May 2025 arXiv papers — page 107

Connor Mooney

We prove that viscosity solutions to the quadratic Hessian equation $$\sigma_2(D^2u) = 1$$ cannot touch a harmonic function on a minimal surface from below. This can be viewed as a form of strict $2$-convexity. We also prove an a priori interior $C^2$ estimate in terms of the $W^{2,\,p}$ norm, for any $p > 2$. Finally, we discuss how these results rule out c

Wenbin Hu, Haoran Li, Huihao Jing, Qi Hu

While Large Language Models (LLMs) exhibit remarkable capabilities, they also introduce significant safety and privacy risks. Current mitigation strategies often fail to preserve contextual reasoning capabilities in risky scenarios. Instead, they rely heavily on sensitive pattern matching to protect LLMs, which limits the scope. Furthermore, they overlook es

Yolanda Cabrera Casado, Maria Inez Cardoso Gonçalves, Daniel Gonçalves, Dolores Martín Barquero

Let A be an evolution algebra (possibly infinite-dimensional) equipped with a fixed natural basis B, and let E be the associated graph defined by Elduque and Labra. We describe the group of automorphisms of A that are diagonalizable with respect to B. This group arises as the inverse limit of a functor (a diagram) from the category associated with the graph

Abhimanyu Talwar, Julien Laasri

Recently proposed neural network architectures like PointNet [QSMG16] and PointNet++ [QYSG17] have made it possible to apply Deep Learning to 3D point sets. The feature representations of shapes learned by these two networks enabled training classifiers for Semantic Segmentation, and more recently for Instance Segmentation via the Similarity Group Proposal N

Yan Wang, Ling Ding, Tien N Nguyen, Shaohua Wang

Large Language Models for code often entail significant computational complexity, which grows significantly with the length of the input code sequence. We propose LeanCode for code simplification to reduce training and prediction time, leveraging code contexts in utilizing attention scores to represent the tokens' importance. We advocate for the selective re

Shangziqi Zhao, Jiahao Yuan, Jinyang Wu, Zhenglin Wang

Long chain-of-thought (Long-CoT) reasoning improves accuracy in LLMs, yet its verbose, self-reflective style often hinders effective distillation into small language models (SLMs). We revisit Long-CoT compression through the lens of capability alignment and ask: Can pruning improve reasoning? We propose Prune-on-Logic, a structure-aware framework that transf

Deemah H. Tashman, Soumaya Cherkaoui, Walaa Hamouda

In this paper, a reinforcement learning technique is employed to maximize the performance of a cognitive radio network (CRN). In the presence of primary users (PUs), it is presumed that two secondary users (SUs) access the licensed band within underlay mode. In addition, the SU transmitter is assumed to be an energy-constrained device that requires harvestin

Yaroslav Marchukov, Luis Montano

Planning in environments with moving obstacles remains a significant challenge in robotics. While many works focus on navigation and path planning in obstacle-dense spaces, traversing such congested regions is often avoidable by selecting alternative routes. This paper presents Traversability-aware FMM (Tr-FMM), a path planning method that computes paths in

Sina Tootoonian, Andreas T. Schaefer

A common view of sensory processing is as probabilistic inference of latent causes from receptor activations. Standard approaches often assume these causes are a priori independent, yet real-world generative factors are typically correlated. Representing such structured priors in neural systems poses architectural challenges, particularly when direct interac

Takuya Isogawa, Guoqing Wang, Boning Li, Zhiyao Hu

Quantum multiparameter estimation promises to extend quantum advantage to the simultaneous high-precision measurements of multiple physical quantities. However, realizing this capability in practical quantum sensors under realistic conditions remains challenging due to intrinsic system imperfections. Here, we experimentally demonstrate multiparameter estimat

Sohaila Eltanbouly, Salam Albatarni, Tamer Elsayed

Research on holistic Automated Essay Scoring (AES) is long-dated; yet, there is a notable lack of attention for assessing essays according to individual traits. In this work, we propose TRATES, a novel trait-specific and rubric-based cross-prompt AES framework that is generic yet specific to the underlying trait. The framework leverages a Large Language Mode

Emerson Gehr, Abigail Terrell, Katrina Vermillion, Alexandria Mendoza

This study investigates the filamentary structural states of microgravity dusty plasma using data from the Plasmakristall-4 (PK-4) facility on board the International Space Station. The dust particles in the PK-4 discharge are observed to form field-aligned filaments and nested (layered) structures in response to changes in the plasma conditions, neutral gas

Simran Kumari, Anand Ronald K., Siddhartha Mukhopadhyay, Ashish R. Hota

Most of the existing state-of-the-art approaches for energy consumption analysis do not account for the effect of lateral dynamics on energy consumption in electric vehicles (EVs) during vehicle maneuvers. This paper aims to validate this effect through an experimental study. We develop a scaled model using a radio-controlled (RC) car, modified to achieve dy

Xi Wang, Susmit Shannigrahi

Efficient data replication in decentralized storage systems must account for diverse policies, especially in multi-organizational, data-intensive environments. This work proposes PSMOA, a novel Policy Support Multi-objective Optimization Algorithm for decentralized data replication that dynamically adapts to varying organizational requirements such as minimi

Simona J. Miller, Maximiliano Isi, Katerina Chatziioannou, Vijay Varma

Robustly measuring binary black hole spins via gravitational waves is key to understanding these systems' astrophysical origins, but remains challenging -- especially for high-mass systems, whose signals are short and dominated by the merger. Nonetheless, events like GW190521 show that strong spin precession can indeed be gleaned. In this work, we track how

Jayroop Ramesh, Valentin Bacher, Mark C. Eid, Hoda Kalabizadeh

The International Society of Ultrasound advocates Intrapartum Ultrasound (US) Imaging in Obstetrics and Gynecology (ISUOG) to monitor labour progression through changes in fetal head position. Two reliable ultrasound-derived parameters that are used to predict outcomes of instrumental vaginal delivery are the angle of progression (AoP) and head-symphysis dis

Ruben Burkard, Benedikt Schneider, Björn Sbierski

For quantum spin systems in equilibrium, the dynamic structure factor (DSF) is among the most feature-packed experimental observables. However, from a theory perspective it is often hard to simulate in an unbiased and accurate way, especially for frustrated and high-dimensional models at intermediate temperature. To address this challenge, we compute the DSF

Richard Kang, Leonardo A. Cunha, Diptarka Hait, Martin Head-Gordon

Quantum mechanical calculations of core electron binding energies (CEBEs) leading to 2p hole states are relevant to interpreting L-edge x-ray photo-electron spectroscopy (XPS), as well as higher edges. Orbital-optimized density functional theory (OO-DFT) accurately predicts K-edge CEBEs but is challenged by the presence of significant spin-orbit coupling (SO

Devansh Bhardwaj, Arjun Beniwal, Shreyas Chaudhari, Ashwin Kalyan

AI agents have become increasingly adept at complex tasks such as coding, reasoning, and multimodal understanding. However, building generalist systems requires moving beyond individual agents to collective inference -- a paradigm where multi-agent systems with diverse, task-specialized agents complement one another through structured communication and colla

Alayt Issak, Uttkarsh Narayan, Ramya Srinivasan, Erica Kleinman

Ethical theories and Generative AI (GenAI) models are dynamic concepts subject to continuous evolution. This paper investigates the visualization of ethics through a subset of GenAI models. We expand on the emerging field of Visual Ethics, using art as a form of critical inquiry and the metaphor of a kaleidoscope to invoke moral imagination. Through formativ

Coleridge Faraday, W. A. Horowitz

We present suppression predictions from our pQCD-based energy loss model, which receives small system size corrections, for high-$p_T$ $\pi$, $D$ and $B$ meson $R_{AB}$ as a function of centrality, flavor, $\sqrt{s_{NN}}$, and $p_T$ from large to small collision systems at RHIC and LHC. A statistical analysis is used to constrain the effective strong couplin

Riku Iimori, Yuta Kodani, Shaojie Hu, Takashi Kimura

2D van der Waals (vdW) ferromagnets have emerged as promising materials for spintronic applications due to their unique magnetic properties and tunability. Controlling ferromagnetism via external stimuli is critical for both fundamental research and device integration. In particular, modulation of magnetic anisotropy and exchange interactions through strain

Andrei Cozma, Landon Harris, Hairong Qi

Reinforcement Learning (RL) has made significant strides in various domains, and policy gradient methods like Proximal Policy Optimization (PPO) have gained popularity due to their balance in performance, training stability, and computational efficiency. These methods directly optimize policies through gradient-based updates. However, developing effective co

Pietro Saggese, Michael Fröwis, Stefan Kitzler, Bernhard Haslhofer

Total Value Locked (TVL) aims to measure the aggregate value of cryptoassets deposited in Decentralized Finance (DeFi) protocols. Although blockchain data is public, the way TVL is computed is not well understood. In practice, its calculation on major TVL aggregators relies on self-reports from community members and lacks standardization, making it difficult

David Krame Kadurha, Domini Jocema Leko Moutouo, Yae Ulrich Gaba

This paper reviews the topological groundwork for the study of reinforcement learning (RL) by focusing on the structure of state, action, and policy spaces. We begin by recalling key mathematical concepts such as complete metric spaces, which form the foundation for expressing RL problems. By leveraging the Banach contraction principle, we illustrate how the

Nicole Vassh, Yilin Wang, Richard M. Woloshyn, Michelle P. Kuchera

We apply the capabilities of machine learning (ML) to discern patterns in order to classify metal-poor stars. To do so, we train an ML model on a bank of nucleosynthesis calculations derived from hydrodynamic simulations for events such as neutron star mergers where the rapid ($r$) neutron capture process can take place. Likewise we consider a bank of calcul

Parthasaarathy Sudarsanam, Irene Martín-Morató, Tuomas Virtanen

This paper proposes a single-stage training approach that semantically aligns three modalities - audio, visual, and text using a contrastive learning framework. Contrastive training has gained prominence for multimodal alignment, utilizing large-scale unlabeled data to learn shared representations. Existing deep learning approach for trimodal alignment invol

Theo Lepage, Reda Dehak

Self-Supervised Learning (SSL) has led to considerable progress in Speaker Verification (SV). The standard framework uses same-utterance positive sampling and data-augmentation to generate anchor-positive pairs of the same speaker. This is a major limitation, as this strategy primarily encodes channel information from the recording condition, shared by the a

Yuan Gao, Wenhan Guo, Yu Sun

Inverse scattering is a fundamental challenge in many imaging applications, ranging from microscopy to remote sensing. Solving this problem often requires jointly estimating two unknowns -- the image and the scattering field inside the object -- necessitating effective image prior to regularize the inference. In this paper, we propose a regularized neural fi

John Rincon, Alexander R. Pelletier, Destiny Gilliland, Wei Wang

Objective: As AI becomes increasingly central to healthcare, there is a pressing need for bioinformatics and biomedical training systems that are personalized and adaptable. Materials and Methods: The NIH Bridge2AI Training, Recruitment, and Mentoring (TRM) Working Group developed a cross-disciplinary curriculum grounded in collaborative innovation, ethical

Maxim Vishnikin, Alexander Okhotin

A categorial grammar assigns one of several syntactic categories to each symbol of the alphabet, and the category of a string is then deduced from the categories assigned to its symbols using two simple reduction rules. This paper investigates a special class of categorial grammars, in which only one category is assigned to each symbol, thus eliminating ambi

Xiangxu Zhang, Lei Li, Xiao Zhou, Zheng Liu

Current medical retrieval benchmarks primarily emphasize lexical or shallow semantic similarity, overlooking the reasoning-intensive demands that are central to clinical decision-making. In practice, physicians often retrieve authoritative medical evidence to support diagnostic hypotheses. Such evidence typically aligns with an inferred diagnosis rather than

Klaus Bering

We prove formulas for the multi-instanton corrections to the overlap and energies of a 1D same-level asymmetric double well using the Euclidean path integral. Both the odd and even instanton sectors are summed to all orders. The double well is same-level asymmetric in the sense that the potentials at neighboring wells have the same bottom level but can have

Marlène Careil, Yohann Benchetrit, Jean-Rémi King

Brain-to-image decoding has been recently propelled by the progress in generative AI models and the availability of large ultra-high field functional Magnetic Resonance Imaging (fMRI). However, current approaches depend on complicated multi-stage pipelines and preprocessing steps that typically collapse the temporal dimension of brain recordings, thereby lim

Yingtao Luo, Shikai Fang, Binqing Wu, Qingsong Wen

Weather forecasting is essential but remains computationally intensive and physically incomplete in traditional numerical weather prediction (NWP) methods. Deep learning (DL) models offer efficiency and accuracy but often ignore physical laws, limiting interpretability and generalization. We propose PhyDL-NWP, a physics-guided deep learning framework that in

D. G. C. McKeon, F. T. Brandt, J. Frenkel, S. Martins-Filho

A Lagrange multiplier field can be used to restrict radiative corrections to the Einstein-Hilbert action to one-loop order. This result is employed to show that it is possible to couple a scalar field to the metric (graviton) field in such a way that the model is both renormalizable and unitary. The usual Einstein equations of motion for the gravitational fi

Abhimanyu Talwar, Julien Laasri

Certain pairs of languages suffer from lack of a parallel corpus which is large in size and diverse in domain. One of the ways this is overcome is via use of a pivot language. In this paper we use Hindi as a pivot language to translate Nepali into English. We describe what makes Hindi a good candidate for the pivot. We discuss ways in which a pivot language

Nazmus Saadat As-Saquib, A N M Nafiz Abeer, Hung-Ta Chien, Byung-Jun Yoon

Training neural networks has traditionally relied on backpropagation (BP), a gradient-based algorithm that, despite its widespread success, suffers from key limitations in both biological and hardware perspectives. These include backward error propagation by symmetric weights, non-local credit assignment, and frozen activity during backward passes. We propos

Alexander Gutfraind, Vicki Bier

Large language models (LLMs) offer unprecedented and growing capabilities, but also introduce complex safety and security challenges that resist conventional risk management. While conventional probabilistic risk analysis (PRA) requires exhaustive risk enumeration and quantification, the novelty and complexity of these systems make PRA impractical, particula

Jiajun Shi, Jian Yang, Jiaheng Liu, Xingyuan Bu

Recent advancements in large language models (LLMs) underscore the need for more comprehensive evaluation methods to accurately assess their reasoning capabilities. Existing benchmarks are often domain-specific and thus cannot fully capture an LLM's general reasoning potential. To address this limitation, we introduce the Knowledge Orthogonal Reasoning Gymna

Petros Drineas, Rohit Nema, Rafail Ostrovsky, Vassilis Zikas

We investigate how a blockchain can distill the collective belief of its nodes regarding the trustworthiness of a (sub)set of nodes into a {\em reputation system} that reflects the probability of correctly performing a task. To address this question, we introduce a framework that breaks it down into two sub-problems: 1. (Information Extraction): How can the

Ivan Biočić, Bruno Toaldo

In this paper, we develop a universal method that identifies the (non-local) governing evolution equations for Continuous Time Random Walks' (CTRWs) limit processes. Given one of these processes, our method provides the form of a non-local operator, acting on space and time variables jointly, such that the (generalized) harmonic problem associated with it re

Dzung Pham, Peter Kairouz, Niloofar Mireshghallah, Eugene Bagdasarian

Large language models (LLMs) are increasingly being used in privacy pipelines to detect and remedy sensitive data leakage. These solutions often rely on the premise that LLMs can reliably recognize human names, one of the most important categories of personally identifiable information (PII). In this paper, we reveal how LLMs can consistently mishandle broad

Bilal Islah, Ahmed Zoulati

We offer evidence that federal emergency assistance (FEMA) in the days following natural disasters mitigate evictions in comparison to similar emergency scenarios where FEMA aid is not provided. We find an approximate 10.9% increase in overall evictions after hurricane natural disaster events driven in large part by areas in close proximity of the hurricane

Noah Krever, Jakub Černý, Moïse Blanchard, Christian Kroer

Game-theoretic algorithms are commonly benchmarked on recreational games, classical constructs from economic theory such as congestion and dispersion games, or entirely random game instances. While the past two decades have seen the rise of security games -- grounded in real-world scenarios like patrolling and infrastructure protection -- their practical eva

Ilias Giannakopoulos, José E. Cruz Serrallés, Jan Paška, Martijn A. Cloos

Objective: Global Maxwell Tomography (GMT) is a noninvasive inverse optimization method for the estimation of electrical properties (EP) from magnetic resonance (MR) measurements. GMT uses the volume integral equation (VIE) in the forward problem and assumes that the sample has negligible effect on the coil currents. Consequently, GMT calculates the coil's i

Slava Pimenov, Angel Toledo

Let $(\mathcal{C}, \otimes)$ be a monoidal dg-category. We construct a complex controlling the deformation of the monoidal structure on $\mathcal{C}$ together with the deformation of the underlying dg-category itself. We show that in the case of a semisimple category $\mathcal{C}$ it reduces to the Davydov-Yetter complex. Furthermore, we study this complex i

Saahil Mahato

Urban traffic congestion, particularly at intersections, significantly affects travel time, fuel consumption, and emissions. Traditional fixed-time signal control systems often lack the adaptability to effectively manage dynamic traffic patterns. This study explores the application of multi-agent reinforcement learning (MARL) to optimize traffic signal coord

Utsav Dutta, Sina Khoshfetrat Pakazad, Henrik Ohlsson

Traditional time series models are task-specific and often depend on dataset-specific training and extensive feature engineering. While Transformer-based architectures have improved scalability, foundation models, commonplace in text, vision, and audio, remain under-explored for time series and are largely restricted to forecasting. We introduce $\textbf{CHA

Stefan Gieseke, Stefan Kiebacher, Simon Plätzer, Jan Priedigkeit

We introduce building blocks for the cluster hadronization model in light of a new structure, focusing on cluster fission and cluster decay. We propose theoretically motivated matrix elements for cluster fission and decay as building blocks and study some first phenomenological implications at different energies. In particular we develop a set of observables

Chuanbo Tang, Zhuoyuan Li, Yifan Bian, Li Li

Efficient video coding is highly dependent on exploiting the temporal redundancy, which is usually achieved by extracting and leveraging the temporal context in the emerging conditional coding-based neural video codec (NVC). Although the latest NVC has achieved remarkable progress in improving the compression performance, the inherent temporal context propag

Fan Yi, Haoran Wan, Kyle Jamieson, Oliver Michel

5G wireless networks are complex, leveraging layers of scheduling, retransmission, and adaptation mechanisms to maximize their efficiency. But these mechanisms interact to produce significant fluctuations in uplink and downlink capacity and latency. This markedly impacts the performance of real-time applications, such as video-conferencing, which are particu

Gaia Belardinelli, Thomas Bolander, Sebastian Watzl

In this work, we present the first general logic of attention. Attention is a powerful cognitive ability that allows agents to focus on potentially complex information, such as logically structured propositions, higher-order beliefs, or what other agents pay attention to. This ability is a strength, as it helps to ignore what is irrelevant, but it can also i

Abouzied M. A. Nasar, Benedict D. Rogers, Georgios Fourtakas, Mladen Ivkovic

This paper highlights first steps towards enabling graphics processing unit (GPU) acceleration of the task-parallel smoothed particle hydrodynamics (SPH) solver SWIFT. Novel combinations of algorithms are presented, enabling SWIFT to function as a truly heterogeneous software leveraging task-parallelism on CPUs for memory-bound computations concurrently with

Yuxuan Wang, Xuanyu Yi, Qingshan Xu, Yuan Zhou

Personalizing 3D scenes from a single reference image enables intuitive user-guided editing, which requires achieving both multi-view consistency across perspectives and referential consistency with the input image. However, these goals are particularly challenging due to the viewpoint bias caused by the limited perspective provided in a single image. Lackin

Agam Goyal, Vedant Rathi, William Yeh, Yian Wang

Large language models (LLMs) are now ubiquitous in user-facing applications, yet they still generate undesirable toxic outputs, including profanity, vulgarity, and derogatory remarks. Although numerous detoxification methods exist, most apply broad, surface-level fixes and can therefore easily be circumvented by jailbreak attacks. In this paper we leverage s

Jiangrong Shen, Yulin Xie, Qi Xu, Gang Pan

Multimodal spiking neural networks (SNNs) hold significant potential for energy-efficient sensory processing but face critical challenges in modality imbalance and temporal misalignment. Current approaches suffer from uncoordinated convergence speeds across modalities and static fusion mechanisms that ignore time-varying cross-modal interactions. We propose

Chih-Yu Chang, Milad Azvar, Chinedum Okwudire, Raed Al Kontar

Bayesian optimization (BO) is a sequential decision-making tool widely used for optimizing expensive black-box functions. Recently, Large Language Models (LLMs) have shown remarkable adaptability in low-data regimes, making them promising tools for black-box optimization by leveraging contextual knowledge to propose high-quality query points. However, relyin

Chongyang Shi, Sharon Lin, Shuang Song, Jamie Hayes

Gemini is increasingly used to perform tasks on behalf of users, where function-calling and tool-use capabilities enable the model to access user data. Some tools, however, require access to untrusted data introducing risk. Adversaries can embed malicious instructions in untrusted data which cause the model to deviate from the user's expectations and mishand

Mohammad Irfan Uddin, Nishad Tasnim, Md Omor Faruk, Zejian Zhou

Agent-based Transformers have been widely adopted in recent reinforcement learning advances due to their demonstrated ability to solve complex tasks. However, the high computational complexity of Transformers often results in significant energy consumption, limiting their deployment in real-world autonomous systems. Spiking neural networks (SNNs), with their

Jonathan Klawitter, Alexei J. Drummond

Credible intervals and credible sets, such as highest posterior density (HPD) intervals, form an integral statistical tool in Bayesian phylogenetics, both for phylogenetic analyses and for development. Readily available for continuous parameters such as base frequencies and clock rates, the vast and complex space of tree topologies poses significant challeng

Shaoye Luo, Xinxin Fan, Quanliang Jing, Chi Lin

Aiming at resisting backdoor attacks in convolution neural networks and vision Transformer-based large model, this paper proposes a generalized and model-agnostic trigger-purification approach resorting to the classic Ising model. To date, existing trigger detection/removal studies usually require to know the detailed knowledge of target model in advance, ac

Zhipeng Yang, Junzhuo Li, Siyu Xia, Xuming Hu

We show that large language models (LLMs) exhibit an $\textit{internal chain-of-thought}$: they sequentially decompose and execute composite tasks layer-by-layer. Two claims ground our study: (i) distinct subtasks are learned at different network depths, and (ii) these subtasks are executed sequentially across layers. On a benchmark of 15 two-step composite

Christian Gouriéroux, Yang Lu

The Determinantal Point Process (DPP) is a parameterized model for multivariate binary variables, characterized by a correlation kernel matrix. This paper proposes a closed form estimator of this kernel, which is particularly easy to implement and can also be used as a starting value of learning algorithms for maximum likelihood estimation. We prove the cons

Nitish Shukla, Arun Ross

A face morph is created by combining two face images corresponding to two identities to produce a composite that successfully matches both the constituent identities. Reference-free (RF) demorphing reverses this process using only the morph image, without the need for additional reference images. Previous RF demorphing methods are overly constrained, as they

Matteo El-Hariry, Antoine Richard, Ricard M. Castan, Luis F. W. Batista

Autonomous robots must navigate and operate in diverse environments, from terrestrial and aquatic settings to aerial and space domains. While Reinforcement Learning (RL) has shown promise in training policies for specific autonomous robots, existing frameworks and benchmarks are often constrained to unique platforms, limiting generalization and fair comparis

Abigail Tadlock, Lori McCabe, Kerstin Nordstrom

We present results of LAMMPS Molecular Dynamics simulations of 2D gravity-driven flows of 30,000 soft uniform spheres through a vertical silo. We vary the gravitational field (g), elastic modulus of the particles (E), and silo outlet diameter (D). We present results on upwards pressure waves observed in the system. We compare our results with previous work o

Richard Šléher, William Brach, Tibor Sloboda, Kristián Košťál

Query routing, the task to route user queries to different large language model (LLM) endpoints, can be considered as a text classification problem. However, out-of-distribution queries must be handled properly, as those could be about unrelated domains, queries in other languages, or even contain unsafe text. Here, we thus study a guarded query routing prob

Michael Sullivan

We make the case for language models over logical forms (LFLMs), arguing that such models are more data-efficient than their textual counterparts. To that end, we introduce the Graph-based Formal-Logical Distributional Semantics (GFoLDS) prototype, a pretrained LM over graph representations of logical forms, as a proof-of-concept of LFLMs. Using GFoLDS, we p

Mahmuda Akhter Nishu, Chenyu Huang, Milad Roohi, Xin Zhong

Wind hazards such as tornadoes and straight-line winds frequently affect vulnerable communities in the Great Plains of the United States, where limited infrastructure and sparse data coverage hinder effective emergency response. Existing forecasting systems focus primarily on meteorological elements and often fail to capture community-specific vulnerabilitie

Zhihao Li, Yufei Wang, Heliang Zheng, Yihao Luo

High-fidelity 3D object synthesis remains significantly more challenging than 2D image generation due to the unstructured nature of mesh data and the cubic complexity of dense volumetric grids. Existing two-stage pipelines-compressing meshes with a VAE (using either 2D or 3D supervision), followed by latent diffusion sampling-often suffer from severe detail

Sayantan Ghosh, Magali Le Goff, Pinaki Chaudhuri, Kirsten Martens

We study the flow behavior and unjamming transition in dense assemblies of actively deforming particles that periodically change size, a process that we refer to as breathing. Using extensive molecular dynamics simulations and a complementary mesoscale elasto-plastic model, we explore how this internal activity influences plasticity and rheology. At low ampl

X. Xu, Y. -D. Liu, S. Shi, Y. -J. Wang

In this work, we propose a general protocol for distributed quantum computing that accommodates arbitrary unknown subroutines. It can be applied to scale up quantum computing through multi-chip interconnection, as well as to tasks such as estimating unknown parameters or processes for circuit depth reduction and constructing secure quantum cryptographic prot

Chun-Yi Kuan, Hung-yi Lee

Recent advancements in audio-aware large language models (ALLMs) enable them to process and understand audio inputs. However, these models often hallucinate non-existent sound events, reducing their reliability in real-world applications. To address this, we propose LISTEN (Learning to Identify Sounds Through Extended Negative Samples), a contrastive-like tr

Jakob Kienegger, Timo Gerkmann

Recent speaker extraction methods using deep non-linear spatial filtering perform exceptionally well when the target direction is known and stationary. However, spatially dynamic scenarios are considerably more challenging due to time-varying spatial features and arising ambiguities, e.g. when moving speakers cross. While in a static scenario it may be easy

Ondřej Ježil

Assuming that no family of polynomial-size Boolean circuits can factorize a constant fraction of all products of two $n$-bit primes, we show that the bounded arithmetic theory $\text{PV}_1$, even when augmented by the sharply bounded choice scheme $BB(\Sigma^b_0)$, cannot prove that every number has some prime divisor. By the completeness theorem, it follows

Imen Sayar, Nan Messe, Sophie Ebersold, Jean-Michel Bruel

Confidentiality, integrity, availability, authenticity, authorization, and accountability are known as security properties that secure systems should preserve. They are usually considered as security final goals that are achieved by system development activities, either in a direct or an indirect manner. However, these security properties are mainly elicited

Yen-Chen Wu, Feng-Ting Liao, Meng-Hsi Chen, Pei-Chen Ho

Transformers, the standard implementation for large language models (LLMs), typically consist of tens to hundreds of discrete layers. While more layers can lead to better performance, this approach has been challenged as far from efficient, especially given the superiority of continuous layers demonstrated by diffusion and flow-based models for image generat

Juliusz Ziomek, George Whittle, Michael A. Osborne

In spite of their prevalence, the behaviour of Neural Networks when extrapolating far from the training distribution remains poorly understood, with existing results limited to specific cases. In this work, we prove general results -- the first of their kind -- by applying Neural Tangent Kernel (NTK) theory to analyse infinitely-wide neural networks trained

Guillaume Vray, Devavrat Tomar, Xufeng Gao, Jean-Philippe Thiran

This paper introduces ReservoirTTA, a novel plug-in framework designed for prolonged test-time adaptation (TTA) in scenarios where the test domain continuously shifts over time, including cases where domains recur or evolve gradually. At its core, ReservoirTTA maintains a reservoir of domain-specialized models -- an adaptive test-time model ensemble -- that

Haishi Bai, Jozo Dujmovic, Jianwu Wang

As machine learning models and autonomous agents are increasingly deployed in high-stakes, real-world domains such as healthcare, security, finance, and robotics, the need for transparent and trustworthy explanations has become critical. To ensure end-to-end transparency of AI decisions, we need models that are not only accurate but also fully explainable an

Matthias Koch, Christian Nettersheim, Thorsten Horstmann, Michael Rademacher

This paper investigates the ongoing use of the A5/1 ciphering algorithm within 2G GSM networks. Despite its known vulnerabilities and the gradual phasing out of GSM technology by some operators, GSM security remains relevant due to potential downgrade attacks from 4G/5G networks and its use in IoT applications. We present a comprehensive overview of a histor

Biman Barua, M. Shamim Kaiser

Handling online travel agents globally requires efficient and flexible software solution architectures. When it needs to handle thousands of agents and billions of clients data globally. Microservices architecture is used to break down a large program into numerous, smaller services which can run individually and perform individual tasks. This paper analyses

Jingyun Chen, David Horowitz, Yading Yuan

Background: Deep learning has potential to improve the efficiency and consistency of radiation therapy planning, but clinical adoption is hindered by the limited model generalizability due to data scarcity and heterogeneity among institutions. Although aggregating data from different institutions could alleviate this problem, data sharing is a practical chal

Katharina Dudde, Mahmoud Elhajhasan, Guillaume Würsch, Julian Themann

In this work, we exemplify on a bulk silicon sample that Raman thermometry is capable of phonon mean free path (PMFP) spectroscopy. Our experimental approach is similar to the variation of different characteristic length scales $l_{c}$ during thermal reflectance measurements in the time or frequency domain and transient thermal grating spectroscopy. In place

Jiale Kang, Ziyin Yue, Qingyu Yin, Jiang Rui

Currently, most multimodal studies are based on large language models (LLMs) with quadratic-complexity Transformer architectures. While linear models like RNNs enjoy low inference costs, their application has been largely limited to the text-only modality. This work explores the capabilities of modern RNN architectures in multimodal contexts. We propose ModR

Luca Spagnuolo, Luca Colombo, Kapil Saha, Gabriel Giribaldi

This paper reports on Solidly-Mounted Bidimensional Mode Resonators (S2MRs) utilizing 30% Scandium-doped Aluminum Nitride on Silicon Carbide, operating near 16 GHz. Experimental results show mechanical quality factors up to 380, electromechanical coupling coefficients of 4%, and an overall Figure of Merit (FOM=Q *kt2) exceeding 15. Additionally, Q Bode calcu

Andrew S Johnson, William Winlow

Conventionally it is assumed that the nerve impulse is an electrical process based upon the observation that electrical stimuli produce an action potential as defined by Hodgkin Huxley (1952) (HH). Consequently, investigations into the computation of nerve impulses have almost universally been directed to electrically observed phenomenon. However, models of

Wenze Liu, Xiangyu Yue

To accelerate diffusion model inference, numerical solvers perform poorly at extremely small steps, while distillation techniques often introduce complexity and instability. This work presents an intermediate strategy, balancing performance and cost, by learning ODE integration using loss functions derived from the derivative-integral relationship, inspired

Thorsten Horstmann, Dominik Brunke, Tobias Kremeyer, Matthias Wilmes

In mobile network research, the integration of real-world components such as User Equipment (UE) with open-source network infrastructure is essential yet challenging. To address these issues, we introduce open5Gcube, a modular framework designed to integrate popular open-source mobile network projects into a unified management environment. Our publicly avail

Paloma Bengoechea, Sebastián Herrero, Özlem Imamoglu

We prove two of Kaneko's conjectures on the "values" $\mathrm{val}(w)$ of the modular $j$ function at real quadratic irrationalities: we prove the lower bound $\mathrm{Re}(\mathrm{val}(w))\geq \mathrm{val}\left(\frac{1+\sqrt{5}}{2}\right)$ for all real quadratics $w$ and the upper bound $\mathrm{Re}(\mathrm{val}(w))\leq \mathrm{val}\left(1+\sqrt{2}\right)$ f

Jun Cao, Jiyi Li, Ziwei Yang, Renjie Zhou

There has been growing interest in Multimodal Aspect-Based Sentiment Analysis (MABSA) in recent years. Existing methods predominantly rely on pre-trained small language models (SLMs) to collect information related to aspects and sentiments from both image and text, with an aim to align these two modalities. However, small SLMs possess limited capacity and kn

Amir Sagiv, Remy Kassem, Michael I Weinstein

We establish dispersive time-decay estimates for periodic Jacobi operators on the discrete half-line, $\N$. Specifically, we prove $t^{-1/2}$ decay in the weighted $\ell^\infty_{-1}$ norm for all such operators. For the global $\ell^1 \to \ell^\infty$ decay estimate, we show that $t^{-1/3}$ decay holds under a nondegeneracy condition on the discriminant. Alt

Ahmad Abdi, Gérard Cornuéjols, Daniel Dadush, Mahsa Dalirrooyfard

A set-system $S\subseteq \{0,1\}^n$ is cube-ideal if its convex hull can be described by capacity and generalized set covering inequalities. In this paper, we use combinatorics, convex geometry, and polyhedral theory to give exponential lower bounds on the size of cube-ideal set-systems, and linear lower bounds on their VC dimension. We then provide applicat

Mingyang Wang, Peng Liu, Jianping Yuan, Ang Li

We report the detection of an anti-glitch with a fractional frequency change of $\Delta\nu/\nu=-3.46(6)\times10^{-9}$ in the rotation-powered pulsar PSR J1835$-$1106 at MJD 55813, based on timing observations collected with the Nanshan 26-m and Parkes 64-m radio telescopes from January 2000 to July 2022. A comparison of the average pulse profiles within $\pm

Junyu Cao, Valentino Tosatti

We prove the optimal $C^{1,1}$ regularity of the volume function on the big cone of a projective manifold, and investigate its regularity when restricted to segments moving in ample directions.

LHCb collaboration, R. Aaij, A. S. W. Abdelmotteleb, C. Abellan Beteta

This article presents doubly differential measurements of the asymmetries in production rates between mesons containing a charm quark and those containing an anticharm quark in proton-proton collisions at a centre-of-mass energy of $\sqrt{s}=13.6$ TeV using data recorded by the LHCb experiment. The asymmetries of $D^0$, $D^+$ and $D_s^+$ mesons are measured

Stephanie Riedmüller, Annika Buchholz, Janina Zittel

The goal to decarbonize the energy sector has led to increased research in modeling and optimizing multi-energy systems. One of the most promising techniques for modeling (multi-)energy optimization problems is mixed-integer programming (MIP), valued for its ability to represent the complexities of integrated energy systems. While the literature often focuse

Murat Abdughani, Shao-Song Tang, Kadirya Tursun, Bin Zhu

Dark matter (DM) annihilation can be significantly enhanced through narrow resonances or the Sommerfeld enhancement effect, with both mechanisms potentially combining in a super-resonant annihilation process. In such scenarios, the conventional assumption that kinetic equilibrium persists until chemical decoupling may not hold, leading to substantial impacts

Daniele Agostini, Pietro Beri, Franco Giovenzana, Ángel David Ríos Ortiz

We study projective models of generalized Kummer fourfolds via O'Grady's theta groups and the classical Coble cubic. More precisely, we establish a duality between two singular models of the generalized Kummer fourfold of a Jacobian abelian surface. We also give projective models for singular Jacobian Kummer varieties of arbitrary dimension. Along the way, w