October 2025 arXiv papers — page 143

Stavros Orfanoudakis, Frans A. Oliehoek, Peter Palensky, Pedro P. Vergara

Electric Vehicles (EVs) offer substantial flexibility for grid services, yet large-scale, uncoordinated charging can threaten voltage stability in distribution networks. Existing Reinforcement Learning (RL) approaches for smart charging often disregard physical grid constraints or have limited performance for complex large-scale tasks, limiting their scalabi

Rui Hu, Yu Chen, Longbo Huang

Many popular practical reinforcement learning (RL) algorithms employ evolving reward functions-through techniques such as reward shaping, entropy regularization, or curriculum learning-yet their theoretical foundations remain underdeveloped. This paper provides the first finite-time convergence analysis of a single-timescale actor-critic algorithm in the pre

Yingjie Ma, Jing Guo, Alexis B. Dubs, Krystian K. Ganko

N-glycosylation is a critical quality attribute of monoclonal antibodies (mAbs), the dominant class of biopharmaceuticals. Controlling glycosylation remains difficult due to intrinsic pathway complexity, limited online measurements, and a lack of tailored control strategies. This work applies an adaptive nonlinear model predictive control (ANMPC) framework t

Mohammadreza Kasaei, Mostafa Ghobadi, Mohsen Khadem

This paper presents a shape-aware whole-body control framework for tendon-driven continuum robots with direct application to endoluminal surgical navigation. Endoluminal procedures, such as bronchoscopy, demand precise and safe navigation through tortuous, patient-specific anatomy where conventional tip-only control often leads to wall contact, tissue trauma

Emeric Bouin, Luca Ziviani

In this article, we prove some convergence results for kinetic Fokker-Planck equations with strong space confinement but fat-tailed local equilibria and non-explicit global steady states. We extend the results of \cite{C21} to a wider class of fat-tailed local equilibria, with rates of convergence in a large class of weighted $\sfL^1$ spaces. We complement o

Jeppe K. Christensen, Simon H. Albrechtsen, Christian E. Petersen, Constant A. Schouder

In 2023, ultrafast pump-probe spectrocopy was used to record the solvation dynamics of a single Na$^+$ ion in a liquid helium droplet, atom-by-atom and with femtosecond time resolution [Albrechtsen \textit{et al., Nature}, 2023, \textbf{623}, 319]. Subsequently, theoretical studies showed that other alkali ions solvate in a similar manner but no experimental

Sourav Mal, Nehad Ahmed, Junaid Jami, Subhankar Mishra

Efficient exploration of the vast chemical space is a fundamental challenge in materials design and discovery, particularly for designing functional inorganic crystalline materials with targeted properties. Diffusion-based generative models have emerged as a powerful route, but most existing approaches require domain-specific constraints and separate diffusi

Quentin Villegas, Laurence Denneulin, Simon Prunet, André Ferrari

In this paper, we propose an approach combining diffusion models and inverse problems for the reconstruction of circumstellar disk images. Our method builds upon the Rhapsodie framework for polarimetric imaging, substituting its classical prior with a diffusion model trained on synthetic data. Our formulation explicitly incorporates stellar leakage while eff

Ruolan Cheng, Yong Deng

Random permutation set (RPS) is a new formalism for reasoning with uncertainty involving order information. Measuring the conflict between two pieces of evidence represented by permutation mass functions remains an open issue in order-dependent uncertain information fusion. This paper analyzes conflicts in RPS from two different perspectives: random finite s

Benjamin Clavié, Sean Lee, Rikiya Takehi, Aamir Shakir

Multi-vector dense retrieval methods like ColBERT systematically use a single-layer linear projection to reduce the dimensionality of individual vectors. In this study, we explore the implications of the MaxSim operator on the gradient flows of the training of multi-vector models and show that such a simple linear projection has inherent, if non-critical, li

Guanxin Jiang, Andreas Brendel, Pablo M. Delgado, Jürgen Herre

This paper presents the Deep learning-based Perceptual Audio Quality metric (DeePAQ) for evaluating general audio quality. Our approach leverages metric learning together with the music foundation model MERT, guided by surrogate labels, to construct an embedding space that captures distortion intensity in general audio. To the best of our knowledge, DeePAQ i

Martin Ryzy, Guqi Yan, István Veres, Thomas Berer

Nanometric layer thickness imaging is crucial for fundamental research and characterization of micro fabricated devices. Here, we assess the potential of a non-contact opto-acoustic frequency domain laser ultrasound (FreDomLUS) microscopy technique for imaging nanometric thickness variations via GHz zero-group velocity (ZGV) elastic plate resonances. The met

Jie Yang, Chenyang Gu, Zixuan Liu

Multimodal recommender systems enhance personalized recommendations in e-commerce and online advertising by integrating visual, textual, and user-item interaction data. However, existing methods often overlook two critical biases: (i) modal confounding, where latent factors (e.g., brand style or product category) simultaneously drive multiple modalities and

Sacha Ikonicoff, Jean-Simon Pacaud Lemay, Tim Van der Linden

A tangent category is a category with an endofunctor, called the tangent bundle functor, which is equipped with various natural transformations that capture essential properties of the classical tangent bundle of smooth manifolds. In this paper, we show that, surprisingly, the category of groups is a tangent category whose tangent bundle functor is induced b

Zirui Guo, Xubin Ren, Lingrui Xu, Jiahao Zhang

Retrieval-Augmented Generation (RAG) has emerged as a fundamental paradigm for expanding Large Language Models beyond their static training limitations. However, a critical misalignment exists between current RAG capabilities and real-world information environments. Modern knowledge repositories are inherently multimodal, containing rich combinations of text

Bingrong Huang, Liangxun Li

In this paper, we give the upper bounds on the variance for cubic moment of Hecke--Maass cusp forms and Eisenstein series respectively. For the cusp form case, the bound comes from a large sieve inequality for symmetric cubes. We also give some nontrivial bounds for higher moments of symmetric cube $L$-functions. For the Eisenstein series case, the upper bou

Yue Zhang, Shanshan Luo, Zhi Geng, Yangbo He

Learning an optimal individualized treatment rule depends on reliable value comparisons across the candidate class. Standard doubly robust estimators are consistent when either the propensity score or outcome regression model is correctly specified, but they do not directly control the remaining bias in value estimation when both models are misspecified. In

Thayse A. Pacheco, Paula R. T. Coelho, Lucimara P. Martins, Ricardo P. Schiavon

Globular clusters (GCs) are fundamental for understanding the integrated light of old stellar populations and galaxy assembly processes. However, the role of hot, evolved stars, such as horizontal branch (HB), extreme HB, and blue stragglers, remains poorly constrained. These stars are often underrepresented or entirely excluded from stellar population model

Joaquín Pérez

This article explains a program to study complete and properly embedded minimal surfaces in $\mathbb{R}^3$ developed jointly with W.H. Meeks and A. Ros in the last three decades. It follows closely the structure of my invited ICM talk with the same title and supplies details and references to the original papers. After recalling the role of the classical Rie

Pål Forr Austnes, Matthieu Jacobs, Lu Wang, Mario Paolone

Since the 1990s, widespread introduction of central (wholesale) electricity markets has been seen across multiple continents, driven by the search for efficient operation of the power grid through competition. The increase of renewables has made significant impacts both on central electricity markets and distribution-level grids as renewable power generation

Masahiro Tsujimoto, Daiki Miura, Hiroya Yamaguchi, Ehud Behar

The advent of the X-ray microcalorimeter spectrometer Resolve onboard the XRISM space telescope opened a new era for high-resolution X-ray spectroscopy of astrophysical plasmas. Many spectral features were newly detected, including the K alpha and K beta inner-shell transition lines of mildly ionized (F- to Li-like) Fe at 6-8 keV in the spectra of X-ray bina

Greta Damo, Elena Cabrio, Serena Villata

Counter-speech generation is at the core of many expert activities, such as fact-checking and hate speech, to counter harmful content. Yet, existing work treats counter-speech generation as pure text generation task, mainly based on Large Language Models or NGO experts. These approaches show severe drawbacks due to the limited reliability and coherence in th

D. Mondal, W. Baker, M. Chaudhuri, J. B. Dent

We report on a search for coherent elastic neutrino--nucleus scattering (CE$\nu$NS) using cryogenic sapphire (Al$_2$O$_3$) detectors deployed at the Mitchell Institute Neutrino Experiment at Reactor (MINER), located near the 1~MW$_\text{th}$ TRIGA research reactor at Texas A\&M University. The experiment operated with a primary detector mass of 72~g and achi

Stephen F. Siegel, Ganesh Gopalakrishnan

This volume contains the proceedings of the Verification of Scientific Software (VSS 2025) workshop, held on 4 May 2025 at McMaster University, Canada, as part of ETAPS 2025. VSS brings together researchers in software verification and scientific computing to address challenges in ensuring the correctness and reliability of large-scale scientific codes. The

Anthony Kiely, Diana A. Chisholm, Akram Touil, Sebastian Deffner

We present a precise characterization of the onset of classicality that combines the formalism of quantum Darwinism with the tools from quantum metrology. We show that the quantum Fisher information provides a useful metric for assessing the rate at which classical objectivity emerges. Furthermore, our formalism allows us to explore how the choice of measure

Florent Delgrange, Raphael Avalos, Willem Röpke

Safe policy improvement (SPI) offers theoretical control over policy updates, yet existing guarantees largely concern offline, tabular reinforcement learning (RL). We study SPI in general online settings, when combined with world model and representation learning. We develop a theoretical framework showing that restricting policy updates to a well-defined ne

Nilo Schwencke, Cyriaque Rousselot, Alena Shilova, Cyril Furtlehner

Recent works have shown that natural gradient methods can significantly outperform standard optimizers when training physics-informed neural networks (PINNs). In this paper, we analyze the training dynamics of PINNs optimized with ANaGRAM, a natural-gradient-inspired approach employing singular value decomposition with cutoff regularization. Building on this

Andrea Tononi, Maciej Lewenstein, Luis Santos

A shell-shaped Bose-Einstein condensate released from its confinement expands radially both outwards and inwards, displaying a self-interference pattern characterized by a density peak surrounded by a halo. Here we analyze how an external imprinting or the thermal fluctuations of the condensate phase influence this expansion. In both cases, we find that the

Mohamed Omran, Farhad Zanjani, Davide Abati, Jens Petersen

This paper describes the Qualcomm AI Research solution to the RealADSim-NVS challenge, hosted at the RealADSim Workshop at ICCV 2025. The challenge concerns novel view synthesis in street scenes, and participants are required to generate, starting from car-centric frames captured during some training traversals, renders of the same urban environment as viewe

Maribel Fernández, Miguel Pagano, Nora Szasz, Álvaro Tasistro

We investigate an extension of nominal many-sorted signatures in which abstraction has a form of instantiation, called generalised concretion, as elimination operator (similarly to lambda-calculi). Expressions are then classified using a system of sorts and sort families that respects alpha-conversion (similarly to dependently-typed lambda-calculi) but not a

Thorsten Altenkirch, Nathaniel Burke, Philip Wadler

Defining substitution for a language with binders like the simply typed $\lambda$-calculus requires repetition, defining substitution and renaming separately. To verify the categorical properties of this calculus, we must repeat the same argument many times. We present a lightweight method that avoids repetition and that gives rise to a simply typed category

Ambrus Kaposi, Szumi Xie

Type theory can be described as a generalised algebraic theory. This automatically gives a notion of model and the existence of the syntax as the initial model, which is a quotient inductive-inductive type. Algebraic definitions of type theory include Ehrhard's definition of model, categories with families (CwFs), contextual categories, Awodey's natural mode

Zhibo Chen, Frank Pfenning

Logical Frameworks such as Automath [de Bruijn, 1968] or LF [Harper et al., 1993] were originally conceived as metalanguages for the specification of foundationally uncommitted deductive systems, yielding generic proof checkers. Their high level of abstraction was soon exploited to also express algorithms over deductive systems such as theorem provers, type-

Sebastián Urciuoli

We develop formal theories of conversion for Church-style lambda-terms with Pi-types in first-order syntax using one-sorted variables names and Stoughton's multiple substitutions. We then formalize the Pure Type Systems along some fundamental metatheoretic properties: weakening, syntactic validity, closure under alpha-conversion and substitution. Finally, we

Zhi Li, Yanan Wang, Hao Niu, Julio Vizcarra

Multimodal large language models have recently achieved remarkable progress in video question answering (VideoQA) by jointly processing visual, textual, and audio information. However, it remains unclear which video representations are most effective for MLLMs, and how different modalities balance task accuracy against computational efficiency. In this work,

Marek Chalupa, Thomas A. Henzinger, Ana Oliveira da Costa

Hypertrace logic is a sorted first-order logic with separate sorts for time and execution traces. Its formulas specify hyperproperties, which are properties relating multiple traces. In this work, we extend hypertrace logic by introducing trace quantifiers that range over the set of all possible traces. In this extended logic, formulas can quantify over two

Nathan Guermond, Gopalan Nadathur

The logic underlying the Abella proof assistant includes mechanisms for interpreting atomic predicates through fixed point definitions that can additionally be treated inductively or co-inductively. However, the original formulation of the logic includes a strict stratification condition on definitions that is too restrictive for some applications such as th

Gaku Kinoshita, Beatriz Sanchez-Cano, Yoshizumi Miyoshi, Laura Rodriguez-Garcia

Interplanetary coronal mass ejections (ICMEs) cause Forbush Decreases (FDs) effects, which are local decreases in background galactic cosmic rays (GCR). Even though FDs can be observed with simple particle instruments, their amplitude and shape provide physical profiles of passing ICMEs. However, in some cases, previous statistical studies of the heliocentri

Hung-Te Henry Su, Chih-Hsueh Lin, Po-Han Lee

Boron neutron capture therapy (BNCT) leverages the nuclear reaction between thermal neutrons and boron-10 (B-10) atoms to induce selective tumor cell death. The spatial and quantum mechanical overlap between the neutron wavefunction and B-10 nuclei encapsulated in carborane cages (C2B10H12) is fundamental to optimizing therapeutic efficacy. This study presen

Martin Ryzy, Guqi Yan, Clemens Grünsteidl, Georg Watzl

In inertial confinement fusion experiments hollow, spherical mm-sized capsules are used as a container for nuclear fuel. To achieve maximum implosion efficiency, a perfect capsule geometry is required. This paper presents a wall thickness measurement method based on zero-group velocity guided elastic wave resonances. They are measured with a non-destructive,

Roberto M. Amadio

The focus of these lecture notes is on abstract models and basic ideas and results that relate to the operational semantics of programming languages largely conceived. The approach is to start with an abstract description of the computation steps of programs and then to build on top semantic equivalences, specification languages, and static analyses. While o

Gergő Balogh, Dávid Kószó, Homayoun Safarpour Motealegh Mahalegi, László Tóth

Understanding how software developers think, make decisions, and behave remains a key challenge in software engineering (SE). Verbalization techniques (methods that capture spoken or written thought processes) offer a lightweight and accessible way to study these cognitive aspects. This paper presents a scoping review of research at the intersection of SE an

Fei Ren, Sifan Wang, Pei-Zhi Zhuang, Hai-Sui Yu

Conventional physics-informed extreme learning machine (PIELM) often faces challenges in solving partial differential equations (PDEs) involving high-frequency and variable-frequency behaviors. To address these challenges, we propose a general Fourier feature physics-informed extreme learning machine (GFF-PIELM). We demonstrate that directly concatenating mu

Min Yuan, Bo Nan Zhang, Yin Huang

Motivated by the interpretation of $X(3872)$ as a $D\bar{D}^{*}$ molecular state, heavy-quark spin symmetry predicts a spin-2 partner, $X_{2}(4013)$, which can be regarded as a $D^{*}\bar{D}^{*}$ molecule with quantum numbers $J^{PC} = 2^{++}$. Its experimental confirmation, however, remains elusive. In this work, we investigate the production mechanisms of

Tsung-En Lin, Kuan-Yi Lee, Hung-Yi Lee

Large Audio-Language Models (LALMs) excel in Audio QA but often suffer from hallucinations ungrounded in the audio. To our knowledge, we are the first to propose applying vector steering to the audio domain to mitigate this. Unlike text-based steering, our silence-anchored contrastive approach steers the model away from hallucinations by contrasting active a

Mingzhu Wang, Yun Shang

Quantum machine learning (QML) holds promise for computational advantage, yet progress on real-world tasks is hindered by classical preprocessing and noisy devices. We introduce ViT-QCNN-FT, a hybrid framework that integrates a fine-tuned Vision Transformer with a quantum convolutional neural network (QCNN) to compress high-dimensional images into features s

Minsung Kho, Norton Lee, Rak-Kyeong Seong

Brane tilings are bipartite periodic graphs on the 2-torus and realize a large family of 4d N=1 supersymmetric gauge theories corresponding to toric Calabi-Yau 3-folds. We present a complete classification of dimer integrable systems corresponding to the 30 brane tilings whose toric Calabi-Yau 3-folds are given by the 16 reflexive polygons in 2 dimensions. F

Tatsuru Kikuchi

This paper develops a nonparametric framework for identifying and estimating spatial boundaries of treatment effects in settings with geographic spillovers. While atmospheric dispersion theory predicts exponential decay of pollution under idealized assumptions, these assumptions -- steady winds, homogeneous atmospheres, flat terrain -- are systematically vio

Qi Zhang, Jia-Wei Ying, Shi-Pu Gu, Xing-Fu Wang

Quantum secret sharing (QSS) plays a critical role in building the distributed quantum networks. Device-independent (DI) QSS provides the highest security level for QSS. However, the photon transmission loss and extremely low multipartite entanglement generation rate largely limit DI QSS's secure photon transmission distance (less than 1 km) and practical ke

Sifan Li, Hongkai Chen, Yujun Cai, Qingwen Ye

Vision Language Models (VLMs) have achieved impressive progress in multimodal reasoning; yet, they remain vulnerable to hallucinations, where outputs are not grounded in visual evidence. In this paper, we investigate a previously overlooked setting: logo hallucination, where models generate brand names or textual content despite logos containing no visible w

Mahamodul Hasan Mahadi, Md. Nasif Safwan, Souhardo Rahman, Shahnaj Parvin

Developing AI systems capable of nuanced ethical reasoning is critical as they increasingly influence human decisions, yet existing models often rely on superficial correlations rather than principled moral understanding. This paper introduces Ethic-BERT, a BERT-based model for ethical content classification across four domains: Commonsense, Justice, Virtue,

Stefano Gagliani, Feliciano Giuseppe Pacifico, Lorenzo Chicchi, Duccio Fanelli

A general class of dynamical systems which can be trained to operate in classification and generation modes are introduced. A procedure is proposed to plant asymptotic stationary attractors of the deterministic model. Optimizing the dynamical system amounts to shaping the architecture of inter-nodes connection to steer the evolution towards the assigned equi

Zeyu Zhao, Ningtao Wang, Xing Fu, Yu Cheng

Encoder-only Transformers have advanced along three axes -- architecture, data, and systems -- yielding Pareto gains in accuracy, speed, and memory efficiency. Yet these improvements have not fully transferred to Chinese, where tokenization and morphology differ markedly from English. We introduce Chinese ModernBERT, a from-scratch Chinese encoder that coupl

Haizhong Li, Hiroshi Tamaru, Zeke Yao

In this paper, we study the Hopf hypersurfaces of the complex hyperbolic quadric $Q^{m*}=SO^o_{2,m}/(SO_2\times SO_m)$ ($m\geq3$) with constant principal curvatures. We classify the Hopf hypersurfaces of $Q^{m*}$ ($m\geq3$) with at most two distinct constant principal curvatures. For Hopf hypersurfaces with three or four distinct constant principal curvature

Jianfeng Dong, Lei Huang, Daizong Liu, Xianke Chen

Almost all previous text-to-video retrieval works ideally assume that videos are pre-trimmed with short durations containing solely text-related content. However, in practice, videos are typically untrimmed in long durations with much more complicated background content. Therefore, in this paper, we focus on the more practical yet challenging task of Partial

Ying A, Wenzhang Sun, Chang Zeng, Chunfeng Wang

Reconstructing dynamic 3D urban scenes is crucial for autonomous driving, yet current methods face a stark trade-off between fidelity and computational cost. This inefficiency stems from their semantically agnostic design, which allocates resources uniformly, treating static backgrounds and safety-critical objects with equal importance. To address this, we i

Ozan K. Tonguz, Federico Taschin

One of the major problems in Machine Learning (ML) and Artificial Intelligence (AI) is the fact that the probability distribution of the test data in the real world could deviate substantially from the probability distribution of the training data set. When this happens, the predictions of an ML system or an AI agent could involve large errors which is very

Xin Wei

In this paper, we introduce a new class of mappings, termed $(\rho,t)$-quasisymmetric mappings, which generalizes the classical concept of quasisymmetric mappings. Using this broader class of mappings, we provide an analytic characterization of $t$-quasicircles. This result can be viewed as a $t$-quasisymmetric analogue of a classical theorem by Tukia and V\

Yosuke Bando, Akinobu Mita, Kazuhiro Hiwada, Shintaro Sano

When key-value (KV) stores use SSDs for storing a large number of items, oftentimes they also require large in-memory data structures including indices and caches to be traversed to reduce IOs. This paper considers offloading most of such data structures from the costly host DRAM to secondary memory whose latency is in the microsecond range, an order of magn

Benedikt Böck, Amar Kasibovic, Wolfgang Utschick

Standardized (link-level) channel models such as the 3GPP TDL and CDL models are frequently used to evaluate machine learning (ML)-based physical-layer methods. However, in this work, we argue that a link-level perspective incorporates limiting assumptions, causing unwanted distributional shifts or necessitating impractical online training. An additional dra

Thomas Benz, Axel Vanoni, Michael Rogenmoser, Luca Benini

With the ever-growing heterogeneity in computing systems, driven by modern machine learning applications, pressure is increasing on memory systems to handle arbitrary and more demanding transfers efficiently. Descriptor-based direct memory access controllers (DMACs) allow such transfers to be executed by decoupling memory transfers from processing units. Cla

Luigi Foscari, Emanuele Guidotti, Nicolò Cesa-Bianchi, Tatjana Chavdarova

We study overpricing in a repeated game between two representative agents: a market maker, who controls market liquidity, and a market taker, who chooses trade quantities. Market prices evolve through the endogenous price impact of trades and exogenous shocks. We define overpricing relative to a counterfactual price path that holds fixed the same sequence of

Fuhao Li, Wenxuan Song, Han Zhao, Jingbo Wang

Vision-language-action (VLA) models have recently shown strong potential in enabling robots to follow language instructions and execute precise actions. However, most VLAs are built upon vision-language models pretrained solely on 2D data, which lack accurate spatial awareness and hinder their ability to operate in the 3D physical world. Existing solutions a

Youhao Si, Yuan Liao, Qiushi Han, Yuhang Yang

The rapid development of auditory attention decoding (AAD) based on electroencephalography (EEG) signals offers the possibility EEG-driven target speaker extraction. However, how to effectively utilize the target-speaker common information between EEG and speech remains an unresolved problem. In this paper, we propose a model for brain-controlled speaker ext

Hao Jiang, Meng Qin, Ruijie Kuai, Dandan Liang

With the rapid growth in computing power demand, cloud native networks have emerged as a promising solution to address the challenges of efficient resource coordination, particularly in coping with the dynamic fluctuations of network bandwidth in clusters. We propose Metronome, a network-aware and priority-aware scheduling mechanism for cloud native networks

Laurin Luttmann, Lin Xie

Self-improvement has emerged as a state-of-the-art paradigm in Neural Combinatorial Optimization (NCO), where models iteratively refine their policies by generating and imitating high-quality solutions. Despite strong empirical performance, existing methods face key limitations. Training is computationally expensive, as policy updates require sampling numero

Shihao Xia, Jingyi Chen, Jincan Chen, Shanhe Su

We establish a finite-time quantum tricycle driven by an external field and investigate its thermodynamic performance in the slow-driving regime. By developing a perturbative expansion of heat with respect to operation time, we capture the dynamics of heat exchange processes beyond the quasistatic limit. Within a geometric framework, we derive fundamental bo

Federico Gabriele, Aldo Glielmo, Marco Taboga

Current macroeconomic models with agent heterogeneity can be broadly divided into two main groups. Heterogeneous-agent general equilibrium (GE) models, such as those based on Heterogeneous Agent New Keynesian (HANK) or Krusell-Smith (KS) approaches, rely on GE and 'rational expectations', somewhat unrealistic assumptions that make the models very computation

Zhaoyang Shang, Sibo Wei, Jianbin Guo, Rui Zhou

Large Language Models (LLMs) excel in general tasks, but adapting them to specialized domains relies on high-quality supervised fine-tuning (SFT) data. Although existing methods can identify subsets of high-quality data and reduce training cost to some extent, their selection process still suffers from over-reliance on LLMs' internal knowledge, weak interpre

Kutay Bölat, Peter Palensky, Simon Tindemans

Accurate intraday forecasts are essential for power system operations, complementing day-ahead forecasts that gradually lose relevance as new information becomes available. This paper introduces a Bayesian updating mechanism that converts fully probabilistic day-ahead forecasts into intraday forecasts without retraining or re-inference. The approach conditio

Manon Lizzana, Fabien Malbet, Alain Leger, Fabrice Pancher

Many different scientific applications require sub-micro arcsecond precision astrometry, including researching rocky exoplanets in the vicinity of the Sun and studying dark matter. The Habitable Worlds Observatory (HWO) is a promising candidate to carry an astrometric instrument because it provides a stable, space-based telescope with a large aperture, which

Pedro Domingos

Progress in AI is hindered by the lack of a programming language with all the requisite features. Libraries like PyTorch and TensorFlow provide automatic differentiation and efficient GPU implementation, but are additions to Python, which was never intended for AI. Their lack of support for automated reasoning and knowledge acquisition has led to a long and

Jeanne Choi, Dasom Choi, Sejun Jeong, Hwajung Hong

People with visual impairments (PVI) use a variety of assistive technologies to navigate their daily lives, and conversational AI (CAI) tools are a growing part of this toolset. Much existing HCI research has focused on the technical capabilities of current CAI tools, but in this paper, we instead examine how PVI themselves envision potential futures for liv

Chenghanyu Zhang, Zekun Li, Peipei Li, Xing Cui

With the increasing integration of Multimodal Large Language Models (MLLMs) into the medical field, comprehensive evaluation of their performance in various medical domains becomes critical. However, existing benchmarks primarily assess general medical tasks, inadequately capturing performance in nuanced areas like the spine, which relies heavily on visual i

Ziyi Han, Huanyu Wang, Zeyu Zhang, Xiangxiang Dai

Low-Rank Adaptation (LoRA) has emerged as a widely used technique for adapting large language models (LLMs) to new domains, due to its modular design and broad availability on platforms such as HuggingFace. This availability has motivated efforts to reuse existing LoRAs for domain generalization. However, existing methods often rely on explicit task labels o

Sami Khairy, Gabriel Mittag, Vishak Gopal, Ross Cutler

The quality of experience (QoE) delivered by video conferencing systems is significantly influenced by accurately estimating the time-varying available bandwidth between the sender and receiver. Bandwidth estimation for real-time communications remains an open challenge due to rapidly evolving network architectures, increasingly complex protocol stacks, and

Deyu Zou, Yongqiang Chen, Jianxiang Wang, Haochen Yang

Active reasoning requires large language model (LLM) agents to interact with external sources and strategically gather information to solve problems in multiple turns. Central to this process is belief tracking: maintaining an accurate representation of the underlying state and uncertainty in understanding and solving the problem. However, due to limited rea

K. Aditya, Sandeep Kataria

We investigate the stability of Milky Way analogs (MWAs) in the \texttt{TNG50} simulation against the growth of local axisymmetric instabilities, tracing their evolution from cosmic noon ($z=2.5$) to the present day ($z=0$). Using a two-component stability criterion that accounts for stars, gas, and the force field of the dark matter halo, we compute the net

Abhimanyu Gupta, Myung Hwan Seo

We develop a class of optimal tests for a structural break occurring at an unknown date in infinite and growing-order time series regression models, such as AR($\infty$), linear regression with increasingly many covariates, and nonparametric regression. Under an auxiliary i.i.d. Gaussian error assumption, we derive an average power optimal test, establishing

Mikko Korhonen

Let $r$ be an odd prime and $\mathbb{F}$ a field containing a primitive $r$th root of unity. Then for all $\ell \geq 1$, there is a faithful representation $f: \operatorname{Sp}_{2\ell}(r) \rightarrow \operatorname{GL}_{r^\ell}(\mathbb{F})$ called the Weil representation. We provide explicit matrices generating $\operatorname{Sp}_{2\ell}(r)$ in $\operatornam

Xiaopeng Liu, Yupei Lin, Sen Zhang, Xiao Wang

Visible-infrared image fusion is crucial in key applications such as autonomous driving and nighttime surveillance. Its main goal is to integrate multimodal information to produce enhanced images that are better suited for downstream tasks. Although deep learning based fusion methods have made significant progress, mainstream unsupervised approaches still fa

Jinlun Ye, Zhuohao Sun, Yiqiao Qiu, Qiu Li

Out-of-distribution (OOD) detection is crucial when deploying deep neural networks in the real world to ensure the reliability and safety of their applications. One main challenge in OOD detection is that neural network models often produce overconfident predictions on OOD data. While some methods using auxiliary OOD datasets or generating fake OOD images ha

Yuto Yokoi, Kazuhiro Hotta

We propose two novel loss functions, Multiplicative Loss and Confidence-Adaptive Multiplicative Loss, for semantic segmentation in medical and cellular images. Although Cross Entropy and Dice Loss are widely used, their additive combination is sensitive to hyperparameters and often performs suboptimally, especially with limited data. Medical images suffer fr

Marco Calzà, Massimiliano Rinaldi, Sunny Vagnozzi

In curved space-time, a scalar field $\phi$ is generically expected to couple to curvature, via a coupling of the form $\xi\phi^2R$. Yet in the study of Hawking emission from regular black holes (RBHs), where scalar fields are often introduced as simple probes of the geometry, and the Ricci scalar is generically non-zero, this non-minimal coupling is almost

Ye Chen, Liming Tan, Yupeng Zhu, Yuanbin Wang

Current video representations heavily rely on unstable and over-grained priors for motion and appearance modelling, \emph{i.e.}, pixel-level matching and tracking. A tracking error of just a few pixels would lead to the collapse of the visual object representation, not to mention occlusions and large motion frequently occurring in videos. To overcome the abo

Blazej Manczak, Eric Lin, Francisco Eiras, James O' Neill

Large language models (LLMs) are rapidly transitioning into medical clinical use, yet their reliability under realistic, multi-turn interactions remains poorly understood. Existing evaluation frameworks typically assess single-turn question answering under idealized conditions, overlooking the complexities of medical consultations where conflicting input, mi

Ningxin He, Yang Liu, Wei Sun, Xiaozhou Ye

Text-to-Image (T2I) models have demonstrated their versatility in a wide range of applications. However, adaptation of T2I models to specialized tasks is often limited by the availability of task-specific data due to privacy concerns. On the other hand, harnessing the power of rich multimodal data from modern mobile systems and IoT infrastructures presents a

Changfu Xu, Jianxiong Guo, Yuzhu Liang, Haiyang Huang

Diffusion Models (DMs), as a leading class of generative models, offer key advantages for reinforcement learning (RL), including multi-modal expressiveness, stable training, and trajectory-level planning. This survey delivers a comprehensive and up-to-date synthesis of diffusion-based RL. We first provide an overview of RL, highlighting its challenges, and t

Yuqi Jia, Yupei Liu, Zedian Shao, Jinyuan Jia

Prompt injection attacks deceive a large language model into completing an attacker-specified task instead of its intended task by contaminating its input data with an injected prompt, which consists of injected instruction(s) and data. Localizing the injected prompt within contaminated data is crucial for post-attack forensic analysis and data recovery. Des

Jiakai Li, Rongzheng Wang, Yizhuo Ma, Shuang Liang

While large language models (LLMs) show considerable promise across various fields, they have notable limitations in handling multi-document question answering (Multi-doc QA) tasks. The first challenge is long-range dependency modeling, where LLMs struggle to focus on key information in long texts, which weakens important semantic connections. Second, most L

Yue Wang, Xiao-Ming Zhang, Xiao Yuan, Qi Zhao

While the preparation of a general quantum state is challenging, realistic problem instances, such as those encountered in quantum chemistry and quantum machine learning-typically exhibit hierarchical amplitude structures, consisting of a small number of large components alongside a vast number of small but non-negligible ones. Standard approaches determinis

Jingyi Wang, Hongyuan Zhu, Ye Niu, Yunhui Deng

Large Language Models (LLMs) have demonstrated profound impact on Natural Language Processing (NLP) tasks. However, their effective deployment across diverse domains often require domain-specific adaptation strategies, as generic models may underperform when faced with specialized data distributions. Recent advances in prompt engineering (PE) offer a promisi

Tao Yin, Xiaohong Zhang, Jiacheng Zhang, Li Huang

Effectively integrating molecular graph structures with Large Language Models (LLMs) is a key challenge in drug discovery. Most existing multi-modal alignment methods typically process these structures by fine-tuning the LLM or adding a static adapter simultaneously. However, these approaches have two main limitations: (1) it optimizes a shared parameter spa

Matthew S. Scott

Qualification conditions (also termed constraint qualifications) help avoid pathological behavior at domain boundaries in convex analysis. By generalizing facial reduction from conic programming to general convex programs of the form $f(x) + g(Ax)$, we provide qualification-free generalizations of several key results: an exact Fenchel-Rockafellar dual, KKT o

Jeanne Choi, DongJae Kang, Yubin Choi, Juhoon Lee

As social media adoption grows globally, online problematic behaviors increasingly escalate into large-scale crises, requiring an evolving set of mitigation strategies. While HCI research often analyzes problematic behaviors with pieces of user-generated content as the unit of analysis, less attention has been given to event-focused perspectives that track h

Håkon R. Fredheim, Simen Kvaal

A mathematical framework for reduced density matrix functional theory (RDMFT) is proposed. The work is inspired by and generalizes the work by E.H.~Lieb [E.H. Lieb, Int. J. Quant. Chem. 24(1983), pp.243--277] on density-functional theory (DFT). We introduce a Banach space for density matrices with finite kinetic energy. The dual space is a rich class of sing

Yuehui Li, Yahao Lu, Haoyuan Wu, Sen Zhang

In the multimedia domain, Infrared Small Target Detection (ISTD) plays a important role in drone-based multi-modality sensing. To address the dual challenges of cross-domain shift and heteroscedastic noise perturbations in ISTD, we propose a doubly wavelet-guided Invariance learning framework(Ivan-ISTD). In the first stage, we generate training samples align

Sanju S Pillai, M Muhsin, M Sahoo

We investigate the dynamics of an inertial active Ornstein-Uhlenbeck particle suspended in a non-Markovian environment. The particle is additionally subjected to external forces, such as harmonic confinement and a magnetic field. Motivated by the importance of understanding the non-Markovian behavior of complex environments, we examine the impact of a viscoe

Loïc Foissy, Yunzhou Xie, Dawei Zhang, Yi Zhang

The concept of weighted infinitesimal bialgebras provides an algebraic framework for understanding the non-homogeneous associative Yang-Baxter equation. In this paper, we endow the space of decorated planar rooted forests with a two-parameters family of coproducts, making it into a weighted infinitesimal bialgebra. A combinatorial characterization of the cop

Boyang Zhang, Zhiguo Wang, Ya-Feng Liu

Chance constrained programming (CCP) is a powerful framework for addressing optimization problems under uncertainty. In this paper, we introduce a novel Gradient-Guided Diffusion-based Optimization framework, termed GGDOpt, which tackles CCP through three key innovations. First, GGDOpt accommodates a broad class of CCP problems without requiring the knowledg