Yong Dou, Kele Xu

Deep learning-based direction-of-arrival (DoA) estimation has gained increasing popularity. A popular family of DoA estimation algorithms is beamforming methods, which operate by constructing a spatial filter that is applied to array signals. However, these spatial filters obtained by traditional model-driven beamforming algorithms fail under demanding condi

John Chiang

Privacy-preserving inference of convolutional neural networks (CNNs) using homomorphic encryption has emerged as a promising approach for enabling secure machine learning in untrusted environments. In our previous work, we introduced a matrix-encoding strategy that allows convolution and matrix multiplication to be efficiently evaluated over encrypted data,

Somayeh Habibi

Building on earlier work concerning the motives of $G$-bundles, we study the structure of motives associated with certain classes of $G$-varieties. In particular, we show that the corresponding motives lie within the category of mixed-Tate motives, under certain condition on the stabilizers. We further discuss some applications and provide some examples to i

Alexei D. Chepelianskii, Dima L. Shepelyansky

The quantum dissipative time evolution of a fluxonium under a pulsed field (kicks) is studied numerically and analytically. In the classical limit the system dynamics is converged to a strange chaotic attractor. The quantum properties of this system are studied for the density matrix in the frame of Lindblad equation. In the case of dissipative quantum evolu

Abdessatar Souissi, Amenallah Andolsi

This paper introduces a hidden quantum Markov models (HQMMs) framework to the Affleck-Kennedy-Lieb-Tasaki (AKLT) state-a cornerstone example of a symmetry-protected topological (SPT) phase. The model's observation system is the physical spin-1 chain, which emerges from a hidden spin-1/2 layer through well-defined quantum emission operation. We show that the

Kai Kohyama, Yoshimitsu Aoki, Guillermo Gallego, Shintaro Shiba

Event cameras offer a high temporal resolution over traditional frame-based cameras, which makes them suitable for motion and structure estimation. However, it has been unclear how event-based 3D Gaussian Splatting (3DGS) approaches could leverage fine-grained temporal information of sparse events. This work proposes GPERT, a framework to address the trade-o

D. Brioschi, A. Di Siena, R. Bilato, A. Bottino

In this work, the interplay between fast ions and safety factor rational surfaces is studied in a turbulent plasma via global nonlinear gyrokinetic simulations. Initially, the fast particles-induced enhancement of shearing structures from turbulence self-interaction is analyzed. Our study takes into account the competition between this mechanism and other fa

Xiyue Bai, Ronghao Yu, Jia Xiu, Pengfei Zhou

Generating or editing images directly from Neural signals has immense potential at the intersection of neuroscience, vision, and Brain-computer interaction. In this paper, We present Uni-Neur2Img, a unified framework for neural signal-driven image generation and editing. The framework introduces a parameter-efficient LoRA-based neural signal injection module

Ryotaro Kawata, Yujin Song, Alberto Bietti, Naoki Nishikawa

Transformers can implement both generalizable algorithms (e.g., induction heads) and simple positional shortcuts (e.g., memorizing fixed output positions). In this work, we study how the choice of pretraining data distribution steers a shallow transformer toward one behavior or the other. Focusing on a minimal trigger-output prediction task -- copying the to