November 2025 arXiv papers — page 150

Fengyi Fu, Mengqi Huang, Lei Zhang, Zhendong Mao

Text-driven multi-object image editing which aims to precisely modify multiple objects within an image based on text descriptions, has recently attracted considerable interest. Existing works primarily follow the localize-editing paradigm, focusing on independent object localization and editing while neglecting critical inter-object interactions. However, th

Xinyuan Liao, Xinyue Zhang, Xing Wei, Junwei Liu

Data-driven health monitoring of power converters remains limited by poor generalization to unseen operating conditions. This work addresses this out-of-distribution (OOD) challenge by building a domain-specific time-series foundation model (PE-TSFM) that learns representations directly from large-scale unlabeled converter data. Unlike generic TSFMs trained

Tommaso Confalone, Flavia Lo Sardo, Yejin Lee, Sanaz Shokri

Recent advances in the manipulation of complex oxide layers, particularly the fabrication of atomically thin cuprate superconducting films via molecular beam epitaxy, have revealed new ways in which nanoscale engineering can govern superconductivity and its interwoven electronic orders. In parallel, the creation of twisted cuprate heterostructures through cr

Kunal Mahatha, Jose Dolz, Christian Desrosiers

Despite recent advances in Open-Vocabulary Semantic Segmentation (OVSS), existing training-free methods face several limitations: use of computationally expensive affinity refinement strategies, ineffective fusion of transformer attention maps due to equal weighting or reliance on fixed-size Gaussian kernels to reinforce local spatial smoothness, enforcing i

Ziyu Ma, Chenhui Gou, Yiming Hu, Yong Wang

Large Multimodal Models (LMMs) have shown promising in-context learning (ICL) capabilities, but scaling to many-shot settings remains difficult due to limited context length and high inference cost. To address these challenges, task-vector-based methods have been explored by inserting compact representations of many-shot in-context demonstrations into model

Jisoo Jang, Tien-Cuong Bui, Yunjun Choi, Wen-Syan Li

This paper introduces an Error Correction through Prompt Tuning for NL-to-SQL, leveraging the latest advancements in generative pre-training-based LLMs and RAG. Our work addresses the crucial need for efficient and accurate translation of natural language queries into SQL expressions in various settings with the growing use of natural language interfaces. We

Elena Lanina, Radomir Stepanov

The Khovanov-Rozansky (KR) link polynomial is a certain $t$-deformation of Wilson loops in 3-dimensional $SU(N)$ Chern--Simons topological field theory, believed to be an observable in the refined Chern-Simons theory, probably described in terms of 4d or 5d QFT and related by a certain procedure to the triply-graded link superpolynomial. This link invariant

Xianshuai Shi, Jianfeng Zhu, Leibo Liu

The Transformer architecture has achieved tremendous success in natural language processing, computer vision, and scientific computing through its self-attention mechanism. However, its core components-positional encoding and attention mechanisms-have lacked a unified physical or mathematical interpretation. This paper proposes a structural theoretical frame

Waseem AlShikh, Muayad Sayed Ali, Brian Kennedy, Dmytro Mozolevskyi

As AI agents proliferate across industries and applications, evaluating their performance based solely on infrastructural metrics such as latency, time-to-first-token, or token throughput is proving insufficient. These metrics fail to capture the quality of an agent's decisions, its operational autonomy, or its ultimate business value. This white paper propo

Zhihao Lin, Lin Wu, Zhen Tian, Jianglin Lan

Exploration in reinforcement learning remains a critical challenge, as naive entropy maximization often results in high variance and inefficient policy updates. We introduce \textbf{PrefPoE}, a novel \textit{Preference-Product-of-Experts} framework that performs intelligent, advantage-guided exploration via the first principled application of product-of-expe

Chenyu Hu, Xiaotong Li, Hao Zhu, Biao Hou

Point cloud processing has become a cornerstone technology in many 3D vision tasks. However, arbitrary rotations introduce variations in point cloud orientations, posing a long-standing challenge for effective representation learning. The core of this issue is the disruption of the point cloud's intrinsic directional characteristics caused by rotational pert

Jonas Pronk, Oliver Porth, Jordy Davelaar

Flexible and accurate interpolation schemes using machine learning could be of great benefit for many use-cases in numerical simulations and post-processing, such as temporal upsampling or storage reduction. In this work, we adapt the physics-informed token transformer (PITT) network for multi-channel data and couple it with Fourier neural operator (FNO). Th

Xiangyang Wu, Liu Liu, Baosheng Yu, Jiayan Qiu

Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook this information when aligning vision and language, thus leading to suboptimal performance. Toward solving this problem,

Humera Hameed, Waqas Aman, Muhammad Mahboob Ur Rahman, Ali Arshad Nasir

The emergence of 6G networks enables simultaneous near-field and far-field communications through extremely large antenna arrays and high carrier frequencies. While these regimes enhance spatial multiplexing and link capacity, their coexistence poses new challenges in ensuring quality-of-service (QoS) guarantees for delay-sensitive applications. This paper p

Atreyee Majumdar, Raghunathan Ramakrishnan

Molecules that violate Hund's rule by exhibiting an inverted singlet-triplet gap (STG), where the first excited singlet (S$_1$) lies below the triplet (T$_1$), are rare but hold great promise as efficient fifth-generation light emitters. Azaphenalenes (APs) represent one of the few known molecular classes capable of such inversion of the S$_1$/T$_1$ energy o

Zhihao Lin

Gaussian policies have dominated continuous control in deep reinforcement learning (RL), yet they suffer from a fundamental mismatch: their unbounded support requires ad-hoc squashing functions that distort the geometry of bounded action spaces. While von Mises-Fisher (vMF) distributions offer a theoretically grounded alternative on the sphere, their relianc

Eito Ogawa, Taiga Hayami, Hiroshi Watanabe

Point cloud surface reconstruction has improved in accuracy with advances in deep learning, enabling applications such as infrastructure inspection. Recent approaches that reconstruct from small local regions rather than entire point clouds have attracted attention for their strong generalization capability. However, prior work typically places local regions

Alkid Baci, Luke Friedrichs, Caglar Demir, Axel-Cyrille Ngonga Ngomo

In this paper, we introduce OWLAPY, a comprehensive Python framework for OWL ontology engineering. OWLAPY streamlines the creation, modification, and serialization of OWL 2 ontologies. It uniquely integrates native Python-based reasoners with support for external Java reasoners, offering flexibility for users. OWLAPY facilitates multiple implementations of c

Devin Hunter, Chinwendu Enyioha

Various neural network architectures are used in many of the state-of-the-art approaches for real-time nonlinear state estimation. With the ever-increasing incorporation of these data-driven models into the estimation domain, model predictions with reliable margins of error are a requirement -- especially for safety-critical applications. This paper discusse

Heyang Liu, Ziyang Cheng, Yuhao Wang, Hongcheng Liu

The development of multi-modal large language models (LLMs) leads to intelligent approaches capable of speech interactions. As one of the most widely spoken languages globally, Mandarin is supported by most models to enhance their applicability and reach. However, the scarcity of comprehensive speech-to-speech (S2S) benchmarks in Mandarin contexts impedes sy

Junkai Lu, Peng Chen, Chenjuan Guo, Yang Shu

Time series forecasting is critical for decision-making across dynamic domains such as energy, finance, transportation, and cloud computing. However, real-world time series often exhibit non-stationarity, including temporal distribution shifts and spectral variability, which pose significant challenges for long-term time series forecasting. In this paper, we

Liam Mazurowski, Xin Zhou

Assume $h$ is a positive function on the unit three-sphere which satisfies the pinching condition $h < h_0 \approx 0.547$. We prove the existence of at least two embedded two-spheres with prescribed mean curvature $h$. The same result holds for sign-changing functions $h$ satisfying $\vert h\vert < h_0$ under a mild assumption on the zero set.

A. T. ~Bajkova, A. A. ~Smirnov, V. V. ~Bobylev

It is shown how the violation of the invariance of the $Z$-component of the orbital angular momentum $L_z$ in the axially symmetric potential of the Galaxy with a bar can serve as an indicator of the degree of orbital chaos of globular clusters in the central region of the Galaxy. In this case, the higher the variations of $L_z$ of the orbit over a certain p

Gang Liao, Jing Wei

In this paper, we investigate the relationship between chaos and homoclinic orbits from a quantitative perspective. Let f be a C^r diffeomorphism (r > 1) on a compact Riemannian manifold preserving an ergodic hyperbolic measure. We show that the homoclinic growth rate is bounded below by the metric entropy. This result generalizes the work of Mendoza from su

Yishan Du, Conrad Borchers, Mutlu Cukurova

As teachers increasingly turn to GenAI in their educational practice, we need robust methods to benchmark large language models (LLMs) for pedagogical purposes. This article presents an embedding-based benchmarking framework to detect bias in LLMs in the context of formative feedback. Using 600 authentic student essays from the AES 2.0 corpus, we constructed

Ignasi Mas, Ivan Huerta, Ramon Morros, Javier Ruiz-Hidalgo

We introduce 2Dto3D-SR, a versatile framework for real-time single-view 3D super-resolution that eliminates the need for high-resolution RGB guidance. Our framework encodes 3D data from a single viewpoint into a structured 2D representation, enabling the direct application of existing 2D image super-resolution architectures. We utilize the Projected Normaliz

Serafino Cicerone, Alessia Di Fonso, Gabriele Di Stefano, Alfredo Navarra

In the field of swarm robotics, one of the most studied problem is Gathering. It asks for a distributed algorithm that brings the robots to a common location, not known in advance. We consider the case of robots constrained to move along the edges of a graph under the well-known OBLOT model. Gathering is then accomplished once all the robots occupy a same ve

Hui-Min Fan, Alejandro Torres-Orjuela, Verónica Vázquez-Aceves, Tian-Xiao Wang

Extremely large mass-ratio inspirals (XMRIs), formed by brown dwarfs inspiraling into a massive black hole, emit gravitational waves (GWs) that fall within the detection band of future space-borne detectors such as LISA, TianQin, and Taiji. Their detection will measure the astrophysical properties of the MBH in the center of our galaxy (SgrA$^\ast$) with unp

Francesco Michele Ventrella, Victor de Jesus Valadão, Guido Boffetta, Stefano Musacchio

Two-dimensional turbulence with linear (Ekman) friction exhibits spectral properties that deviate from the classical Kraichnan prediction for the direct enstrophy cascade. In particular, for sufficiently small viscosity and large friction, the enstrophy flux is suppressed in the cascade and, as a consequence, the small-scale vorticity field becomes passively

Rune Rost, Lorenzo Branca, Tobias Buck

Radiative transfer is a fundamental process in astrophysics, essential for both interpreting observations and modeling thermal and dynamical feedback in simulations via ionizing radiation and photon pressure. However, numerically solving the underlying radiative transfer equation is computationally intensive due to the complex interaction of light with matte

John Beirne, Haroon Mumtaz, Donghyun Park, Gazi Salah Uddin

This paper investigates the heterogeneous effects of military spending news shocks on household income and wealth inequality for a large, panel of advanced and emerging economies. Confirming prior literature, we find that military spending news shocks lead to persistent increases in aggregate output and Total Factor Productivity. Our primary contribution is

Gleb V. Solovev, Alina B. Zhidkovskaya, Anastasia Orlova, Nina Gubina

Hit identification is a central challenge in early drug discovery, traditionally requiring substantial experimental resources. Recent advances in artificial intelligence, particularly large language models (LLMs), have enabled virtual screening methods that reduce costs and improve efficiency. However, the growing complexity of these tools has limited their

Thomas J. Maullin-Sapey, Fabian J. E. Telschow

In scientific disciplines such as neuroimaging, climatology, and cosmology it is useful to study the uncertainty of excursion sets of imaging data. While the case of imaging data obtained from a single study condition has already been intensively studied, confidence statements about the intersection, or union, of the excursion sets derived from different sub

Rizal Khoirul Anam

The proliferation of digital food applications necessitates robust methods for automated nutritional analysis and culinary guidance. This paper presents a comprehensive comparative evaluation of a decoupled, multimodal pipeline for food recognition. We evaluate a system integrating a specialized visual backbone (EfficientNet-B4) with a powerful generative la

Yi Huang, Zhan Qu, Lihui Jiang, Bingbing Liu

End-to-end autonomous driving systems, predominantly trained through imitation learning, have demonstrated considerable effectiveness in leveraging large-scale expert driving data. Despite their success in open-loop evaluations, these systems often exhibit significant performance degradation in closed-loop scenarios due to causal confusion. This confusion is

Rajit Rajpal, Benedict Leimkuhler, Yuanhao Jiang

Bayesian neural networks (BNNs) require scalable sampling algorithms to approximate posterior distributions over parameters. Existing stochastic gradient Markov Chain Monte Carlo (SGMCMC) methods are highly sensitive to the choice of stepsize and adaptive variants such as pSGLD typically fail to sample the correct invariant measure without addition of a cost

Kaito Kokubu

We study travelling wave solutions to Korteweg--de Vries type equations which have double power nonlinearities with integer indices, such as the Gardner equation, and fractional dispersion. Whether these equations have ground state solutions depends on signatures of nonlinearities and parity combinations of the two indices. The aim of this study is to give t

Taisuke Izumi, Naoki Kitamura, Yutaro Yamaguchi

Finding a maximum cardinality matching in a graph is one of the most fundamental problems. An algorithm proposed by Micali and Vazirani (1980) is well-known to solve the problem in $O(m\sqrt{n})$ time, which is still one of the fastest algorithms in general. While the MV algorithm itself is not so complicated and is indeed convincing, its correctness proof i

Shinichi Arita, Kenichi Ito

We present direct and elementary commutator techniques for the Dirac equation with long-range electric and mass perturbations. The main results are absence of generalized eigenfunctions and locally uniform resolvent estimates, both in terms of the optimal Besov-type spaces. With an additional massless assumption, we also obtain an algebraic radiation conditi

Devriş İşler, Elina van Kempen, Seoyeon Hwang, Nikolaos Laoutaris

Federated learning (FL) offers privacy preserving, distributed machine learning, allowing clients to contribute to a global model without revealing their local data. As models increasingly serve as monetizable digital assets, the ability to prove participation in their training becomes essential for establishing ownership. In this paper, we address this emer

Xiao Yang, Xuejiao Zhao, Zhiqi Shen

Structured Electronic Health Record (EHR) data stores patient information in relational tables and plays a central role in clinical decision-making. Recent advances have explored the use of large language models (LLMs) to process such data, showing promise across various clinical tasks. However, the absence of standardized evaluation frameworks and clearly d

Silvie Illésová, Tomáš Bezděk, Vojtěch Novák, Ivan Zelinka

This work addresses the challenge of enabling practitioners without quantum expertise to transition from classical to hybrid quantum-classical machine learning workflows. We propose a three-stage framework: starting with a classical self-training model, then introducing a minimal hybrid quantum variant, and finally applying diagnostic feedback via QMetric to

Shivam Rawat, Lucie Flek, Akbar Karimi

Scientific literature in astronomy is rapidly expanding, making it increasingly important to automate the extraction of key entities and contextual information from research papers. In this paper, we present an encoder-based system for extracting knowledge from astronomy articles. Our objective is to develop models capable of classifying telescope references

Kyle Fogarty, Jack Foster, Boqiao Zhang, Jing Yang

Despite their impressive results, large-scale image-to-3D generative models remain opaque in their inductive biases. We identify a significant limitation in image-conditioned 3D generative models: a strong canonical view bias. Through controlled experiments using simple 2D rotations, we show that the state-of-the-art Hunyuan3D 2.0 model can struggle to gener

Zun-Yan Di, Zhi-Gang Wang

In this article, we take the $X(4140)$ as the axialvector tetraquark state with the symbolic quark structure $[sc]_S[\bar{s}\bar{c}]_A+[sc]_A[\bar{s}\bar{c}]_S$, and calculate the width of the two-body strong decay $X(4140)\rightarrow J/\psi \phi$ within the framework of the light-cone sum rules. Different from the traditional light-cone sum rules, at the ph

Or Peretz, Tai Dinh, Michal Koren

We investigate whether shallow quantum circuits can accurately reproduce the short-horizon dynamics of discrete-time Markov chains derived from fashion electronic-commerce recommendation links. Transition operators are compiled into block-encoded circuits and iterated using fixed-point oblivious amplitude amplification, and amplitude-encoded marginals are us

Arzu Burcu Güven, Anna Rogers, Rob van der Goot

We examine the syntactic properties of BabyLM corpus, and age-groups within CHILDES. While we find that CHILDES does not exhibit strong syntactic differentiation by age, we show that the syntactic knowledge about the training data can be helpful in interpreting model performance on linguistic tasks. For curriculum learning, we explore developmental and sever

Omri Isac, Idan Refaeli, Haoze Wu, Clark Barrett

The widespread adoption of deep neural networks (DNNs) requires efficient techniques for verifying their safety. DNN verifiers are complex tools, which might contain bugs that could compromise their soundness and undermine the reliability of the verification process. This concern can be mitigated using proofs: artifacts that are checkable by an external and

Bangti Jin, Fengru Wang, Jun Zou

We propose in this work a novel iterative direct sampling method for imaging moving inhomogeneities in parabolic problems using boundary measurements. It can efficiently identify the locations and shapes of moving inhomogeneities when very limited data are available, even with only one pair of lateral Cauchy data, and enjoys remarkable numerical stability fo

Arnav Aditya, Nitin Kumar, Saurabh Shigwan

Driven by advancements in deep learning, computer-aided diagnoses have made remarkable progress. However, outside controlled laboratory settings, algorithms may encounter several challenges. In the medical domain, these difficulties often stem from limited data availability due to ethical and legal restrictions, as well as the high cost and time required for

Antoine Monier, Kévin Gutierrez, Cyrille Claudet, Franck Celestini

Surface bubbles are an abundant source of aerosols, with important implications for climate processes. In this context, we investigate the stability and thinning dynamics of soap films under effective gravity fields. Experiments are performed using a centrifugal thin-film balance capable of generating accelerations from 0.2 up to 100 times standard gravity,

Chen Chen, Kejun Liu, Dezhou Deng, Shuchang Ma

We present a compact design of dual-beam Zeeman slower optimized for efficient production of cold atom applications. Traditional single-beam configurations face challenges from substantial residual atomic flux impacting downstream optical windows, resulting in increased system size, atomic deposition contamination, and a reduced operational lifetime. Our app

Lydia Kakampakou, Jennifer L. Wadsworth

Recent developments in extreme value statistics have established the so-called geometric approach as a powerful modelling tool for multivariate extremes. We tailor these methods to the case of spatial modelling and examine their efficacy at inferring extremal dependence and performing extrapolation. The geometric approach is based around a limit set describe

Ruihan Zhang, Jun Sun, Ee-Peng Lim, Peixin Zhang

The recent success of machine learning models, especially large-scale classifiers and language models, relies heavily on training with massive data. These data are often collected from online sources. This raises serious concerns about the protection of user data, as individuals may not have given consent for their data to be used in training. To address thi

M. Mangut, O. Gurtug, M. Halilsoy

The scalar quantum probe method developed by Horowitz and Marolf is applied to the cylindrically symmetric Curzon solution. The main cause for choosing the Curzon solution is that it is the best known example that exhibits directional singularity. Interestingly the singularity at $r=0$, for the uncharged Curzon spacetime, which is classically very strong wit

Mingyuan Song, Huan Shen, Jinghui Jiang, Qiang Su

Distributed in-network programs are increasingly deployed in data centers for their performance benefits, but shifting application logic to switches also enlarges the failure domain. Ensuring their correctness before deployment is thus critical for reliability. While prior verification frameworks can efficiently detect bugs for programs running on a single s

Sourav Mukherjee, Bho Matthiesen, Armin Dekorsy, Petar Popovski

6G networks are expected to integrate low Earth orbit satellites to ensure global connectivity by extending coverage to underserved and remote regions. However, the deployment of dense mega-constellations introduces severe interference among satellites operating over shared frequency bands. This is, in part, due to the limited flexibility of conventional fre

Oguz Umut Salman, Aylin Ahadi, Lev Truskinovsky

Quasi-brittle plastic yielding is a salient feature of well-annealed glassy materials. Here we show that the same behavior is characteristic of perfect crystals after they experience mechanically driven elastic instability leading to massive nucleation of dislocations. We argue that such 'preparation' effectively converts an atomic configuration from crystal

Yunhui Zhu, Buliao Huang

Modern oriented object detectors typically predict a set of bounding boxes and select the top-ranked ones based on estimated localization quality. Achieving high detection performance requires that the estimated quality closely aligns with the actual localization accuracy. To this end, existing approaches predict the Intersection over Union (IoU) between the

Tai Hoang, Alessandro Trenta, Alessio Gravina, Niklas Freymuth

Learning to simulate complex physical systems from data has emerged as a promising way to overcome the limitations of traditional numerical solvers, which often require prohibitive computational costs for high-fidelity solutions. Recent Graph Neural Simulators (GNSs) accelerate simulations by learning dynamics on graph-structured data, yet often struggle to

Sameeksha Sriram, Ayush Paliwal, Alexander S. Ecker, Chase van de Geijn

Rotary Positional Embeddings (RoPE) have demonstrated exceptional performance as a positional encoding method, consistently outperforming their baselines. While recent work has sought to extend RoPE to higher-dimensional inputs, many such extensions are non-commutative, thereby forfeiting RoPE's shift-equivariance property. Spherical RoPE is one such non-com

Kentaro Fukumoto

When scholars suspect units are dependent on each other within clusters but independent of each other across clusters, they employ cluster-robust standard errors (CRSEs). Nevertheless, what to cluster over is sometimes unknown. For instance, in the case of cross-sectional survey samples, clusters may be households, municipalities, counties, or states. A few

Seung Hwan Cho, Yujin Yang, Danik Baeck, Minjoo Kim

Recommender systems (RS) are currently being studied to mitigate limitations during cold-start conditions by leveraging modality information or introducing Agent concepts based on the exceptional reasoning capabilities of Large Language Models (LLMs). Meanwhile, food and beverage recommender systems have traditionally used knowledge graph and ontology concep

Guido Masarotto

This paper presents a comprehensive algorithm for fitting generative models whose likelihood, moments, and other quantities typically used for inference are not analytically or numerically tractable. The proposed method aims to provide a general solution that requires only limited prior information on the model parameters. The algorithm combines a global sea

Kaitao Huang, Yan Yan, Jing-Hao Xue, Hanzi Wang

3D GAN inversion projects a single image into the latent space of a pre-trained 3D GAN to achieve single-shot novel view synthesis, which requires visible regions with high fidelity and occluded regions with realism and multi-view consistency. However, existing methods focus on the reconstruction of visible regions, while the generation of occluded regions r

Yasmine Elfares, Gül Çalikli, Mohamed Khamis

AI-powered coding assistants, like GitHub Copilot, are increasingly used to boost developers' productivity. However, their output quality hinges on the contextual richness of the prompts. Meanwhile, gaze behaviour carries rich cognitive information, providing insights into how developers process code. We leverage this in Real-time GazeCopilot, a novel approa

Jan T. Weber, Kristin Kliemt, Sergey L. Bud'ko, Paul C. Canfield

Cerium diantimonide (CeSb$_2$) is a layered heavy-fermion Kondo lattice material that hosts complex magnetism and pressure-induced superconductivity. The interpretation of its in-plane anisotropy has remained unsettled due to structural twinning, which superimposes orthogonal magnetic responses. Here we combine controlled crystal growth with magnetization an

Annarien G. Headley, James O. Chibueze

We investigated the distributions of classified young stellar object (YSO) in nine star-forming regions associated with Hii regions, namely Sh2-22, Sh2-19, Sh2-17, M16, M8, IC5070, Sh2-252, NGC2467 and M42, as a means of exploring the star formation scenarios (triggered or spontaneous) in the various regions. The YSO distributions of nine regions along the g

Hang Xu, Kai Li, Haobo Fu, Qiang Fu

Counterfactual regret minimization (CFR) is a family of algorithms for effectively solving imperfect-information games. To enhance CFR's applicability in large games, researchers use neural networks to approximate its behavior. However, existing methods are mainly based on vanilla CFR and struggle to effectively integrate more advanced CFR variants. In this

Samet Hicsonmez, Abd El Rahman Shabayek, Djamila Aouada

Detecting visual anomalies in diverse, multi-class real-world images is a significant challenge. We introduce \ours, a novel unsupervised multi-class visual anomaly detection framework. It integrates a Latent Diffusion Model (LDM) with a Vision-Language Model (VLM) for enhanced anomaly localization and detection. Specifically, a pre-trained VLM with a simple

Georgios Pantazopoulos, Eda B. Özyiğit

Visual grounding is the task of localising image regions from natural language queries and is critical for reasoning capable Graphical User Interface agents. Many existing methods rely on massive, noisy synthetic datasets. This work introduces an efficient training pipeline that combines model-based data filtering with parameter-efficient fine-tuning. From 4

Bangti Jin, Fengru Wang, Jun Zou

We develop a novel iterative direct sampling method (IDSM) for solving linear or nonlinear elliptic inverse problems with partial Cauchy data. It integrates three innovations: a data completion scheme to reconstruct missing boundary information, a heterogeneously regularized Dirichlet-to-Neumann map to enhance the near-orthogonality of probing functions, and

Mingyu Sung, Suhwan Im, Vikas Palakonda, Jae-Mo Kang

Split computing distributes deep neural network inference between resource-constrained edge devices and cloud servers but faces significant communication bottlenecks when transmitting intermediate features. To this end, in this paper, we propose a novel lightweight compression framework that leverages Range Asymmetric Numeral Systems (rANS) encoding with asy

Zhi Chen, Yadan Luo, Zi Huang, Jingjing Li

In this paper, we propose a Distributed Zero-Shot Learning (DistZSL) framework that can fully exploit decentralized data to learn an effective model for unseen classes. Considering the data heterogeneity issues across distributed nodes, we introduce two key components to ensure the effective learning of DistZSL: a cross-node attribute regularizer and a globa

Xinhui Yin, Qifei Li, Yilin Guo, Hongxia Xie

Image composition aims to seamlessly integrate a foreground object into a background, where generating realistic and geometrically accurate shadows remains a persistent challenge. While recent diffusion-based methods have outperformed GAN-based approaches, existing techniques, such as the diffusion-based relighting framework IC-Light, still fall short in pro

Ryusuke Mizutani, Kazuaki Matano, Tsugumi Kadowaki, Haruki Tenya

This project was conducted as a 2nd-term adopted project of the "Post-5G Information and Communication System Infrastructure Enhancement R&D Project Development of Competitive Generative AI Foundation Models (GENIAC)," a business of the Ministry of Economy, Trade and Industry (METI) and the New Energy and Industrial Technology Development Organization (NEDO)

Feng Lu, Ao Zhou, Shujie Cheng, Gao Xianlong

We study the topological properties of rhombohedral-stacked N-layer Su-Schrieffer-Heeger networks with interlayer coupling. We find that these systems exhibit $2N$-fold degenerate zero-energy edge states with winding number $W=N$, providing a direct route to high-winding-number topological phases where $W$ equals the layer number. Using effective Hamiltonian

Neetesh Dhakar, Sunil Kumar

Following the ultrafast photoexcitation of a semiconductor, it embodies competing dynamics among photocarriers, many-body transient states of highly energetic excitons, and electron-hole liquid. Here, we show that femtosecond optical pulse excitation induces transient excitonic shift current contributing to stronger THz emission from a single crystalline bul

Meng Zhan, Miao Han, Yayao Zhang, Hongsheng Xu

The change of electric power generation - from synchronous generator (SG) to converter - is generally regarded as the second revolution of power system. Different from rotor swing of SG in traditional grids mainly described by the swing equation (SE), the converter dynamics plays an indispensable role in modern renewable dominated power systems (RDPS). The h

Thi Tam Dang, Lukas Einkemmer, Alexander Ostermann

Treating diffusion and advection/reaction separately is an effective strategy for solving semilinear advection-diffusion-reaction equations. However, such an approach is prone to suffer from order reduction, especially in the presence of inhomogeneous Dirichlet boundary conditions. In this paper, we extend an approach of Einkemmer and Ostermann [SIAM J. Sci.

Ning Wang, Long Yu, Cong Hua, Guangming Zhu

Zero-shot learning (ZSL) aims to recognize unseen classes with zero samples by transferring semantic knowledge from seen classes. Current approaches typically correlate global visual features with semantic information (i.e., attributes) or align local visual region features with corresponding attributes to enhance visual-semantic interactions. Although effec

Stefan Hetzl, Johannes Kloibhofer

We consider a class of formula equations in first-order logic, Horn formula equations, which are defined by a syntactic restriction on the occurrences of predicate variables. Horn formula equations play an important role in many applications in computer science. We state and prove a fixed-point theorem for Horn formula equations in first-order logic with a l

Hyunsuk Hong, Jae Sung Lee, Hyunggyu Park

We study a generalized Kuramoto model in which each oscillator carries two coupled phase variables, representing a minimal swarmalator system. Assuming perfect correlation between the intrinsic frequencies associated with each phase variable, we identify a novel dynamic mode characterized by bounded oscillatory motion that breaks the $\pi$-reflection symmetr

Argyris Deligkas, Eduard Eiben, Tiger-Lily Goldsmith, Dušan Knop

In standard fair division models, we assume that all agents are selfish. However, in many scenarios, division of resources has a direct impact on the whole group or even society. Therefore, we study fair allocations of indivisible items that, at the same time, maximize social impact. In this model, each agent is associated with two additive functions that de

Ludwik Dąbrowski, Sugato Mukhopadhyay, Filip Požar

We compute the nonvanishing spectral torsion functional of the internal part of the noncommutative geometry behind the Standard Model. We show that with a suitable modification of the usual differential graded calculus it matches an analogous functional constructed in terms of the connection. We study also the impact of the torsion on the other spectral fuct

George Stamatelis, George C. Alexandropoulos

This paper presents the Jump Markov Filtering Network (JMFNet), a novel model-based deep learning framework for real-time state-state estimation in jump Markov systems with unknown noise statistics and mode transition dynamics. A hybrid architecture comprising two Recurrent Neural Networks (RNNs) is proposed: one for mode prediction and another for filtering

Kelun Lei, Hailong Yang, Kaige Zhang, Kejie Ma

Sparse matrix-dense matrix multiplication (SpMM) is a critical kernel in both scientific computing and emerging graph learning workloads. The recent Armv9 architecture introduces Scalable Matrix Extension (SME), enabling tile-based matrix operations with high throughput. However, effectively exploiting both SME and traditional SIMD resources for unstructured

Esha Gupta, Yu Zhou

For $d\geq 1$, we define semibricks and wide subcategories in the $d$-extended hearts of bounded $t$-structures on a triangulated category. We show that these semibricks are in bijection with finite-length wide subcategories. When the $d$-extended heart is the $d$-extended module category $d\mbox{-}\mathrm{mod}\Lambda$ of a finite-dimensional algebra $\Lambd

Chenying Liu, Wei Huang, Xiao Xiang Zhu

Land Use and Land Cover (LULC) mapping is a fundamental task in Earth Observation (EO). However, current LULC models are typically developed for a specific modality and a fixed class taxonomy, limiting their generability and broader applicability. Recent advances in foundation models (FMs) offer promising opportunities for building universal models. Yet, tas

Abhijay Ghildyal, Rajesh Sureddi, Nabajeet Barman, Saman Zadtootaghaj

Evaluating the perceptual quality of Novel View Synthesis (NVS) images remains a key challenge, particularly in the absence of pixel-aligned ground truth references. Full-Reference Image Quality Assessment (FR-IQA) methods fail under misalignment, while No-Reference (NR-IQA) methods struggle with generalization. In this work, we introduce a Non-Aligned Refer

Roman Cherniha

The Comments are devoted to the recently published paper 'Modelling and nonclassical symmetry analysis of a complex porous media flow in a dilating channel' (Physica D. 481 (2025) 134834), in which a model describing an unsteady two-dimensional viscous incompressible fluid flow through a porous medium is studied. The main theoretical results of that study co

Jun Sun, Xinxin Zhang, Simin Hong, Jian Zhu

Multimodal learning, while contributing to numerous success stories across various fields, faces the challenge of prohibitively expensive manual annotation. To address the scarcity of annotated data, a popular solution is unsupervised domain adaptation, which has been extensively studied in unimodal settings yet remains less explored in multimodal settings.

Xuchen Li, Ruitao Wu, Xuanbo Liu, Xukai Wang

Recent advances in large language models have enabled AI systems to achieve expert-level performance on domain-specific scientific tasks, yet these systems remain narrow and handcrafted. We introduce SciAgent, a unified multi-agent system designed for generalistic scientific reasoning-the ability to adapt reasoning strategies across disciplines and difficult

Xinpeng Zhao, Zhaochun Ren, Yukun Zhao, Zhenyang Li

Generative retrieval (GR) reframes document retrieval as an end-to-end task of generating sequential document identifiers (DocIDs). Existing GR methods predominantly rely on left-to-right auto-regressive decoding, which suffers from two fundamental limitations: (i) a \emph{mismatch between DocID generation and natural language generation}, whereby an incorre

Helen Christodoulidi, Sergej Flach

We assess the ergodic properties of the Fermi-Pasta-Ulam-Tsingou-$\alpha$ model for generic initial conditions using a Toda integral. It serves as an adiabatic invariant for the system and a suitable observable to measure its equilibrium time. Over this timescale, the onset of action diffusion results in ergodic temporal fluctuations. We compare this timesca

Andrija Stanisic, Stefan Nastic

Integration of edge, cloud and space devices into a unified 3D continuum imposes significant challenges for client selection in federated learning systems. Traditional approaches rely on continuous monitoring and historical data collection, which becomes impractical in dynamic environments where satellites and mobile devices frequently change operational con

R. Marqués-Gómez, J. Melchor, A. C. Moreno Maldonado, C. Marquina

We report experimental evidence of mechanical wave generation by magnetic nanoparticles (MNPs) under radiofrequency electromagnetic fields (EMFs, 65mT, 800 kHz, at 100ms bursts) in the ultrasonic (US) frequency range. We developed an experimental setup to detect the second harmonic (SH) signal under isothermal conditions. This SH-US signal was observed only

Kunal Kingkar Das, Manoj Balaji Jagadeeshan, Nallani Chakravartula Sahith, Jivnesh Sandhan

Large Language Models (LLMs) are increasingly treated as universal, general-purpose solutions across NLP tasks, particularly in English. But does this assumption hold for low-resource, morphologically rich languages such as Sanskrit? We address this question by comparing instruction-tuned and in-context-prompted LLMs with smaller task-specific encoder-decode

Paula Gómez-Álvarez, Miguel J. Torrejón, Jesús Algaba, Felipe J. Blas

We investigate the three-phase coexistence line of ethane (C$_2$H$_6$) hydrate through molecular dynamics simulations using the direct coexistence approach. In this framework, C$_2$H$_6$ sI hydrate, aqueous, and pure guest phases are constructed within a single simulation box, allowing us to monitor their mutual stability. From the temporal evolution of the

Qiankun Pi, Yepeng Sun, Jicang Lu, Qinlong Fan

Large Language Models (LLMs) have demonstrated their remarkable capabilities in document understanding. However, recent research reveals that LLMs still exhibit performance gaps in Document-level Relation Extraction (DocRE) as requiring fine-grained comprehension. The commonly adopted "extract entities then predict relations" paradigm in LLM-based methods le