May 2025 arXiv papers — page 21

Bei-Lei Liu, Yue-Chao Wang, Yuan-Ji Xu, Xingyu Gao

The pressure-driven insulator-metal transition is a crucial topic in condensed matter physics. However, even for the prototypical strongly correlated system, NiO, the critical pressure for transition remains debated. In this work, we evaluated the electronic interactions over a wide range of pressures based on our developed doubly-screened Coulomb correction

Zi-An Wang, Shihao Zou, Shiyao Yu, Mingyuan Zhang

Recent advances in interactive technologies have highlighted the prominence of audio signals for semantic encoding. This paper explores a new task, where audio signals are used as conditioning inputs to generate motions that align with the semantics of the audio. Unlike text-based interactions, audio provides a more natural and intuitive communication method

S. Bogdanov, E. Manuylovich, S. K. Turitsyn

Nonlinear systems, transforming an input signal into a high-dimensional output feature space, can be used for non-conventional computing. This approach, however, requires a change of system parameters during training rather than coefficients in a software program. We propose here to use available off-the-shelf high-speed optical communication devices and tec

Han Zhou, Sebastian G. Gruber, Teodora Popordanoska, Matthew B. Blaschko

Several variants of reweighted risk functionals, such as focal loss, inverse focal loss, and the Area Under the Risk Coverage Curve (AURC), have been proposed for improving model calibration; yet their theoretical connections to calibration errors remain under-explored. In this paper, we revisit a broad class of weighted risk functions and find a principled

Runyi Li, Bin Chen, Jian Zhang, Radu Timofte

Blind face restoration from low-quality images is a challenging task that requires not only high-fidelity image reconstruction, but also preservation of facial identity. Although diffusion models like Stable Diffusion have shown promise in generating high-quality images, their VAE modules are typically trained on broad natural-image data dominated by HQ cont

Chuanyuan Tan, Wenbiao Shao, Hao Xiong, Tong Zhu

Handling unanswerable questions (UAQ) is crucial for LLMs, as it helps prevent misleading responses in complex situations. While previous studies have built several datasets to assess LLMs' performance on UAQ, these datasets lack factual knowledge support, which limits the evaluation of LLMs' ability to utilize their factual knowledge when handling UAQ. To a

Safwan Labbi, Paul Mangold, Daniil Tiapkin, Eric Moulines

We provide global convergence rates for vanilla and entropy-regularized federated softmax stochastic policy gradient (FedPG) with local training. We show that FedPG converges to a near-optimal policy in terms of the average agent value, with a gap controlled by the level of heterogeneity. Remarkably, we obtain the first convergence rates for entropy-regulari

Kevin Frans, Seohong Park, Pieter Abbeel, Sergey Levine

At the core of reinforcement learning is the idea of learning beyond the performance in the data. However, scaling such systems has proven notoriously tricky. In contrast, techniques from generative modeling have proven remarkably scalable and are simple to train. In this work, we combine these strengths, by deriving a direct relation between policy improvem

Simón Martínez-Rozas, David Alejo, José Javier Carpio, Fernando Caballero

Unmanned Aerial Vehicles (UAVs) have become essential tools in inspection and emergency response operations due to their high maneuverability and ability to access hard-to-reach areas. However, their limited battery life significantly restricts their use in long-duration missions. This paper presents a tethered marsupial robotic system composed of a UAV and

Paul Dupuis, Benjamin J. Zhang

We introduce and develop a novel particle exchange Monte Carlo method. Whereas existing methods apply to eigenfunction problems where the eigenvalue is known (e.g., integrals with respect to a Gibbs measure, which can be interpreted as corresponding to eigenvalue zero), here the focus is on problems where the eigenvalue is not known a priori. To obtain an ap

Narmin Nasibova, Xerxes D. Arsiwalla

Within the framework of the thermal soft-wall model of AdS/QCD, we investigate phenomenological properties of pions at finite temperature. This includes the electromagnetic (EM) form factor $F_{\pi}(Q^{2}, T)$, the thermal mass $M_{\pi}(T)$, charge radius $r_{\pi}(T)$, the generalized parton distribution (GPD) $H_{\pi}(x, Q^{2}, T)$, the charge density $\rho

Yanbin Wang, Xingyu Chen, Yumiao Wang, Xiang Wang

We propose the LCB-CV-UNet to tackle performance degradation caused by High Dynamic Range (HDR) radar signals. Initially, a hardware-efficient, plug-and-play module named Logarithmic Connect Block (LCB) is proposed as a phase coherence preserving solution to address the inherent challenges in handling HDR features. Then, we propose the Dual Hybrid Dataset Co

Ahmed Y. Radwan, Mustafa Yildirim, Navid Hasanzadeh, Hina Tabassum

Wi-Fi technology has evolved from simple communication routers to sensing devices. Wi-Fi sensing leverages conventional Wi-Fi transmissions to extract and analyze channel state information (CSI) for applications like proximity detection, occupancy detection, activity recognition, and health monitoring. By leveraging existing infrastructure, Wi-Fi sensing off

Ying Liu, Rui Zhang, Wen-Quan Yang, Ya-Feng Jiao

We propose a theoretical scheme to enhance the sensitivity of a quantum optomechanical gyroscope (QOMG) by optical Kerr effect. We utilize quantum Fisher information (QFI) to evaluate the metrological potential of the QOMG scheme. It is found that the Kerr interaction can significantly enhances the sensitivity of the QOMG. We observe the super-Hesenberg scal

Quim Motger, Marc Oriol, Max Tiessler, Xavier Franch

Opinion mining plays a vital role in analysing user feedback and extracting insights from textual data. While most research focuses on sentiment polarity (e.g., positive, negative, neutral), fine-grained emotion classification in app reviews remains underexplored. Fine-grained emotion classification is thus needed to better understand users' affective respon

Shuzhou Sun, Li Liu, Tianpeng Liu, Shuaifeng Zhi

Existing two-stage Scene Graph Generation (SGG) frameworks typically incorporate a detector to extract relationship features and a classifier to categorize these relationships; therefore, the training paradigm follows a causal chain structure, where the detector's inputs determine the classifier's inputs, which in turn influence the final predictions. Howeve

Arne Tillmann

Multi-agent large language models (MA-LLMs) are a rapidly growing research area that leverages multiple interacting language agents to tackle complex tasks, outperforming single-agent large language models. This literature review synthesizes the latest research on agent profiles, communication structures, and decision-making processes, drawing insights from

Zhejian Yang, Yongchao Chen, Xueyang Zhou, Jiangyue Yan

Long-horizon robotic manipulation poses significant challenges for autonomous systems, requiring extended reasoning, precise execution, and robust error recovery across complex sequential tasks. Current approaches, whether based on static planning or end-to-end visuomotor policies, suffer from error accumulation and lack effective verification mechanisms dur

Fanxiao Li, Jiaying Wu, Canyuan He, Wei Zhou

Multimodal large language models (MLLMs) have demonstrated impressive capabilities in visual reasoning and text generation. While previous studies have explored the application of MLLM for detecting out-of-context (OOC) misinformation, our empirical analysis reveals two persisting challenges of this paradigm. Evaluating the representative GPT-4o model on dir

Dota Tianai Dong, Yifan Luo, Po-Ya Angela Wang, Asli Ozyurek

Multimodal language models (MLMs) increasingly demonstrate human-like communication, yet their use of everyday perspectival words remains poorly understood. To address this gap, we compare humans and MLMs in their use of three word types that impose increasing cognitive demands: vocabulary (for example, "boat" or "cup"), possessives (for example, "mine" vers

Pirzada Suhail, Rehna Afroz, Gouranga Bala, Amit Sethi

Out-of-distribution (OOD) detection and uncertainty estimation (UE) are critical components for building safe machine learning systems, especially in real-world scenarios where unexpected inputs are inevitable. However the two problems have, until recently, separately been addressed. In this work, we propose a novel framework that combines network inversion

Jiayi Zeng, Yizhe Feng, Mengliang He, Wenhui Lei

Large language models (LLMs) have demonstrated significant advancements in error handling. Current error-handling works are performed in a passive manner, with explicit error-handling instructions. However, in real-world scenarios, explicit error-handling instructions are usually unavailable. In this paper, our work identifies this challenge as how to conduc

Sara Johansson Fernstad, Sarah Alsufyani, Silvia Del Din, Alison Yarnall

This paper contributes a set of quality metrics for identification and visual analysis of structured missingness in high-dimensional data. Missing values in data are a frequent challenge in most data generating domains and may cause a range of analysis issues. Structural missingness in data may indicate issues in data collection and pre-processing, but may a

Yiping Meng, Chulin Jiang, Courtney Jayne Scurr, Farzad Pour Rahimian

Engineered timber is pivotal to low-carbon construction, but moisture uptake during its service life can compromise structural reliability and impede reuse within a circular economy model. Despite growing interest, quantitative standards for classifying the reusability of moisture-exposed timber are still lacking. This study develops a probabilistic framewor

Bálint Soczó, Ildikó Pethes

Describing the interactions of water molecules is one of the most common, yet critical, tasks in molecular dynamics simulations. Because of its unique properties, hundreds of attempts have been made to construct an ideal interaction potential model for water. In various studies, the models have been evaluated based on their ability to reproduce different pro

Nasir Kenarangui, Walter C. Daugherity, Arthur Powalka, Laszlo B. Kish

Instantaneous Noise-Based Logic (INBL) represents a computational paradigm that offers a deterministic alternative to quantum computing, potentially challenging the notion of quantum supremacy without relying on quantum hardware. INBL encodes logical information in orthogonal stochastic processes ("noise-bits") and exploits their superpositions and nonlinear

Adrien Majka, El-Mahdi El-Mhamdi

Goodhart's law is a famous adage in policy-making that states that ``When a measure becomes a target, it ceases to be a good measure''. As machine learning models and the optimisation capacity to train them grow, growing empirical evidence reinforced the belief in the validity of this law without however being formalised. Recently, a few attempts were made t

Runmin Jiang, Genpei Zhang, Yuntian Yang, Siqi Wu

Single-particle cryo-electron microscopy (cryo-EM) has become a cornerstone of structural biology, enabling near-atomic resolution analysis of macromolecules through advanced computational methods. However, the development of cryo-EM processing tools is constrained by the scarcity of high-quality annotated datasets. Synthetic data generation offers a promisi

Benyamin Trachtenberg, Nir Rosenfeld

In strategic classification, the standard supervised learning setting is extended to support the notion of strategic user behavior in the form of costly feature manipulations made in response to a classifier. While standard learning supports a broad range of model classes, the study of strategic classification has, so far, been dedicated mostly to linear cla

Linyu Li, Zhi Jin, Yuanpeng He, Dongming Jin

Knowledge graph completion (KGC) has attracted considerable attention in recent years because it is critical to improving the quality of knowledge graphs. Researchers have continuously explored various models. However, most previous efforts have neglected to take advantage of regularization from a deeper perspective and therefore have not been used to their

Jiaqi Chen, Yufei Shan, Yinghui Ye

Volume comparison theorem is a type of fundamental results in Riemannian geometry. In this article, we extend the volume comparison result in \cite{Besse2008} to the comparison of total $\sigma_l$-curvature with respect to $\sigma_k$-curvature ($l<k$). In particular, we prove the comparison holds for metrics close to strictly stable positive Einstein metric

Ben Li, Minqi Li, Jie Ren, Kaibing Zhang

Image-based virtual try-on aims to fit a target garment to a specific person image and has attracted extensive research attention because of its huge application potential in the e-commerce and fashion industries. To generate high-quality try-on results, accurately warping the clothing item to fit the human body plays a significant role, as slight misalignme

Lingyan Ran, Yali Li, Tao Zhuo, Shizhou Zhang

In semi-supervised semantic segmentation (SSSS), data augmentation plays a crucial role in the weak-to-strong consistency regularization framework, as it enhances diversity and improves model generalization. Recent strong augmentation methods have primarily focused on intensity-based perturbations, which have minimal impact on the semantic masks. In contrast

Aladin Djuhera, Swanand Ravindra Kadhe, Farhan Ahmed, Syed Zawad

Fine-tuning large language models (LLMs) on telecom datasets is a common practice to adapt general-purpose models to the telecom domain. However, little attention has been paid to how this process may compromise model safety. Recent research has shown that even benign fine-tuning can degrade the safety alignment of LLMs, causing them to respond to harmful or

Daniel Jarne Ornia, Nicholas Bishop, Joel Dyer, Wei-Chen Lee

Advanced reasoning models with agentic capabilities (AI agents) are deployed to interact with humans and to solve sequential decision-making problems under (approximate) utility functions and internal models. When such problems have resource or failure constraints where action sequences may be forcibly terminated once resources are exhausted, agents face imp

Alireza Zabihi, Luis Badesa, Araceli Hernandez

Voltage unbalance, caused by variations in voltage magnitude and phase angle, is a significant power quality issue in three-phase systems, leading to equipment inefficiencies and increased system losses. The integration of distributed energy resources (DER) into the grid adds complexity, as DER can either reduce or worsen voltage unbalance, depending on fact

Tianhang Wang, Fan Lu, Sanqing Qu, Guo Yu

Existing neural rendering-based urban scene reconstruction methods mainly focus on the Interpolated View Synthesis (IVS) setting that synthesizes from views close to training camera trajectory. However, IVS can not guarantee the on-par performance of the novel view outside the training camera distribution (\textit{e.g.}, looking left, right, or downwards), w

Jian Yao, Ran Cheng, Xingyu Wu, Jibin Wu

The reasoning capabilities of large language models (LLMs) have advanced rapidly, particularly following the release of DeepSeek R1, which has inspired a surge of research into data quality and reinforcement learning (RL) algorithms. Despite the pivotal role diversity plays in RL, its influence on LLM reasoning remains largely underexplored. To bridge this g

L. Elisa Celis, Lingxiao Huang, Nisheeth K. Vishnoi

The rapid rise of Generative AI (GenAI) tools has sparked debate over their role in complementing or replacing human workers across job contexts. We present a mathematical framework that models jobs, workers, and worker-job fit, introducing a novel decomposition of skills into decision-level and action-level subskills to reflect the complementary strengths o

Amer Krivošija, Alexander Munteanu, André Nusser, Chris Schwiegelshohn

This paper introduces $k$-Dynamic Time Warping ($k$-DTW), a novel dissimilarity measure for polygonal curves. $k$-DTW has stronger metric properties than Dynamic Time Warping (DTW) and is more robust to outliers than the Fr\'{e}chet distance, which are the two gold standards of dissimilarity measures for polygonal curves. We show interesting properties of $k

Vinay Sharma, Sergei I Chernyshenko

With an increasing interest in the design of long and complex space missions, the search for orbits that require the least amount of fuel is of fundamental interest. This paper develops existing computational models for locating Unstable Periodic Orbits (UPOs) in polynomial dynamical systems using Sum-of-Squares (SOS) optimization technique and proposes a nu

Laura Grigori, Muhammad Hassan

The density matrix renormalization group (DMRG) algorithm is a popular alternating minimization scheme for solving high-dimensional optimization problems in the tensor train format. Classical DMRG, however, is based on sequential minimization, which raises challenges in its implementation on parallel computing architectures. To overcome this, we propose a no

Monika Gahalawat, Maneesh Bilalpur, Raul Fernandez Rojas, Jeffrey F. Cohn

Depression is a debilitating mood disorder negatively impacting millions worldwide. While researchers have explored multiple verbal and non-verbal behavioural cues for automated depression assessment, head motion has received little attention thus far. Further, the common practice of validating machine learning models via a single dataset can limit model gen

Yinuo Wang, Likun Wang, Mining Tan, Wenjun Zou

Due to their expressive capacity, diffusion models have shown great promise in offline RL and imitation learning. Diffusion Actor-Critic with Entropy Regulator (DACER) extended this capability to online RL by using the reverse diffusion process as a policy approximator, achieving state-of-the-art performance. However, it still suffers from a core trade-off:

Cen Mo, Liang Li

Fast and accurate muon reconstruction is crucial for neutrino telescopes to improve experimental sensitivity and enable online triggering. This paper introduces a hybrid-graph neural network (GNN) method tailored for efficient muon track reconstruction, leveraging the robustness of GNNs, alongside traditional physics-based approaches. The "light GNN model" a

A. N. Azarevich, N. B. Bolotina, O. N. Khrykina, A. V. Bogach

A review of long-term studies of ZrB$_{12}$ and LuB$_{12}$ superconductors with very similar conduction bands and phonon spectra, but with radically different (by a factor of 15-20) critical temperatures and magnetic fields is presented. A detailed analysis of well-known studies in combination with new results of structural, thermodynamic and charge transpor

Tianrui Dai, Elisa Francini, Sergio Vessella

We investigate the Strong Unique Continuation Property (SUCP) for elliptic equations with piecewise Lipschitz coefficients exhibiting jump discontinuities across a regular interface. We prove SUCP at the interface using a doubling inequality derived from a Carleman estimate with a singular weight. This result is intended as a first step toward solving the in

Tobias Lindenbauer, Georg Groh, Hinrich Schütze

We introduce CTIM-Rover, an AI agent for Software Engineering (SE) built on top of AutoCodeRover (Zhang et al., 2024) that extends agentic reasoning frameworks with an episodic memory, more specifically, a general and repository-level Cross-Task-Instance Memory (CTIM). While existing open-source SE agents mostly rely on ReAct (Yao et al., 2023b), Reflexion (

Qi Li, Runpeng Yu, Xinchao Wang

Multimodal large language models (MLLMs) demonstrate remarkable capabilities in handling complex multimodal tasks and are increasingly adopted in video understanding applications. However, their rapid advancement raises serious data privacy concerns, particularly given the potential inclusion of sensitive video content, such as personal recordings and survei

Zheming Zhang, Yan Jiang, Qingshan Li, Ai Han

In China's competitive fresh e-commerce market, optimizing operational strategies, especially inventory management in front-end warehouses, is key to enhance customer satisfaction and to gain a competitive edge. Front-end warehouses are placed in residential areas to ensure the timely delivery of fresh goods and are usually in small size. This brings the cha

Marco Gaido, Sara Papi, Luisa Bentivogli, Alessio Brutti

Training large-scale models presents challenges not only in terms of resource requirements but also in terms of their convergence. For this reason, the learning rate (LR) is often decreased when the size of a model is increased. Such a simple solution is not enough in the case of speech-to-text (S2T) trainings, where evolved and more complex variants of the

Linghao Zhang, Shilin He, Chaoyun Zhang, Yu Kang

The issue-resolving task, where a model generates patches to fix real-world bugs, has emerged as a critical benchmark for evaluating the capabilities of large language models (LLMs). While SWE-bench and its variants have become standard in this domain, they suffer from key limitations: they have not been updated since their initial releases, cover a narrow s

Vittorio D'Agostino, Eugenio Fasci, Muhammad Asad Khan, Stefania Gravina

We report on a comb-locked cavity ring-down spectrometer developed for high-precision molecular spectroscopy at 2 ${\mu}$m. It is based on the use of an external-cavity diode laser that is offset-frequency locked to the signal output of a singly-resonant optical parametric oscillator. This latter acts as reference laser, being locked to a self-referenced opt

Danilo Ribeiro, Thayssa Rocha, Gustavo Pinto, Bruno Cartaxo

Artificial Intelligence (AI) governance is the practice of establishing frameworks, policies, and procedures to ensure the responsible, ethical, and safe development and deployment of AI systems. Although AI governance is a core pillar of Responsible AI, current literature still lacks synthesis across such governance frameworks and practices. Objective: To i

Jang-Hyun Kim, Jinuk Kim, Sangwoo Kwon, Jae W. Lee

Transformer-based large language models (LLMs) cache context as key-value (KV) pairs during inference. As context length grows, KV cache sizes expand, leading to substantial memory overhead and increased attention latency. This paper introduces KVzip, a query-agnostic KV cache eviction method enabling effective reuse of compressed KV caches across diverse qu

Gaspard Oliviers, Mufeng Tang, Rafal Bogacz

Predictive coding (PC) is an influential computational model of visual learning and inference in the brain. Classical PC was proposed as a top-down generative model, where the brain actively predicts upcoming visual inputs, and inference minimises the prediction errors. Recent studies have also shown that PC can be formulated as a discriminative model, where

Luis A. Anchordoqui, Francis Halzen, Dieter Lust

The quantum gravity scale within the dark dimension scenario ($M_* \sim 10^{9}~{\rm GeV}$) roughly coincides with the energy scale of the KM3-230213A neutrino ($E_\nu \sim 10^{8}~{\rm GeV}$). We propose an interpretation for this intriguing coincidence in terms of Hawking evaporation of five-dimensional (5D) primordial black holes (PBHs). 5D PBHs are bigger,

Michael J. Keith, Renée Spiewak, Andrew G. Lyne, Patrick Weltevrede

Time-correlated variations in the pulse profiles of radio pulsars provide insights into changes in their magnetospheres. For a small number of pulsars (~20), these variations have been shown to correlate with spin-down rate. Many of these profile changes involve small (few percent) variations in the relative intensity of different profile components, and hen

Srishti Gupta, Daniele Angioni, Maura Pintor, Ambra Demontis

Class-incremental learning (CIL) poses significant challenges in open-world scenarios, where models must not only learn new classes over time without forgetting previous ones but also handle inputs from unknown classes that a closed-set model would misclassify. Recent works address both issues by (i)~training multi-head models using the task-incremental lear

Luke Qi, Ali Khalatpour, Jason Herrmann, Taewon Park

We present low-loss ($<1.5\%$) and power-efficient Mach-Zehnder interferometers (MZIs) on thin-film lithium niobate. To accurately measure low MZI losses, we develop a self-calibrated method using tunable Sagnac loop reflectors (SLRs) to build cavities. Fabry-P\'erot cavities constructed from these fabrication-tolerant SLRs achieve an intrinsic quality facto

Peter Samoaa, Marcus Vukojevic, Morteza Haghir Chehreghani, Antonio Longa

Graph-level regression underpins many real-world applications, yet public benchmarks remain heavily skewed toward molecular graphs and citation networks. This limited diversity hinders progress on models that must generalize across both homogeneous and heterogeneous graph structures. We introduce RelSC, a new graph-regression dataset built from program graph

Xuan Gong, Hanbo Huang, Shiyu Liang

Factual knowledge extraction aims to explicitly extract knowledge parameterized in pre-trained language models for application in downstream tasks. While prior work has been investigating the impact of supervised fine-tuning data on the factuality of large language models (LLMs), its mechanism remains poorly understood. We revisit this impact through systema

Benno Bock, Simon Ohler, Michael Fleischhauer

We derive a parton mean-field Hamiltonian for Rydberg excitations on a honeycomb lattice with nearest and density-dependent, complex next-nearest neighbor hopping. Numerical results obtained from exact diagonalization of small systems have given indications for a ground state that is a chiral spin liquid (CSL) [Phys.Rev.Res. 5, 013157 (2023)]. Here we provid

Siying Xu, Marcel Früh, Kerstin Hammernik, Andreas Lingg

We propose a self-supervised feature learning assisted reconstruction (SSFL-Recon) framework for MRI reconstruction to address the limitation of existing supervised learning methods. Although recent deep learning-based methods have shown promising performance in MRI reconstruction, most require fully-sampled images for supervised learning, which is challengi

Eda Vurgun, Domingo García-Senz, Manuel Linares, K. Yavuz Eksi

The stability of an accretion disc surrounding a millisecond pulsar is analysed from an energetic point of view, using magnetohydrodynamic simulations that consider realistic disc structures and a variety of magnetic field inclination angles. The time-averaged components of the magnetic field interact with the disc through ohmic dissipation, which causes hea

Binyamin Manela, Sharon Gannot, Ethan Fetyaya

Visual dubbing, the synchronization of facial movements with new speech, is crucial for making content accessible across different languages, enabling broader global reach. However, current methods face significant limitations. Existing approaches often generate talking faces, hindering seamless integration into original scenes, or employ inpainting techniqu

Sapolnach Prompiengchai, Charith Narreddy, Steve Joordens

Formative assessment is a cornerstone of effective teaching and learning, providing students with feedback to guide their learning. While there has been an exponential growth in the application of generative AI in scaling various aspects of formative assessment, ranging from automatic question generation to intelligent tutoring systems and personalized feedb

Mingyu Yu, Wei Wang, Yanjie Wei, Sujuan Qin

Recent advancements in adversarial jailbreak attacks have exposed critical vulnerabilities in Large Language Models (LLMs), enabling the circumvention of alignment safeguards through increasingly sophisticated prompt manipulations. Our experiments find that the effectiveness of jailbreak strategies is influenced by the comprehension ability of the target LLM

Deng Lin-fang, Zhang He-yao, Long Chao-yun

We extend the Banados-Silk-West effect to the static spherically symmetric black hole and rotating black hole in gravity's rainbow. Through systematic investigation that the effects of different rainbow functions on the center-of-mass energy of two test particles colliding outside the event horizon, we discussed the possibility of infinite center-of-mass ene

Hichem Hajaiej, Jun Wang, Zhaoyang Yin

In this paper, we develop a novel idea to study $y$-dependence for the logarithmic Schr\"odinger equation on $\mathbb{R}^d \times \mathbb{T}^n$. Unlike \cite{STNT2014}(Analysis \& PDE, 2014) and \cite{HHYL2024}(SIAM J. Math. Anal., 2024), the heart of the matter is that the scaling argument is invalid. Moreover, we also consider the Cauchy problem, which tra

G. Lusetti, M. Brüggen, H. W. Edler, F. de Gasperin

The galaxy cluster CIZA J2242.8+5301 is a well-studied merging galaxy cluster that hosts prominent double radio relics including the famous sausage relic, as well as other diffuse radio sources. Observations at frequencies below 100 MHz are essential for investigating the physics of radio relics as they provide unique access to the low-energy population of c

Marta Bílková, Wesley Fussner, Roman Kuznets

We define a new type of proof formalism for multi-agent modal logics with S5-type modalities. This novel formalism combines the features of hypersequents to represent S5 modalities with nested sequents to represent the T-like modality alternations. We show that the calculus is sound and complete, cut-free, and terminating and yields decidability and the fini

Sanggyun Ma, Wonjoon Choi, Jihun Park, Jaeyeul Kim

We present Bridging Geometric and Semantic (BriGeS), an effective method that fuses geometric and semantic information within foundation models to enhance Monocular Depth Estimation (MDE). Central to BriGeS is the Bridging Gate, which integrates the complementary strengths of depth and segmentation foundation models. This integration is further refined by ou

Jusheng Zhang, Yijia Fan, Wenjun Lin, Ruiqi Chen

We propose GAM-Agent, a game-theoretic multi-agent framework for enhancing vision-language reasoning. Unlike prior single-agent or monolithic models, GAM-Agent formulates the reasoning process as a non-zero-sum game between base agents--each specializing in visual perception subtasks--and a critical agent that verifies logic consistency and factual correctne

Marianne Bauer, William Bialek, Chase Goddard, Caroline M. Holmes

Many biological systems perform close to their physical limits, but promoting this optimality to a general principle seems to require implausibly fine tuning of parameters. Using examples from a wide range of systems, we show that this intuition is wrong. Near an optimum, functional performance depends on parameters in a "sloppy'' way, with some combinations

Xingguang Wei, Haomin Wang, Shenglong Ye, Ruifeng Luo

We study the task of panoptic symbol spotting, which involves identifying both individual instances of countable things and the semantic regions of uncountable stuff in computer-aided design (CAD) drawings composed of vector graphical primitives. Existing methods typically rely on image rasterization, graph construction, or point-based representation, but th

Longzhen Han, Awes Mubarak, Almas Baimagambetov, Nikolaos Polatidis

Multimodal Generative Models (MGMs) have rapidly evolved beyond text generation, now spanning diverse output modalities including images, music, video, human motion, and 3D objects, by integrating language with other sensory modalities under unified architectures. This survey categorises six primary generative modalities and examines how foundational techniq

Zhenjun Dong, Zhiwen Zhou, Yong Zeng

This paper proposes a novel multi-antenna architecture, termed ray antenna array (RAA), which practically enables flexible beamforming and also enhances wireless communication performance for high frequency systems in a cost-effective manner. RAA consists of a large number of inexpensive antenna elements and a few radio frequency (RF) chains. These antenna e

Hongrui Peng, Haolang Lu, Yuanlong Yu, Weiye Fu

Knowledge graphs (KGs) are ubiquitous in numerous real-world applications, and watermarking facilitates protecting intellectual property and preventing potential harm from AI-generated content. Existing watermarking methods mainly focus on static plain text or image data, while they can hardly be applied to dynamic graphs due to spatial and temporal variatio

Enzo Cerullo, Klaus Linde, Haley E. Jones, Efthymia Derezea

Standard (network) meta-analysis methods for medical test accuracy evaluation analyse the data separately for each test threshold - wasting data - unless every study reports all thresholds. Previously proposed "multiple threshold" models either fail to provide threshold-specific summary estimates, or they assume that ordinal tests (e.g., questionnaires) are

Yun-Cheng Tsai

Purpose: Accurate wound segmentation is essential for automated DESIGN-R scoring. However, existing models such as FUSegNet, which are trained primarily on foot ulcer datasets, often fail to generalize to wounds on other body sites. Methods: We propose an annotation-efficient pipeline that combines a lightweight YOLOv11n-based detector with the pre-trained F

Niklas Knobel

We study the Sobolev stability thresholds of 2d dissipative fluid equations around Couette flow on the domain $\mathbb T\times \mathbb R$. We prove a bound for general nonlinear interactions, which, for several fluid equations, reduces the proof of nonlinear stability to a linear stability analysis. We apply this approach to the examples of Navier-Stokes, Bo

M. Gorgone, G. Inferrera

In this paper, non-variational systems of differential equations containing small terms are considered, and a consistent approach for deriving approximate conservation laws through the introduction of approximate Lagrange multipliers is developed. The proposed formulation of the approximate direct method starts by assuming the Lagrange multipliers to be depe

Ivana Nikoloska, Hamdi Joudeh, Ruud van Sloun, Osvaldo Simeone

Quantum sensing exploits non-classical effects to overcome limitations of classical sensors, with applications ranging from gravitational-wave detection to nanoscale imaging. However, practical quantum sensors built on noisy intermediate-scale quantum (NISQ) devices face significant noise and sampling constraints, and current variational quantum sensing (VQS

Francesco Benini, Ohad Mamroud, Tomas Reis, Marco Serone

We study the $O(2N)$ symmetric Gross-Neveu model at finite density in the presence of a $U(1)$ chemical potential $h$ for a generic number $a \leq N-2$ of fermion fields. By combining perturbative quantum field theory, semiclassical large $N$, and Bethe ansatz techniques, we show that at finite $N$ two new dynamically generated scales $\Lambda_\mathrm{n}$ an

Mingzhe Du, Luu Anh Tuan, Yue Liu, Yuhao Qing

Large Language Models (LLMs) generate functionally correct solutions but often fall short in code efficiency, a critical bottleneck for real-world deployment. In this paper, we introduce a novel test-time iterative optimization framework to address this, employing a closed-loop system where LLMs iteratively refine code based on empirical performance feedback

Han Bao, Qinying Wang, Zhi Chen, Qingming Li

Not Safe/Suitable for Work (NSFW) content is rampant on social networks and poses serious harm to citizens, especially minors. Current detection methods mainly rely on deep learning-based image recognition and classification. However, NSFW images are now presented in increasingly sophisticated ways, often using image details and complex semantics to obscure

Konstantin Baune

We show that building blocks for open- and closed-string amplitudes on AdS are generated by the Drinfeld and Deligne associator, respectively. Our formalism lifts the known associator recursions for flat-space string amplitudes to the AdS picture. This delivers another proof that the AdS building blocks admit low-energy expansions with (single-valued) multip

Sahar Diskin, Michael Krivelevich, Itay Markbreit, Maksim Zhukovskii

We show that there exist constants $\delta_1,\delta_2>0$ such that if $G$ is an $(n,d,\lambda)$-graph with $\lambda/d\le\delta_1$, then $G$ contains an induced cycle of length at least $\delta_2n/d$. We further demonstrate that, up to a constant factor, this is best possible. Utilising our techniques, we derive that the number of non-isomorphic induced subgr

Ahmad Anaqreh, Shih-Kai Chou, Blaž Bertalanič, Mihael Mohorčič

Modeling propagation is the cornerstone for designing and optimizing next-generation wireless systems, with a particular emphasis on 5G and beyond era. Traditional modeling methods have long relied on statistic-based techniques to characterize propagation behavior across different environments. With the expansion of wireless communication systems, there is a

Anam Hashmi, Julia Dietlmeier, Kathleen M. Curran, Noel E. O'Connor

Attention is a fundamental component of the human visual recognition system. The inclusion of attention in a convolutional neural network amplifies relevant visual features and suppresses the less important ones. Integrating attention mechanisms into convolutional neural networks enhances model performance and interpretability. Spatial and channel attention

Mai Yashiki

The tension between the Hubble constant ($H_0$) inferred from the cosmic microwave background (CMB) and that measured from late-time observations, such as the local distance ladder, is a major challenge in modern cosmology. Early dark energy (EDE) has been proposed as a possible resolution to the $H_0$ tension, but it typically worsens the $S_8$ tension by e

Bowen Ping, Minnan Luo, Zhuohang Dang, Chenxi Wang

Geometry problem solving presents distinctive challenges in artificial intelligence, requiring exceptional multimodal comprehension and rigorous mathematical reasoning capabilities. Existing approaches typically fall into two categories: neural-based and symbolic-based methods, both of which exhibit limitations in reliability and interpretability. To address

Weijia Mao, Zhenheng Yang, Mike Zheng Shou

Unified multimodal large language models such as Show-o and Janus have achieved strong performance across both generation and understanding tasks. However, these models typically rely on large-scale datasets and require substantial computation during the pretraining stage. In addition, several post-training methods have been proposed, but they often depend o

Yao Guo, Yang Ai, Rui-Chen Zheng, Hui-Peng Du

This paper proposes a novel vision-integrated neural speech codec (VNSC), which aims to enhance speech coding quality by leveraging visual modality information. In VNSC, the image analysis-synthesis module extracts visual features from lip images, while the feature fusion module facilitates interaction between the image analysis-synthesis module and the spee

Shanon J. Rubin

One generally expects that the techniques of arboreal singularities and gluing of local differential graded categories will result in a useful global invariant for all Weinstein manifolds. In this paper we construct explicit models for the homotopy limits of diagrams of microlocal sheaf categories which arise from Weinstein surfaces with arboreal skeleta. Th

Khattiya Pongsirijinda, Zhiqiang Cao, Billy Pik Lik Lau, Ran Liu

Collaborative multiple robots for unknown environment exploration have become mainstream due to their remarkable performance and efficiency. However, most existing methods assume perfect robots' communication during exploration, which is unattainable in real-world settings. Though there have been recent works aiming to tackle communication-constrained situat

Mikkel Abrahamsen, Florestan Brunck, Jacobus Conradi, Benedikt Kolbe

We present the winning implementation of the Seventh Computational Geometry Challenge (CG:SHOP 2025). The task in this challenge was to find non-obtuse triangulations for given planar regions, respecting a given set of constraints consisting of extra vertices and edges that must be part of the triangulation. The goal was to minimize the number of introduced

R. Arcidiacono, G. Bardelli, M. Bartolini, M. Boscardin

In the past 10 years, two design innovations, the introduction of low internal gain (LGAD) and of resistive read-out (RSD), have radically changed the performance of silicon detectors. The LGAD mechanism, increasing the signal-to-noise ratio by about a factor of 20, leads to improved time resolution (typically 30 ps for a 50-$\mu$m thick sensor), while resis

Tangyou Huang, Jing-Jun Zhu, Zhong-Yi Ni

Hybrid quantum-classical algorithms hold great promise for solving quantum control problems on near-term quantum computers. In this work, we employ the hybrid framework that integrates digital quantum simulation with classical optimization to achieve optimal engineering of quantum many-body systems. To evaluate the overall performance of this method, we intr