May 2025 arXiv papers — page 87

Valentin Schmuker, Alex Hoi Hang Chan, Bastian Goldluecke, Urs Waldmann

In this paper, we present a texture-independent approach to estimate and track 3D joint positions of multiple pigeons. For this purpose, we build upon the existing 3D-MuPPET framework, which estimates and tracks the 3D poses of up to 10 pigeons using a multi-view camera setup. We extend this framework by using a segmentation method that generates silhouettes

Carla Pracias, Maycol Falla Luza

In this paper, we study homogeneous convex foliations on the complex projective plane $\mathbb{P}^2$. A foliation is called convex if all of its leaves, except straight lines, have no inflection points, and such foliations form a Zariski closed subset in the space of degree $d$ foliations on $\mathbb{P}^2$. Using projective duality, every foliation can be as

Jonghwi Kim, Deokhyung Kang, Seonjeong Hwang, Yunsu Kim

Despite bilingual speakers frequently using mixed-language queries in web searches, Information Retrieval (IR) research on them remains scarce. To address this, we introduce MiLQ, Mixed-Language Query test set, the first public benchmark of mixed-language queries, qualified as realistic and relatively preferred. Experiments show that multilingual IR models p

Sushant Gautam, Cise Midoglu, Vajira Thambawita, Michael A. Riegler

The integration of artificial intelligence in sports analytics has transformed soccer video understanding, enabling real-time, automated insights into complex game dynamics. Traditional approaches rely on isolated data streams, limiting their effectiveness in capturing the full context of a match. To address this, we introduce SoccerChat, a multimodal conver

Wen Chen, Jinjin Liang, Erxiao Wang

We develop the basic and new tools for classifying non-side-to-side tilings of the sphere by congruent triangles. Then we prove that, if the triangle has any irrational angle in degree, such tilings are: a sequence of 1-parameter families of triangles each admitting many 2-layer earth map tilings with $2n$($n\geq3$) tiles, together with rotational modificati

Sandra Müller, Bartosz Wcisło

We describe a construction of a model of second order arithmetic in which (boldface) $\bm{\Pi^1_n}$-determinacy holds, but (lightface) $\Pi^1_{n+2}$-$\mathsf{DC}$ fails, thus showing that no projective level of determinacy implies full $\mathsf{DC}_{\mathbb{R}}$. The construction builds upon the work of Gitman, Friedman, and Kanovei.

Vivek Kumar, Abhishek Jangid, Manas Sharma, Manvi Verma

Two-dimensional magnetic materials (2D-MM) are an exciting playground for fundamental research, and for spintronics and quantum sensing. However, their large-grain large-area synthesis using scalable vapour deposition methods is still an unsolved challenge. Here, we develop a tailored approach for centimetre-scale growth of semiconducting 2D-MM CrCl3 films o

Gregorio García-Valladares, Antonio Prados, Alessandro Manacorda, Carlos A. Plata

We investigate the search of a target with a given spatial distribution in a finite one-dimensional domain. The searcher follows Brownian dynamics and is always reset to its initial position when reaching the boundaries of the domain (boundary resetting). In addition, the searcher may be reset to its initial position from any internal point of the domain (bu

Luyang Cao, Jianwei Li, Yinghuan Shi

Semi-supervised medical image segmentation (SSMIS) leverages unlabeled data to reduce reliance on manually annotated images. However, current SOTA approaches predominantly focus on foreground-oriented modeling (i.e., segmenting only the foreground region) and have largely overlooked the potential benefits of explicitly modeling the background region. Our stu

Francesco Dalla Serra, Patrick Schrempf, Chaoyang Wang, Zaiqiao Meng

We present a novel approach to Chest X-ray (CXR) Visual Question Answering (VQA), addressing both single-image image-difference questions. Single-image questions focus on abnormalities within a specific CXR ("What abnormalities are seen in image X?"), while image-difference questions compare two longitudinal CXRs acquired at different time points ("What are

Pavel Kurasov, Omer Farooq, Michał Ławniczak, Szymon Bauch

A concept of germ graphs and the M-function formalism are employed to construct large families of isospectral and isoscattering graphs. This approach represents a complete departure from the original approach pioneered by Sunada, where isospectral graphs are obtained as quotients of a certain large symmetric graph. Using the M-function formalism and the symm

Benjamin Herdeanu, Juan Nathaniel, Carla Roesch, Jatan Buch

Causal discovery for dynamical systems poses a major challenge in fields where active interventions are infeasible. Most methods used to investigate these systems and their associated benchmarks are tailored to deterministic, low-dimensional and weakly nonlinear time-series data. To address these limitations, we present CausalDynamics, a large-scale benchmar

Gavin Farrell, Eleni Adamidi, Rafael Andrade Buono, Mihail Anton

Artificial intelligence (AI) has recently seen transformative breakthroughs in the life sciences, expanding possibilities for researchers to interpret biological information at an unprecedented capacity, with novel applications and advances being made almost daily. In order to maximise return on the growing investments in AI-based life science research and a

Anthony Leverrier

While 2-level systems, aka qubits, are a natural choice to perform a logical quantum computation, the situation is less clear at the physical level. Encoding information in higher-dimensional physical systems can indeed provide a first level of redundancy and error correction that simplifies the overall fault-tolerant architecture. A challenge then is to ens

Peter Petrov, Mikhail Shkolnikov

In this letter, we establish a general fact about the convergence of images of families of closed analytic surfaces in the special linear group $\operatorname{SL}_2(\mathbb{C})$ under the quotient by its maximal compact subgroup $\operatorname{SU}(2)$ subject to a contracting scaling sequence.

Javier Perez, Dimme de Groot, Jorge Martinez

Objective speech quality measures are widely used to assess the performance of video conferencing platforms and telecommunication systems. They predict human-rated speech quality and are crucial for assessing the systems quality of experience. Despite the widespread use, the quality measures are developed on a limited set of languages. This can be problemati

Kacper Prech, Joël Aschwanden, Patrick P. Potts

The laws of thermodynamics are a cornerstone for describing nanoscale and open quantum systems. However, formulating these laws for systems under continuous feedback control and under experimentally relevant conditions is challenging. In this work, we lay out a formalism for the laws of thermodynamics in an open quantum system under continuous measurement an

J Cameron Patterson, William J Buchanan, Callum Turino

The emergence of quantum computing and Shor's algorithm necessitates an imminent shift from current public key cryptography techniques to post-quantum robust techniques. NIST has responded by standardising Post-Quantum Cryptography (PQC) algorithms, with ML-KEM (FIPS-203) slated to replace ECDH (Elliptic Curve Diffie-Hellman) for key exchange. A key practica

Federica Spoto, Francesca Dominici, Tarik Benmarhnia, Danielle Braun

There is extensive evidence that long-term exposure to all-source PM2.5 increases mortality. However, to date, no study has evaluated whether this effect is exacerbated in the presence of wildfire events. Here, we study 60+ million older US adults and find that wildfire events increase the harmful effects of long-term all-source PM2.5 exposure on mortality,

Daniel C. Cabra, Gerardo L. Rossini

Multiferroic materials exhibit the coexistence of magnetic and electric order. They are at the forefront of modern condensed matter physics due to their potential applications in next-generation technologies such as data storage, sensors, and actuators. Despite significant progress, understanding and optimizing the coupling mechanisms between electric polari

Jing Ye, Lu Xiang, Yaping Zhang, Chengqing Zong

Effective emotional support hinges on understanding users' emotions and needs to provide meaningful comfort during multi-turn interactions. Large Language Models (LLMs) show great potential for expressing empathy; however, they often deliver generic and one-size-fits-all responses that fail to address users' specific needs. To tackle this issue, we propose a

Huacen Wang, Jiarui Zou, Zeju Zheng, Hongqiang Wang

Electrostatic adhesion is widely used in mobile robotics, haptics, and robotic end effectors for its adaptability to diverse substrates and low energy consumption. Force sensing is important for feedback control, interaction, and monitoring in the EA system. However, EA force monitoring often relies on bulky and expensive sensors, increasing the complexity a

Jing Ouyang, Yunxiao Chen, Chengcheng Li, Gongjun Xu

International Large-scale Assessments (ILSAs), such as the Program for International Student Assessment (PISA) and the Trends in International Mathematics and Science Study (TIMSS), are cornerstone tools for global educational research and policy-making. By benchmarking educational quality and performance trends, these assessments enable countries to evaluat

Yuzhu Wang, Archontis Politis, Konstantinos Drossos, Tuomas Virtanen

This paper addresses the problem of single-channel speech separation, where the number of speakers is unknown, and each speaker may speak multiple utterances. We propose a speech separation model that simultaneously performs separation, dynamically estimates the number of speakers, and detects individual speaker activities by integrating an attractor module.

Ana Carolina Mançur

We verify that LA-Courant algebroids provide the Manin triple framework for double Lie bialgebroids. Specifically, we establish a correspondence between double Lie bialgebroids and LA-Manin triples, i.e., LA-Courant algebroids equipped with a pair of complementary LA-Dirac structures. As an application, LA-Courant algebroids and CA-groupoids given by Drinfel

Nora Brambilla, Panayiotis Panayiotou, Saga Säppi, Antonio Vairo

The physics of quarkonium created in heavy-ion collisions is intrinsically connected to the correlation functions of adjoint chromoelectric fields in quantum chromodynamics. We study such correlation functions in a weak-coupling expansion in a thermal medium. We identify three distinct gauge-invariant correlators, and evaluate them to next-to-leading order.

Nora Brambilla, Saumen Datta, Marc Janer, Viljami Leino

We perform a lattice calculation of the correlators of two chromoelectric fields in the adjoint representation connected by adjoint Wilson lines at non-zero temperature. These correlators arise in the study of quarkonium dynamics and of adjoint heavy quark diffusion in deconfined matter. We work in SU(3) gauge theory using either gradient flow or multi-level

Bohan Zhou, Yi Zhan, Zhongbin Zhang, Zongqing Lu

Egocentric hand-object motion generation is crucial for immersive AR/VR and robotic imitation but remains challenging due to unstable viewpoints, self-occlusions, perspective distortion, and noisy ego-motion. Existing methods rely on predefined 3D object priors, limiting generalization to novel objects, which restricts their generalizability to novel objects

V. I. Yukalov

Major issues arising in the theory of Bose-Einstein condensation are reviewed. These issues, although being principally important, are very often misunderstood, which results in wrong conclusions. The basic point is global gauge symmetry breaking that is a necessary and sufficient condition for Bose-Einstein condensation. Paying no attention to this basic po

Yao Huang, Si-Yuan Cao, Yaqing Ding, Hao Yin

Planar homography, with eight degrees of freedom (DOFs), is fundamental in numerous computer vision tasks. While the positional offsets of four corners are widely adopted (especially in neural network predictions), this parameterization lacks geometric interpretability and typically requires solving a linear system to compute the homography matrix. This pape

Chuhongxu Chen, Ziwei Wang, Guilin Chen, Zhijia Zhang

With the advancement of nanoscience, silver/silver chloride (Ag/AgCl) electrodes have become widely utilised in microscale and nanoscale fluidic experiments, because of their stability. However, our findings reveal that the dissolution of AgCl from the electrode in \ch{Cl-}-rich solutions can lead to significant silver contamination, through the formation of

Wilbert Peter Empleo, Yitaek Kim, Hansoul Kim, Thiusius Rajeeth Savarimuthu

Robot-Assisted Minimally Invasive Surgery is currently fully manually controlled by a trained surgeon. Automating this has great potential for alleviating issues, e.g., physical strain, highly repetitive tasks, and shortages of trained surgeons. For these reasons, recent works have utilized Artificial Intelligence methods, which show promising adaptability.

Jia Li, Chao Xia

In this paper, we prove that a complete, two-sided, stable anisotropic minimal immersed hypersurface in $\mathbb{R}^{5}$ or $\mathbb{R}^{6}$ is flat, provided the anisotropic area functional is $C^4$-close to the area functional.

Vignesh Gopinathan, Urs Zimmermann, Michael Arnold, Matthias Rottmann

Video captioning models have seen notable advancements in recent years, especially with regard to their ability to capture temporal information. While many research efforts have focused on architectural advancements, such as temporal attention mechanisms, there remains a notable gap in understanding how models capture and utilize temporal semantics for effec

Md Nurul Molla, Joydwip Singh

In this paper, we study the boundedness of Bochner-Riesz commutator $$[b, S^{\alpha}(\mathcal{L})](f) = b S^{\alpha}(\mathcal{L})(f) - S^{\alpha}(\mathcal{L})(bf)$$ of a $BMO^{\varrho}(\mathbb{R}^d)$ function $b$ and the Bochner-Riesz operator $S^{\alpha}(\mathcal{L})$ associated to the Grushin operator $\mathcal{L}$ on $\mathbb{R}^d$ with $d:= d_1 +d_2$. We

Isabella Kemajou-Brown, Romario Gildas Foko Tiomela, Olawale Nasiru Lawal, Samson Adekola Alagbe

This study explores the application of Pontryagin's Maximum Principle to derive optimal strategies for controlling the spread of COVID-19, leveraging a novel compartmental model to capture the disease dynamics. We prioritize three key criteria: cost, effectiveness, and feasibility, each examined independently to evaluate their unique contributions to pandemi

Shijia Zhou, Siyao Peng, Simon M. Luebke, Jörg Haßler

Media framing refers to the emphasis on specific aspects of perceived reality to shape how an issue is defined and understood. Its primary purpose is to shape public perceptions often in alignment with the authors' opinions and stances. However, the interaction between stance and media frame remains largely unexplored. In this work, we apply an interdiscipli

Bowen Jiang, Runchuan Zhu, Jiang Wu, Zinco Jiang

We introduce KoLasSimpleQA, the first benchmark evaluating the multilingual factual ability of Large Language Models (LLMs). Inspired by existing research, we created the question set with features such as single knowledge point coverage, absolute objectivity, unique answers, and temporal stability. These questions enable efficient evaluation using the LLM-a

Renyi Zhong, Yichen Li, Guangba Yu, Wenwei Gu

Developers use logging statements to create logs that document system behavior and aid in software maintenance. As such, high-quality logging is essential for effective maintenance; however, manual logging often leads to errors and inconsistency. Recent methods emphasize using large language models (LLMs) for automated logging statement generation, but these

Martino Garonzi, Andrea Lucchini, Nowras Otmen

Given a profinite group $G$ and a family $\mathcal{F}$ of finite groups closed under taking subgroups, direct products and quotients, denote by $\mathcal{F}(G)$ the set of elements $g \in G$ such that $\{x \in G\ |\ \langle g,x \rangle \ \mbox{is a pro-}\mathcal{F} \mbox{ group}\}$ has positive Haar measure. We investigate the properties of $\mathcal{F}(G)$

Michael Weiler

The computation of magnitudes and distances from low signal-to-noise observations is known to be problematic, in the sense that the magnitudes and distances tend to assume extreme values, or are even undefined or unphysical in the case of negative observed fluxes or parallaxes. In this work we show that magnitudes can be computed consistently at all signal-t

Yongqi Zhao, Ji Zhou, Dong Bi, Tomislav Mihalj

The safety and reliability of Automated Driving Systems (ADSs) must be validated prior to large-scale deployment. Among existing validation approaches, scenario-based testing has been regarded as a promising method to improve testing efficiency and reduce associated costs. Recently, the emergence of Large Language Models (LLMs) has introduced new opportuniti

A. A. Rota, N. van der Marel, A. Garufi, C. Carrasco-González

Magnetothermal (MHD) winds and jets originate in a wide range of regions of protoplanetary disks (1-30 au) and are thought to be the primary mechanisms driving accretion onto the central star. One indirect signature of these processes is the free-free emission from ionized gas close to the star. We analyze a sample of 31 Class II disks: 18 full disks (FD) an

Weizhe Lin, Xing Li, Zhiyuan Yang, Xiaojin Fu

Large Reasoning Models (LRMs) demonstrate exceptional capability in tackling complex mathematical, logical, and coding tasks by leveraging extended Chain-of-Thought (CoT) reasoning. Test-time scaling methods, such as prolonging CoT with explicit token-level exploration, can push LRMs' accuracy boundaries, but they incur significant decoding overhead. A key i

Sky Cao, Ron Nissim, Scott Sheffield

We extend the parameter regimes for which area law is proven for pure $\mathrm{U}(N)$ lattice Yang-Mills theories, in particular when $N$ is large. This improves on a classical result of Osterwalder-Seiler from 1978. To do so, we view the master loop equation as a linear inhomogeneous equation for Wilson string expectations, and then prove an a priori bound

Shpresim Sadiku, Kartikeya Chitranshi, Hiroshi Kera, Sebastian Pokutta

Plausible counterfactual explanations (p-CFEs) are perturbations that minimally modify inputs to change classifier decisions while remaining plausible under the data distribution. In this study, we demonstrate that classifiers can be trained on p-CFEs labeled with induced \emph{incorrect} target classes to classify unperturbed inputs with the original labels

Jianbiao Mei, Tao Hu, Daocheng Fu, Licheng Wen

Large Language Models (LLMs), despite their advancements, are fundamentally limited by their static parametric knowledge, hindering performance on tasks requiring open-domain up-to-date information. While enabling LLMs to interact with external knowledge environments is a promising solution, current efforts primarily address closed-end problems. Open-ended q

Max Weltevrede, Moritz A. Zanger, Matthijs T. J. Spaan, Wendelin Böhmer

In the zero-shot policy transfer setting in reinforcement learning, the goal is to train an agent on a fixed set of training environments so that it can generalise to similar, but unseen, testing environments. Previous work has shown that policy distillation after training can sometimes produce a policy that outperforms the original in the testing environmen

Yuki K. Wakabayashi, Akihira Munakata, Yoshitaka Taniyasu, Masaki Kobayashi

SrTiO3 has attracted considerable interest as a wide-band gap semiconductor for advanced high-k capacitors and photocatalytic applications. Although previous angle-resolved photoemission spectroscopy (ARPES) studies have characterized the valence band structure originating from O 2p orbitals, the conduction band arising from Ti 3d orbitals upon electron dopi

Siqu Ou, Hongcheng Liu, Pingjie Wang, Yusheng Liao

While chains-of-thought (CoT) have advanced complex reasoning in multimodal large language models (MLLMs), existing methods remain confined to text or static visual domains, often faltering in dynamic spatial reasoning tasks. To bridge this gap, we present GRASSLAND, a novel maze navigation benchmark designed to evaluate dynamic spatial reasoning. Our experi

Hikaru Kawai, Nobuyoshi Ohta

Recently it is asserted that the standard beta function does not describe the correct running of the coupling constant in some theories. We show that the problem arises from the assumption $\mu=p$ ($\mu$ is a renormalization point) and that a suitable choice of $\mu$ gives the correct running. It is also claimed that neither the cosmological constant nor New

Yu Zuo, Dalin Qin, Yi Wang

The growing complexity of power systems has made accurate load forecasting more important than ever. An increasing number of advanced load forecasting methods have been developed. However, the static design of current methods offers no mechanism for human-model interaction. As the primary users of forecasting models, system operators often find it difficult

Romario Gildas Foko Tiomela, Samson Adekola Alagbe, Olawale Nasiru Lawal, Serges Love Teutu Talla

This study introduces a comparative modeling framework using stationary and non-stationary transition probabilities within a Markov Decision Process (MDP) to assess COVID-19 disease dynamics. Stationary transition probabilities assume constant transition rates, while non-stationary transitions reflect time-dependent behaviors including policy interventions o

Spencer Hong, Meng Luo, Xinyi Wan

Determining the veracity of atomic claims is an imperative component of many recently proposed fact-checking systems. Many approaches tackle this problem by first retrieving evidence by querying a search engine and then performing classification by providing the evidence set and atomic claim to a large language model, but this process deviates from what a hu

Alberto Jimenez-Ruiz, Federico Milano

The rising demand of computing power leads to the installation of a large number of Data Centers (DCs). Their Fault-Ride-Through (FRT) behavior and their unique power characteristics, especially for DCs catered to Artificial Intelligence (AI) workloads, pose a threat to the stability of power systems. To ensure its stability, it is required accurate models o

Barak Budnick, Ofer Biham, Eytan Katzav

We present analytical results for the effect of preferential node deletion on the structure of networks that evolve via node addition and preferential attachment. To this end, we consider a preferential-attachment-preferential-deletion (PAPD) model, in which at each time step, with probability $P_{\rm add}$ there is a growth step where an isolated node is ad

Yi Hu, Hanchi Ren, Jingjing Deng, Xianghua Xie

Stock price prediction is a critical area of financial forecasting, traditionally approached by training models using the historical price data of individual stocks. While these models effectively capture single-stock patterns, they fail to leverage potential correlations among stock trends, which could improve predictive performance. Current single-stock le

Eugenio Dellepiane, Marco M. Peloso, Anita Tabacco

This work is devoted to the comparison of de Branges--Rovnyak $H(b)$ spaces harmonically weighted Dirichlet spaces $\mathcal{D}_\mu$. We completely characterize which $H(b)$ spaces are also harmonically weighted Dirichlet spaces $\mathcal{D}_\mu$, when $\mu$ is a finite sum of atoms. This is a generalization of a previous result by Costara--Ransford \cite{co

Anna Brandenberger, Simon Briend, Hannah Cairns, Robin Khanfir

Uniform attachment with freezing is an extension of the classical model of random recursive trees, in which trees are recursively built by attaching new vertices to old ones. In the model of uniform attachment with freezing, vertices are allowed to freeze, in the sense that new vertices cannot be attached to already frozen ones. We study the impact of removi

Dongyang Fan, Vinko Sabolčec, Martin Jaggi

Large Language Models (LLMs) are commonly pretrained on vast corpora of text without utilizing contextual metadata such as source, quality, or topic, leading to a context-free learning paradigm. While recent studies suggest that adding metadata like URL information as context (i.e., auxiliary inputs not used in the loss calculation) can improve training effi

Liudmila Bishler, Andrei Mironov, Alexei Morozov

Vogel's universality implies a unified description of the adjoint sector of representation theory for simple Lie algebras in terms of three parameters $\alpha,\beta,\gamma$, which are homogeneous coordinates of Vogel's plane. Actually this is true (if at all) only for a piece of representation theory captured by knot/Chern-Simons theory, where some irreducib

Özgür Esentepe

In a pandemic era preprint, Dao showed showed two remarkable properties of Arf rings: under some mild conditions, they admit finitely many indecomposable reflexive modules up to isomorphism and every reflexive module is actually isomorphic to its own dual. In fact, the latter property characterises Arf rings. Arf rings are one dimensional rings and it is nat

Thibaud Gloaguen, Mark Vero, Robin Staab, Martin Vechev

Finetuning open-weight Large Language Models (LLMs) is standard practice for achieving task-specific performance improvements. Until now, finetuning has been regarded as a controlled and secure process in which training on benign datasets leads to predictable behaviors. In this paper, we demonstrate, for the first time, that an adversary can create compromis

Dongwon Noh, Donghyeok Koh, Junghun Yuk, Gyuwan Kim

Prior benchmarks for evaluating the domain-specific knowledge of large language models (LLMs) lack the scalability to handle complex academic tasks. To address this, we introduce \texttt{ScholarBench}, a benchmark centered on deep expert knowledge and complex academic problem-solving, which evaluates the academic reasoning ability of LLMs and is constructed

Nina Shvetsova, Goutam Bhat, Prune Truong, Hilde Kuehne

We tackle the problem of monocular-to-stereo video conversion and propose a novel architecture for inpainting and refinement of the warped right view obtained by depth-based reprojection of the input left view. We extend the Stable Video Diffusion (SVD) model to utilize the input left video, the warped right video, and the disocclusion masks as conditioning

Francisco J. Aragón-Artacho, Heinz H. Bauschke, Rubén Campoy, César López-Pastor

In this paper, we develop a general analysis for the fixed points of the operators defining the graph splitting methods from [SIAM J. Optim., 34 (2024), pp. 1569-1594] by Bredies, Chenchene and Naldi. We particularize it to the case where the maximally monotone operators are normal cones of closed linear subspaces and provide an explicit formula for the limi

Son Tran, Edjard Mota, Artur d'Avila Garcez

Knowledge representation and reasoning in neural networks have been a long-standing endeavor which has attracted much attention recently. The principled integration of reasoning and learning in neural networks is a main objective of the area of neurosymbolic Artificial Intelligence (AI). In this chapter, a simple energy-based neurosymbolic AI system is descr

Chen Gong, Rui Xing, Zhenzhe Zheng, Fan Wu

The demand for machine learning (ML) model training on edge devices is escalating due to data privacy and personalized service needs. However, we observe that current on-device model training is hampered by the under-utilization of on-device data, due to low training throughput, limited storage and diverse data importance. To improve data resource utilizatio

Haihao Shi, Zhenyang Huang, Qiyu Yan, Jun Li

In axion models, interactions between axions and electromagnetic waves induce frequency-dependent time delays determined by the axion mass and decay constant. These small delays are difficult to detect, limiting the effectiveness of traditional methods. We compute such delays under realistic radio telescope conditions and identify a prominent dispersive feat

Jannis Becktepe, Leona Hennig, Steffen Oeltze-Jafra, Marius Lindauer

Medical Image Segmentation (MIS) includes diverse tasks, from bone to organ segmentation, each with its own challenges in finding the best segmentation model. The state-of-the-art AutoML-related MIS-framework nnU-Net automates many aspects of model configuration but remains constrained by fixed hyperparameters and heuristic design choices. As a full-AutoML f

Nao Mochizuki, Marvin Plogmann

We prove that, for a proper connective dg algebra $A$ with cohomology concentrated in degrees between $1-d$ and $0$, the extended heart $\mathcal{D}^{fd}(A)^{(-d,0]}\subseteq \mathcal{D}^{fd}(A)$ is an extriangulated category with almost-split conflations. We also prove a version of the 1st Brauer--Thrall Conjecture in this context.

Biao Yi, Tiansheng Huang, Baolei Zhang, Tong Li

Fine-tuning-as-a-service, while commercially successful for Large Language Model (LLM) providers, exposes models to harmful fine-tuning attacks. As a widely explored defense paradigm against such attacks, unlearning attempts to remove malicious knowledge from LLMs, thereby essentially preventing them from being used to perform malicious tasks. However, we hi

Xin Wu, Qi Chen, Ye Xing, Zhi-Peng Xing

This paper present a systematic study of hyperon non-leptonic two-body decays induced by light quark transitions, particularly the $s \rightarrow u\bar{u}d$ process, within the framework of SU(3) flavor symmetry. The effective weak Hamiltonian is decomposed into irreducible SU(3) representations, including the 27-plet and octet components, and applied to ana

Junchi Yao, Jianhua Xu, Tianyu Xin, Ziyi Wang

The rise of Large Language Model-based Multi-Agent Planning has leveraged advanced frameworks to enable autonomous and collaborative task execution. Some systems rely on platforms like review sites and social media, which are prone to fraudulent information, such as fake reviews or misleading descriptions. This reliance poses risks, potentially causing finan

Loris Belcastro, Fabrizio Marozzo, Alessio Orsino, Domenico Talia

The edge-cloud continuum has emerged as a transformative paradigm that meets the growing demand for low-latency, scalable, end-to-end service delivery by integrating decentralized edge resources with centralized cloud infrastructures. Driven by the exponential growth of IoT-generated data and the need for real-time responsiveness, this continuum features mul

Dongryul M. Kim, Andrew Zimmer

In this paper we introduce Patterson--Sullivan systems, which consist of a group action on a compact metrizable space and a quasi-invariant measure which behaves like a classical Patterson--Sullivan measure. For such systems we prove a generalization of Tukia's measurable boundary rigidity theorem. We then apply this generalization to (1) study the singulari

Arash Yavari, Alain Goriely

We study particle dynamics under curl forces. These forces are a class of non-conservative, non-dissipative, position-dependent forces that cannot be expressed as gradient of a potential function. We show that the fundamental quantity of particle dynamics under curl forces is a work $1$-form. By using the Darboux classification of differential $1$-forms on $

Francesco Ancilotto

In a recent paper (Y. Ma and X. Cui, Phys. Rev. Lett. 134, 043402 (2025)), a new type of shell-shaped Bose-Einstein condensate with a self-bound character has been proposed, made of three-component $Na^{23}K^{39}K^{41}$ Bose mixture (species (1,2,3) in the following), where the mixtures (1, 2) and (2, 3) both form quantum droplets. The proposed structures ar

Ziyang Yan, Jason L. Sanders

The Gaia dataset has revealed many intricate Milky Way substructures in exquisite detail, including moving groups and the phase spiral. Precise characterisation of these features and detailed comparisons to theoretical models require engaging with Gaia's heteroscedastic noise model, particularly in more distant parts of the Galactic disc and halo. We propose

Wenhui Tan, Jiaze Li, Jianzhong Ju, Zhenbo Luo

Large Language Models (LLMs) achieve superior performance through Chain-of-Thought (CoT) reasoning, but these token-level reasoning chains are computationally expensive and inefficient. In this paper, we introduce Compressed Latent Reasoning (CoLaR), a novel framework that dynamically compresses reasoning processes in latent space through a two-stage trainin

David Carral, Lukas Gerlach, Lucas Larroque, Michaël Thomazo

The chase is a fundamental algorithm with ubiquitous uses in database theory. Given a database and a set of existential rules (aka tuple-generating dependencies), it iteratively extends the database to ensure that the rules are satisfied in a most general way. This process may not terminate, and a major problem is to decide whether it does. This problem has

Maximilian Inckmann, Nicolas Blumenröhr, Rossella Aversa

FAIR Digital Objects support research data management aligned with the FAIR principles. To be machine-actionable, they must support operations that interact with their contents. This can be achieved by associating operations with FAIR-DO data types. However, current typing models and Data Type Registries lack support for type-associated operations. In this w

Trung V. Phan, George A. Kevrekidis, Soledad Villar, Yannis G. Kevrekidis

The machine learning methods for data-driven identification of partial differential equations (PDEs) are typically defined for a given number of spatial dimensions and a choice of coordinates the data have been collected in. This dependence prevents the learned evolution equation from generalizing to other spaces. In this work, we reformulate the problem in

Lucas Maystre, Gabriel Barello, Tudor Berariu, Aleix Cambray

We address the problem of incremental sequence classification, where predictions are updated as new elements in the sequence are revealed. Drawing on temporal-difference learning from reinforcement learning, we identify a temporal-consistency condition that successive predictions should satisfy. We leverage this condition to develop a novel loss function for

Nitesh Subedi, Hsin-Jung Yang, Devesh K. Jha, Soumik Sarkar

Autonomous harvesting in the open presents a complex manipulation problem. In most scenarios, an autonomous system has to deal with significant occlusion and require interaction in the presence of large structural uncertainties (every plant is different). Perceptual and modeling uncertainty make design of reliable manipulation controllers for harvesting chal

Jun Maeda, Yuya Tanizaki

For quantum field theories with global symmetry, we can study the behavior of the partition function with the background gauge field to diagnose different quantum phases. For the case of discrete symmetries, we find that the symmetry-twisted partition function works as an order parameter that discriminates spontaneous symmetry breaking (SSB), symmetry-protec

Baibhab Bose, Devvrat Tiwari, Subhashish Banerjee

The theory of a two-level $\eta$-Hermitian Hamiltonian with $\mathcal{PT}$ symmetry is reviewed and extended to include open system dynamics. A first-principles derivation of the generalized Gorini-Kossakowski-Sudarshan-Lindblad master equation appropriate for a $\mathcal{PT}-$symmetric Hamiltonian is presented. Inspired by a simple light matter interaction

Zhe Liu, Shashi B. Mishra, Jae-Mo Lihm, Samuel Poncé

Topological Weyl semimetals represent a novel class of quantum materials that exhibit remarkable properties arising from their unique electronic structure. In this work, we employ state-of-the-art ab initio methods to investigate the role of the electron-phonon interactions on the charge transport properties of TaAs. Our calculations of the temperature-depen

Robb Calder, Oliver Shorttle, Sean Jordan, Paul Rimmer

Ozone is a potential biosignature and disambuguator between Earth-like and Venus-like exoplanets due to its association on Earth with photosynthetically produced oxygen (O$_2$). However, the existence of ozone in Venus's observable atmosphere, a planet with no known life, raises the possibility of ozone biosignature false-positives on Venus-like exoplanets.

Johannes Ebert

Ruberman constructed families $\{g_n\vert n \in \mathbb{N}\} \subset \mathcal{R}^+ (M)$ of metrics of positive scalar curvature on certain $4$-manifolds which are concordant but lie in different path components of $\mathcal{R}^+ (M)$. We prove a cancellation result along the following lines. For each closed manifold $N$, there is a map $\nu_N: \mathcal{R}^+

Kayn A. Forbes, David L. Andrews

For paraxial light beams and electromagnetic fields, the Stokes vector and polarization matrix provide equivalent scalar measures of optical chirality, widely used in linear optics. However, growing interest in non-paraxial fields, with fully three-dimensional polarization components, necessitates an extended framework. Here, we develop a general theory for

Inbal Cohen, Boaz Meivar, Peihan Tu, Shai Avidan

Segment Anything Models (SAM) have achieved remarkable success in object segmentation tasks across diverse datasets. However, these models are predominantly trained on large-scale semantic segmentation datasets, which introduce a bias toward object shape rather than texture cues in the image. This limitation is critical in domains such as medical imaging, ma

Deepak Kanneganti, Sajib Mistry, Sheik Mohammad Mostakim Fattah, Aneesh Krishna

The dynamic nature of Internet of Things (IoT) environments challenges the long-term effectiveness of Machine Learning as a Service (MLaaS) compositions. The uncertainty and variability of IoT environments lead to fluctuations in data distribution, e.g., concept drift and data heterogeneity, and evolving system requirements, e.g., scalability demands and res

C. Duque-Arribas, H. M. Tabernero, D. Montes, J. A. Caballero

M dwarfs are the most abundant stars in the Galaxy and serve as key targets for stellar and exoplanetary studies. It is particularly challenging to determine their metallicities because their spectra are complex. For this reason, several authors have focused on photometric estimates of the M-dwarf metallicity. Although artificial neural networks have been us

Ercong Nie, Helmut Schmid, Hinrich Schütze

Language confusion -- where large language models (LLMs) generate unintended languages against the user's need -- remains a critical challenge, especially for English-centric models. We present the first mechanistic interpretability (MI) study of language confusion, combining behavioral benchmarking with neuron-level analysis. Using the Language Confusion Be

Gertrude Hattoh, Jeremiah Ayensu, Nyarko Prince Ofori, Solomon Eshun

Advances in AI, particularly LLMs, have dramatically shortened drug discovery cycles by up to 40% and improved molecular target identification. However, these innovations also raise dual-use concerns by enabling the design of toxic compounds. Prompting Moremi Bio Agent without the safety guardrails to specifically design novel toxic substances, our study gen

Varatharaja Nallathambi, Se-Ho Kim, Baptiste Gault, Sven Reichenberger

The ability to tailor nanoscale surface atom arrangements through multi-elemental compositional control provides high-entropy nanoalloys with promising functional properties. Developing a fundamental understanding of nanoalloy formation mechanisms during synthesis is therefore essential for effectively engineering the surface composition and resulting functi

Yao Xu, Mingyu Xu, Fangyu Lei, Wangtao Sun

Recently, models such as OpenAI-o1 and DeepSeek-R1 have demonstrated remarkable performance on complex reasoning tasks through Long Chain-of-Thought (Long-CoT) reasoning. Although distilling this capability into student models significantly enhances their performance, this paper finds that fine-tuning LLMs with full parameters or LoRA with a low rank on long

Felipe Hawthorne, Pablo de Castro, José A. Freire

Self-propelled particles, like motile cells and artificial colloids, can spontaneously form macroscopic clusters. This phenomenon is called motility-induced phase separation (MIPS) and occurs even without attractive forces, provided that the self-propulsion direction fluctuates slowly. In addition to rotational noise, these particles may experience translati

Asrar Alruwayqi

We present a novel framework for dynamic 3D scene reconstruction that integrates three key components: an explicit tri-plane deformation field, a view-conditioned canonical radiance field with spherical harmonics (SH) attention, and a temporally-aware latent diffusion prior. Our method encodes 4D scenes using three orthogonal 2D feature planes that evolve ov