May 2025 arXiv papers — page 115

Pengfei Shi, Fei Shang, Haohua Du

Environmental sensing is an important research topic in the integrated sensing and communication (ISAC) system. Current works often focus on static environments, such as buildings and terrains. However, dynamic factors like rainfall can cause serious interference to wireless signals. In this paper, we propose a system called RainfalLTE that utilizes the down

Feng Li, Zhaoyue Wang, Enyuan Zhang, Mohammad Masum Billah

BEV-based 3D perception has emerged as a focal point of research in end-to-end autonomous driving. However, existing BEV approaches encounter significant challenges due to the large feature space, complicating efficient modeling and hindering effective integration of global attention mechanisms. We propose a novel modeling strategy, called InstanceBEV, that

James Glimm, Jarret Petrillo

Smooth solutions of the Navier-Stokes equation with smooth but otherwise unconstrained initial conditions are constructed, to solve the Millennium fluids problem in the positive. The smooth solutions are the mean values of general weak solutions and are alternately characterized as the entropy production minimizing solutions. The construction occurs in a fin

Zexin Pan

Recent advances in quasi-Monte Carlo integration demonstrate that the median of linearly scrambled digital net estimators achieves near-optimal convergence rates for high-dimensional integrals without requiring a priori knowledge of the integrand's smoothness. Building on this framework, we prove that the median estimator attains dimension-independent conver

Jihwan Lee, Kevin Huang, Kleanthis Avramidis, Simon Pistrosch

We present a model for predicting articulatory features from surface electromyography (EMG) signals during speech production. The proposed model integrates convolutional layers and a Transformer block, followed by separate predictors for articulatory features. Our approach achieves a high prediction correlation of approximately 0.9 for most articulatory feat

Matthew Raffel, Lizhong Chen

The Kolmogorov-Arnold Network (KAN) has been gaining popularity as an alternative to the multilayer perceptron (MLP) due to its greater expressiveness and interpretability. Even so, KAN suffers from training instability and being orders of magnitude slower due to its increased computational cost, limiting its applicability to large-scale tasks. Recently, the

Parikshit Bansal, Sujay Sanghavi

Fine-tuning a language model often results in a degradation of its existing performance on other tasks, due to a shift in the model parameters; this phenomenon is often referred to as (catastrophic) forgetting. We are interested in mitigating this, in settings where we only have access to the model weights but no access to its training data/recipe. A natural

Xiaofei Qi, Yuyang Pang, Jinchuan Hou

Multipartite quantum entanglement, as a core quantum resource, is fundamental to the advancement of quantum science and technology. In multipartite quantum systems, there are two kinds of quantum entanglement: $k$-nonseparability and $k$-partite entanglement. In this paper, we propose sufficient criteria for detecting $k$-nonseparability and $k$-partite enta

Faramarz Safi Esfahani, Ghassan Beydoun, Morteza Saberi, Brad McCusker

Metaheuristic algorithms are widely used for solving complex optimization problems, yet their effectiveness is often constrained by fixed structures and the need for extensive tuning. The Polymorphic Metaheuristic Framework (PMF) addresses this limitation by introducing a self-adaptive metaheuristic switching mechanism driven by real-time performance feedbac

Noah A. Rubin, Yeshaiahu Fainman

Polarization control and switchability are among the most unique features of "metasurfaces" as compared with diffractive optics technologies of the past. Here, we review how the polarization control afforded by the advent of present-day metasurfaces compares to diffractive elements of previous decades, clarifying from a functional perspective what is new, an

Jonah Guse, David Jiang, David Keating

We study the coupling of pairs of reverse plane partitions of the same shape by assigning a certain local interaction between the reverse plane partitions. We show that they are in bijection with a certain Yang-Baxter integrable colored vertex model. By utilizing the Yang-Baxter equation for this colored vertex model, we are able to compute the generating fu

Yu Pan, Yanni Hu, Yuguang Yang, Jixun Yao

Despite great advances, achieving high-fidelity emotional voice conversion (EVC) with flexible and interpretable control remains challenging. This paper introduces ClapFM-EVC, a novel EVC framework capable of generating high-quality converted speech driven by natural language prompts or reference speech with adjustable emotion intensity. We first propose EVC

Sk Tanzir Mehedi, Raja Jurdak, Chadni Islam, Gowri Ramachandran

Securing software supply chains is a growing challenge due to the inadequacy of existing datasets in capturing the complexity of next-gen attacks, such as multiphase malware execution, remote access activation, and dynamic payload generation. Existing datasets, which rely on metadata inspection and static code analysis, are inadequate for detecting such atta

Yecheng Zhang, Rong Zhao, Zimu Huang, Xinyu Wang

Generative artificial intelligence (GenAI) models are increasingly used for scientific data generation, yet their alignment with empirical knowledge in urban science remains unclear. We therefore ask whether generated urban data reproduce empirical regularities and support repeatable experiments. We introduce AI4US, a framework for evaluating data synthesis

Michael Röckner, Deng Zhang, Guohuan Zhao

We prove the existence and conditional uniqueness in the Krylov class for SDEs with singular divergence-free drifts in the endpoint critical Lorentz space $L^{\infty}(0,T; L^{d,\infty}(\mathbb{R}^d))$, $d \geqslant 2$, which particularly includes the $2$D Biot-Savart law. The uniqueness result is shown to be optimal in dimensions $d \geqslant 3$, by construc

Jun Yan Lee, Duo Wu, Xuanrui Guo, Jian Ding Tan

The fifth-generation (5G) network faces limitations in supporting emerging applications, such as artificial intelligence (AI), virtual reality (VR) and digital twins. To overcome these confines, sub-Terahertz (sub-THz) and Terahertz (THz) technologies are considered to be key enablers of effective 6G wireless communications, offering higher transmission spee

João Paulo C Moura, Lanna EB Lucchetti, Caio M Fernandes, Aline B Trench

The oxygen reduction reaction (ORR) plays a pivotal role in clean energy generation and sustainable chemical production, particularly in the synthesis of hydrogen peroxide (H\textsubscript{2}O\textsubscript{2}). In this study, WO\textsubscript{3}/Vulcan-XC72 electrocatalysts were synthesized and characterized for ORR applications, evaluating the WO\textsubsc

Boqing Deng

We proved that for any finite collection of sparse subgraphs $(D_m)_{m=1}^\ell$ of the complete graph $K_{2n}$, and a uniformly chosen perfect matching $R$ in $K_{2n}$, the random vector $(|E(R \cap D_m)|)_{m=1}^\ell$ jointly converges to a vector of independent Poisson random variables with mean $|E(D_m)|/(2n)$. We also showed a similar result when $K_{2n}$

Ruby A. Shi

In type-II superconductors, magnetic flux penetrates in the form of quantized vortices whose dissipative motion, driven by the Lorentz force, can degrade superconductivity. Understanding vortex dynamics in both homogeneous regions and near unavoidable structural defects is crucial for superconducting applications. This study examines a scenario in which a su

Xiao Lin, Zhining Liu, Ze Yang, Gaotang Li

Warning: This paper contains examples of harmful language and images. Reader discretion is advised. Recently, vision-language models have demonstrated increasing influence in morally sensitive domains such as autonomous driving and medical analysis, owing to their powerful multimodal reasoning capabilities. As these models are deployed in high-stakes real-wo

B. Farley, U. T. Ahmed, A. M. Hopkins, M. Cowley

We introduce a new approach to quantifying dust in galaxies by combining information from the Balmer decrement (BD) and the dust mass ($M_d$). While there is no explicit correlation between these two properties, they jointly probe different aspects of the dust present in galaxies. We explore two new parameters that link BD with $M_d$ by using star formation

Cedric Chauve, Louxin Zhang

The d-neighborhood of a word W in the Levenshtein distance is the set of all words at distance at most d from W. Generating the neighborhood of a word W, or related sets of words such as the condensed neighborhood or the super-condensed neighborhood has applications in the design of approximate pattern matching algorithms. It follows that bounds on the maxim

Zheng Wang, Jingshen Yan, Shu-Shen Lyu, Kaixuan Chen

Monolayer V2O3, a two-dimensional magnetic topological insulator with intrinsic ferromagnetic order and a nontrivial band gap, offers a promising platform for realizing quantum anomalous Hall (QAH) states. Using first-principles density functional theory calculations, we systematically investigate the impact of substrate selection on its electronic and topol

Qi Cheng, Licheng Liu, Qing Zhu, Runlong Yu

Evaluating ecological time series is critical for benchmarking model performance in many important applications, including predicting greenhouse gas fluxes, capturing carbon-nitrogen dynamics, and monitoring hydrological cycles. Traditional numerical metrics (e.g., R-squared, root mean square error) have been widely used to quantify the similarity between mo

J C Bray, U Kolb, S A Mills

With the planned launch of the PLAnetary Transit and Oscillation of stars (PLATO) satellite mission in 2026, an understanding of the stellar properties and spatial distribution of astrophysical false positives (FPs) is essential to ensure the limited ground-based spectroscopy resources are used efficiently to target the most likely genuine planetary transit

Mohammad Rubyet Islam

The pursuit of alpha returns that exceed market benchmarks has undergone a profound transformation, evolving from intuition-driven investing to autonomous, AI powered systems. This paper introduces a comprehensive five stage taxonomy that traces this progression across manual strategies, statistical models, classical machine learning, deep learning, and agen

Siddhant Bhambri, Upasana Biswas, Subbarao Kambhampati

Recent advances in reasoning-focused Large Language Models (LLMs) have introduced Chain-of-Thought (CoT) traces - intermediate reasoning steps generated before a final answer. These traces, as in DeepSeek R1, guide inference and train smaller models. A common but under-examined assumption is that these traces are both semantically correct and interpretable t

Manshi Limbu, Diwita Banerjee

Medical image captioning is a challenging task that requires generating clinically accurate and semantically meaningful descriptions of radiology images. While recent vision-language models (VLMs) such as BLIP, BLIP2, Gemini and ViT-GPT2 show strong performance on natural image datasets, they often produce generic or imprecise captions when applied to specia

Austin H. Cheng, Chong Sun, Alán Aspuru-Guzik

Generative models of 3D molecular structure play a rapidly growing role in the design and simulation of molecules. Diffusion models currently dominate the space of 3D molecule generation, while autoregressive models have trailed behind. In this work, we present Quetzal, a simple but scalable autoregressive model that builds molecules atom-by-atom in 3D. Trea

Yu-Bo Hou, Xiaoan Ai, Ruizhe You, Changchun Zhong

Quantum systems are inherently susceptible to noise -- a notorious factor that induces decoherence and limits the performance of quantum applications. To mitigate its detrimental effects, various techniques have been developed, including cryogenic cooling, bath engineering, and quantum error correction. In this paper, we demonstrate that by exploiting noise

Joshua Hinman

We prove that every 4-polytope is determined by its edge-polygon incidences, solving an open problem of Gr\"unbaum. For each $d \geq 3$, we show that not every $d$-polytope is determined by its $(d-3)$-skeleton and dual $(d-3)$-skeleton together, answering a question of Samper. In the simplicial realm, we prove that for $d \geq 4$ and $\lceil \frac{d}{2} \rc

Yongshuo Zong, Qin Zhang, Dongsheng An, Zhihua Li

This work presents a simple yet effective workflow for automatically scaling instruction-following data to elicit pixel-level grounding capabilities of VLMs under complex instructions. In particular, we address five critical real-world challenges in text-instruction-based grounding: hallucinated references, multi-object scenarios, reasoning, multi-granularit

Chris Cundy, Adam Gleave

As AI systems become more capable, deceptive behaviors can undermine evaluation and mislead users at deployment. Recent work has shown that lie detectors can accurately classify deceptive behavior, but they are not typically used in the training pipeline due to concerns around contamination and objective hacking. We examine these concerns by incorporating a

Yu Tan, Lei Li, Zi-Xuan Yang, Tao Huang

Spin states are pivotal in modulating the electrocatalytic activity of transition-metal (TM)-based compounds, yet quantitatively evaluating the activity-spin state correlation remains a formidable challenge. Here, we propose an 'activity index n' as a descriptor, to assess the activity of the spin states for the hydrogen evolution reaction (HER). n descripto

Tung Tran

We present twistor BV actions that encompasses many classically consistent bosonic holomorphic twistorial higher-spin theories with vanishing cosmological constant. Upon quantization, these actions are shown to be quantum consistent, i.e. no gauge anomaly, for some subclasses of twistorial higher-spin theories. Anomaly-free twistorial theories can be identif

Dinh Nam Pham, Eleftherios Avramidis

Sign Language Recognition (SLR) systems primarily focus on manual gestures, but non-manual features such as mouth movements, specifically mouthing, provide valuable linguistic information. This work directly classifies mouthing instances to their corresponding words in the spoken language while exploring the potential of transfer learning from Visual Speech

Junze Zhou, John Thomas, Thomas P. Darlington, Edward S. Barnard

Excitons in 2D material bubbles-nanoscale deformations in atomically thin materials, typically exhibiting a dome-like shape-are confined by the strain effect, exhibiting extraordinary emission properties, such as single photon generation, enhanced light emission, and spectrally tunable excitonic states. While the strain profiles of these bubbles have been ex

Zongyuan Shen, James P. Wilson, Shalabh Gupta

The paper presents a novel sample-based algorithm, called C*, for real-time coverage path planning (CPP) of unknown environments. C* is built upon the concept of a Rapidly Covering Graph (RCG), which is incrementally constructed during robot navigation via progressive sampling of the search space. By using efficient sampling and pruning techniques, the RCG i

Pieter E. Vermaas, Luca Possati, Zeki C. Seskir

The development of quantum technologies has been accelerating in the last decade, turning them into emerging technologies that need explicit attention by decision-makers at national funding agencies, companies and governments. In this paper we consider the governance of quantum internet, a new type of communication network developed for the promise that it c

Anthony Caine, Tom Needham, Clayton Shonkwiler

A Parseval frame is a spanning set for a Hilbert space which satisfies the Parseval identity: a vector can be expressed as a linear combination of the frame whose coefficients are inner products with the frame vectors. There is considerable interest within the signal processing community in the structural properties of the space of finite-dimensional Parseva

C. A. Lindstrøm, E. Adli, H. B. Anderson, P. Drobniak

Plasma accelerators promise greatly reduced size and cost for future particle-accelerator facilities. However, several challenges remain to be solved; in particular that of coupling beams between plasma stages (i.e., staging) without beam-quality degradation, and that of ensuring a stable acceleration process. In order to mature the technology, it is also ke

Viktor Holubec, Alexander Fischer, Giovanni Volpe, Frank Cichos

Self-propelled active particles exhibit delayed responses to environmental changes, modulating their propulsion speed through intrinsic sensing and feedback mechanisms. This adaptive behavior fundamentally determines their dynamics and self-organization in active matter systems, with implications for biological microswimmers and engineered microrobots. Here,

D. Cano-Ott, S. Cebrián, P. Dimitriou, M. Gromov

Understanding the radiogenic neutron production rate through the ($α$, n) reaction is crucial in many areas of physics, including dark matter searches, neutrino studies, and nuclear astrophysics. In addition to its relevance for fundamental research, the ($α$, n) reaction also plays a significant role in nuclear energy technologies, for example by contributi

Carolyn Chun, James Dylan Douthitt, Wayne Ge, Tony Huynh

We formulate a geometric version of the Erd\H{o}s-Hajnal conjecture that applies to finite projective geometries rather than graphs, in both its usual 'induced' form and the multicoloured form. The multicoloured conjecture states, roughly, that a colouring $c$ of the points of $\mathsf{PG}(n-1,q)$ containing no copy of a fixed colouring $c_0$ of $\mathsf{PG}

Hanlun Lei, Evgeni Grishin

Triple systems with low hierarchical structure are common throughout the Universe, including examples such as high-altitude lunar satellites influenced by the Earth, planetary satellites perturbed by the Sun, and stellar binaries affected by a supermassive black hole. In these systems, nonlinear perturbations are significant, making classical double-averaged

Gwyn Bellamy, Ruslan Maksimau, Travis Schedler

In this article we describe completely the singularities appearing in Calogero--Moser varieties associated (at any parameter) to the wreath product symplectic reflection groups. We do so by parameterizing the symplectic leaves in the variety, describing combinatorially the resulting closure relation and computing a transverse slice to each leaf. We also show

Guoheng Sun, Ziyao Wang, Bowei Tian, Meng Liu

As post-training techniques evolve, large language models (LLMs) are increasingly augmented with structured multi-step reasoning abilities, often optimized through reinforcement learning. These reasoning-enhanced models outperform standard LLMs on complex tasks and now underpin many commercial LLM APIs. However, to protect proprietary behavior and reduce ver

Subash Khanal, Srikumar Sastry, Aayush Dhakal, Adeel Ahmad

We present Sat2Sound, a unified multimodal framework for geospatial soundscape understanding, designed to predict and map the distribution of sounds across the Earth's surface. Existing methods for this task rely on paired satellite images and geotagged audio samples, which often fail to capture the full diversity of sound at a location. Sat2Sound overcomes

Bangti Jin, Jing Li, Yifeng Xu, Shengfeng Zhu

In this work, we develop an adaptive nonconforming finite element algorithm for the numerical approximation of phase-field parameterized topology optimization governed by the Stokes system. We employ the conforming linear finite element space to approximate the phase field, and the nonconforming linear finite elements (Crouzeix-Raviart elements) and piecewis

Zidi Xiong, Shan Chen, Zhenting Qi, Himabindu Lakkaraju

Large Reasoning Models (LRMs) have significantly enhanced their capabilities in complex problem-solving by introducing a thinking draft that enables multi-path Chain-of-Thought explorations before producing final answers. Ensuring the faithfulness of these intermediate reasoning processes is crucial for reliable monitoring, interpretation, and effective cont

Ryan Bowers, Richard Agbeyibor, Jack Kolb, Karen Feigh

We compare three methods of familiarizing a human with an artificial intelligence (AI) teammate ("agent") prior to operation in a collaborative, fast-paced intelligence, surveillance, and reconnaissance (ISR) environment. In a between-subjects user study (n=60), participants either read documentation about the agent, trained alongside the agent prior to the

Dimitris Roussis, Leon Voukoutis, Georgios Paraskevopoulos, Sokratis Sofianopoulos

We introduce Llama-Krikri-8B, a cutting-edge Large Language Model tailored for the Greek language, built on Meta's Llama 3.1-8B. Llama-Krikri-8B has been extensively trained on high-quality Greek data to ensure superior adaptation to linguistic nuances. With 8 billion parameters, it offers advanced capabilities while maintaining efficient computational perfo

Wanli Sun, Anton Ragni

Noise contrastive estimation (NCE) is a popular method for training energy-based models (EBM) with intractable normalisation terms. The key idea of NCE is to learn by comparing unnormalised log-likelihoods of the reference and noisy samples, thus avoiding explicitly computing normalisation terms. However, NCE critically relies on the quality of noisy samples

Jin Du, Li Chen, Xun Xian, An Luo

Reliable causal inference is essential for making decisions in high-stakes areas like medicine, economics, and public policy. However, it remains unclear whether large language models (LLMs) can handle rigorous and trustworthy statistical causal inference. Current benchmarks usually involve simplified tasks. For example, these tasks might only ask LLMs to id

Yonghoon Lee, Edgar Dobriban, Eric Tchetgen Tchetgen

We consider the problem of comparing a reference distribution with several other distributions. Given a sample from both the reference and the comparison groups, we aim to identify the comparison groups whose distributions differ from that of the reference group. Viewing this as a multiple testing problem, we introduce a methodology that provides exact, dist

Ruiquan Huang, Donghao Li, Chengshuai Shi, Cong Shen

This paper investigates a hybrid learning framework for reinforcement learning (RL) in which the agent can leverage both an offline dataset and online interactions to learn the optimal policy. We present a unified algorithm and analysis and show that augmenting confidence-based online RL algorithms with the offline dataset outperforms any pure online or offl

Shivani Shukla, Himanshu Joshi, Romilla Syed

The rapid adoption of Large Language Models(LLMs) for code generation has transformed software development, yet little attention has been given to how security vulnerabilities evolve through iterative LLM feedback. This paper analyzes security degradation in AI-generated code through a controlled experiment with 400 code samples across 40 rounds of "improvem

Celso Jorge Villas-Boas, Ciro Micheletti Diniz

We develop a quantum-optical framework demonstrating that thermal radiation can confine a significant portion of its energy in dark collective modes -- highly entangled photon states that, despite their photonic nature, remain decoupled from matter through standard electromagnetic interactions. In a system comprising $M$ thermal field modes, we show that onl

Zhaoyang Chen, Cody Fleming

Classifier free guidance has shown strong potential in diffusion-based reinforcement learning. However, existing methods rely on joint training of the guidance module and the diffusion model, which can be suboptimal during the early stages when the guidance is inaccurate and provides noisy learning signals. In offline RL, guidance depends solely on offline d

Avinash Patil

Software Quality Assurance (SQA) is critical for delivering reliable, secure, and efficient software products. The Software Quality Assurance Process aims to provide assurance that work products and processes comply with predefined provisions and plans. Recent advancements in Large Language Models (LLMs) present new opportunities to enhance existing SQA proc

Hainan Xu, Vladimir Bataev, Lilit Grigoryan, Boris Ginsburg

We propose Windowed Inference for Non-blank Detection (WIND), a novel strategy that significantly accelerates RNN-T inference without compromising model accuracy. During model inference, instead of processing frames sequentially, WIND processes multiple frames simultaneously within a window in parallel, allowing the model to quickly locate non-blank predicti

Andrea De Simone, Giovanna Turvani, Fabrizio Riente

Efficiently supporting remote firmware updates in Internet of Things (IoT) devices remains a significant challenge due to the limitations of many IoT communication protocols, which often make it impractical to transmit full firmware images. Techniques such as firmware partitioning have been introduced to mitigate this issue, but they frequently fall short, e

Li Ji-An, Hua-Dong Xiong, Robert C. Wilson, Marcelo G. Mattar

Large language models (LLMs) can sometimes report the strategies they actually use to solve tasks, yet at other times seem unable to recognize those strategies that govern their behavior. This suggests a limited degree of metacognition - the capacity to monitor one's own cognitive processes for subsequent reporting and self-control. Metacognition enhances LL

Siyuan Zhang, Yufei Zhang, Junlin Lyu, Sunil K. Agrawal

This paper presents the development of a wearable ankle rehabilitation robot based on a 3-RRR spherical parallel mechanism (SPM) to support multi-DOF recovery through pitch, roll, and yaw motions. The system features a compact, ergonomic structure designed for comfort, safety, and compatibility with ankle biomechanics. A complete design-to-dynamics pipeline

Jacob Kleiman, Kevin Frank, Joseph Voyles, Sindy Campagna

Simulations, although powerful in accurately replicating real-world systems, often remain inaccessible to non-technical users due to their complexity. Conversely, large language models (LLMs) provide intuitive, language-based interactions but can lack the structured, causal understanding required to reliably model complex real-world dynamics. We introduce ou

Drona Khurana, Anish Thilagar, Dhamma Kimpara, Rafael Frongillo

The statistical consistency of surrogate losses for discrete prediction tasks is often checked via the condition of calibration. However, directly verifying calibration can be arduous. Recent work shows that for polyhedral surrogates, a less arduous condition, indirect elicitation (IE), is still equivalent to calibration. We give the first results of this ty

Caio Machado Fernandes, Eduardo M. Rodrigues, Odivaldo C. Alves, Flavio Garcia

Ammonia poses significant environmental challenges due to its role in water pollution, contributing to eutrophication and several detrimental environmental and ecological issues. Addressing the efficient removal or conversion of ammonia is, therefore, critical. Among various methods, the ammonia electro-oxidation reaction stands out due to its potential for

Kaan Kale, Kyle Mylonakis, Jay Roberts, Sidhartha Roy

In this work, we consider an inversion attack on the obfuscated input embeddings sent to a language model on a server, where the adversary has no access to the language model or the obfuscation mechanism and sees only the obfuscated embeddings along with the model's embedding table. We propose BeamClean, an inversion attack that jointly estimates the noise p

Runchu Tian, Xueqiang Xu, Bowen Jin, SeongKu Kang

Scientific retrieval is essential for advancing scientific knowledge discovery. Within this process, document reranking plays a critical role in refining first-stage retrieval results. However, standard LLM listwise reranking faces challenges in the scientific domain. First-stage retrieval is often suboptimal in the scientific domain, so relevant documents a

J. S. Urquhart, C. Koenig, D. Colombo, A. Karska

The Outer Galaxy High-Resolution Survey (OGHReS) covers 100 square degrees ($180^\circ < \ell < 280^\circ$) in the (2--1) transitions of three CO-isotopologues. We use the spectra to refine the velocities and physical properties to 6706 \higal\ clumps located in the OGHReS region. In a previous paper, we analysed 3584 clumps between $\ell = 250^\circ$ and $2

Jeffrey Lai, Anthony Bao, William Gilpin

Chaotic systems are intrinsically sensitive to small errors, challenging efforts to construct predictive data-driven models of real-world dynamical systems such as fluid flows or neuronal activity. Prior efforts comprise either specialized models trained on individual time series, or foundation models trained on vast time series databases with little underly

Devendra Parkar, Anya Chaturvedi, Joshua J. Daymude

We present the first unsupervised learning model for Maximum-Independent-Set (MaxIS) in dynamic graphs where edges change over time. Our method combines structural learning from graph neural networks (GNNs) with a learned distributed update mechanism that, given an edge addition or deletion event, modifies nodes' internal memories and infers their MaxIS memb

Caleb Traxler, Minh Ton, Nameer Ahmed, Sasha Prostota

This paper presents a detailed mathematical investigation into the dynamics of COVID-19 infections through extended Susceptible-Infected-Recovered (SIR) and Susceptible-Exposed-Infected-Recovered (SEIR) epidemiological models. By incorporating demographic factors such as birth and death rates, we enhance the classical Kermack-McKendrick framework to realisti

Denny M. Oliveira, Eftyhia Zesta, Katherine Garcia-Sage

The exponential increase of low-Earth orbit (LEO) satellites in the past 5 years has brought into intense focus the need for reliable monitoring and reentry prediction to safeguard from space collisions and ground debris impacts. However, LEO satellites fly within the upper atmosphere region that exerts significant drag forces to their orbits, reducing their

Mohammad Nooraiepour, Karol M. Dąbrowski, Mohammad Masoudi, Szymon Kuczyński

Emerging carbon capture and storage (CCS) markets face critical challenges in developing systematic methodologies to assess geological CO2 storage potential under conditions of limited data availability, evolving regulatory frameworks, and nascent infrastructure development. This study establishes an assessment framework designed for lower-maturity CCS regio

Tapan Srivastava, Jacopo Tagliabue, Ciro Greco

Due to the variety of its target use cases and the large API surface area to cover, a data lakehouse (DLH) is a natural candidate for a composable data system. Bauplan is a composable DLH built on "spare data parts" and a unified Function-as-a-Service (FaaS) runtime for SQL queries and Python pipelines. While FaaS simplifies both building and using the syste

Yousef Shakiba, Henry Sinclair-Banks, Georg Zetzsche

In many kinds of infinite-state systems, the coverability problem has significantly lower complexity than the reachability problem. In order to delineate the border of computational hardness between coverability and reachability, we propose to place these problems in a more general context, which makes it possible to prove complexity dichotomies. The more ge

Xiang Zhang, Chenlin Xu, Zhouxiao Lu, Haonan Wang

Quantifying similarity between population spike patterns is essential for understanding how neural dynamics encode information. Traditional approaches, which combine kernel smoothing, PCA, and CCA, have limitations: smoothing kernel bandwidths are often empirically chosen, CCA maximizes alignment between patterns without considering the variance explained wi

Davide Costa, Francesco Fallavollita, Hubert Kroha, Oliver Kortner

Resistive Plate Chambers (RPCs) featuring 1 mm gas volumes combined with high-pressure phenolic laminate (HPL) electrodes provide excellent timing resolution down to a few hundred picoseconds, along with spatial resolution on the order of a few millimeters. Thanks to their relatively low production cost and robust performance in high-background environments,

Muhammad Monjurul Karim, Yan Shi, Shucheng Zhang, Bingzhang Wang

Roadway safety and mobility remain critical challenges for modern transportation systems, demanding innovative analytical frameworks capable of addressing complex, dynamic, and heterogeneous environments. While traditional engineering methods have made progress, the complexity and dynamism of real-world traffic necessitate more advanced analytical frameworks

Lei Zhu, Bernhard Rauer, Hilton B. De Aguiar, Sylvain Gigan

Two-photon (2P) microscopy is a powerful technique for deep-tissue fluorescence imaging; however, tissue scattering limits its effectiveness for depth imaging using conventional approaches. Despite typical strategies having been put forward to extend depth imaging capabilities based on wave-front shaping (WFS), computationally recovering images remains a sig

Satoshi Kondo

Surgical phase recognition from video is a technology that automatically classifies the progress of a surgical procedure and has a wide range of potential applications, including real-time surgical support, optimization of medical resources, training and skill assessment, and safety improvement. Recent advances in surgical phase recognition technology have f

Jorge Suárez-Recio, Pablo M. Piaggi, Francisco J. Domínguez-Gutiérrez, Raquel Gonzalez-Arrabal

Nanostructured tungsten has been reported as a possible alternative plasma-facing material due to its potential ability to self-heal radiation-induced defects, a property that is attributed to its high density of grain boundaries (GB). Here, we study the initial stages of self-healing at tungsten interfaces with molecular dynamics simulations driven by a mac

Francisco Fuica, Stefan Volkwein

We analyze a pointwise tracking multiobjective optimal control problem subject to the Poisson problem and bilateral control constraints. To approximate Pareto optimal points and the Pareto front numerically, we consider two different finite element-based scalarization techniques, namely the weighted-sum method and the reference point method, where in both me

Andrew Nam, Declan Campbell, Thomas Griffiths, Jonathan Cohen

Neural networks are powerful tools for cognitive modeling due to their flexibility and emergent properties. However, interpreting their learned representations remains challenging due to their sub-symbolic semantics. In this work, we introduce a novel probabilistic framework for interpreting latent task representations in neural networks. Inspired by Bayesia

Gaspard Goupy, Pierre Tirilly, Ioan Marius Bilasco

Direct training of Spiking Neural Networks (SNNs) on neuromorphic hardware can greatly reduce energy costs compared to GPU-based training. However, implementing Backpropagation (BP) on such hardware is challenging because forward and backward passes are typically performed by separate networks with distinct weights. To compute correct gradients, forward and

Chenning Yu, Sicun Gao

We introduce a novel resampling criterion using lift scores, for improving compositional generation in diffusion models. By leveraging the lift scores, we evaluate whether generated samples align with each single condition and then compose the results to determine whether the composed prompt is satisfied. Our key insight is that lift scores can be efficientl

Caio Machado Fernandes, Joao Paulo C. Moura, Aline B. Trench, Odivaldo C. Alves

The current study illustrates the successful synthesis of Ce$_{1.0}$Mn$_{0.9}$Co$_{0.1}$ nanoparticles, characterized through XRD, EPR, magnetization curves, and TEM/HRTEM/EDX analyses. These nanoparticles were then loaded into the carbon Vulcan XC72 and the carbon Printex L6 matrices in varying amounts (1, 3, 5, and 10% w/w) via wet impregnation method to f

Shane Bergsma, Nolan Dey, Gurpreet Gosal, Gavia Gray

Efficient LLM pre-training requires well-tuned hyperparameters (HPs), including learning rate $\eta$ and weight decay $\lambda$. We study scaling laws for HPs: formulas for how to scale HPs as we scale model size N, dataset size D, and batch size B. Recent work suggests the AdamW timescale, $\tau = B/(\eta \lambda D)$, should remain constant across training

Andrew Nam, Henry Conklin, Yukang Yang, Thomas Griffiths

We present causal head gating (CHG), a scalable method for interpreting the functional roles of attention heads in transformer models. CHG learns soft gates over heads and assigns them a causal taxonomy - facilitating, interfering, or irrelevant - based on their impact on task performance. Unlike prior approaches in mechanistic interpretability, which are hy

Hajime Otsuka, Yutaka Sakamura

We investigate the response of the Kaluza-Klein (KK) mass spectrum to various deformations of the rugby-ball background in 6-dimensional supergravity. We derived the mode equations that contain the 3-dimensional scale factor and the lapse function. By solving these, we numerically evaluate the KK masses for a bulk scalar and a spinor when the background has

Andrei Poblaguev

The analysis of RHIC hydrogen gas jet target polarimeter measurements of transverse analyzing powers $A_\text{N}(t)$ in proton-nucleus scattering requires accurate Coulomb corrections to both spin-flip and non-flip amplitudes. These corrections must cover a wide range of nuclear charges $Z$ and form factor slopes, with flexibility to vary form factors during

Mehdi Ghorbani, Fatemeh Alikhani, Saad Varsaie

The Euler-Poincare characteristic, or Euler characteristic in short, is a fundamental topological invariant of compact manifolds that plays a crucial role in a variety of geometric and topological situations. From this point of view, we tried to expand on this important concept in supergeometry. In this article, we introduce the Euler-Poincare characteristic

Mithun Roy, Tianyi Zeng, Zhenyang Xiao, Chao Dong

Broadband light sources with well-defined spectral structures are vital for science and technology. However, the evenly spaced lines of frequency combs represent only a small subset of all possible structured white-light sources. We demonstrate liquid combs: optical states that preserve spectral equidistance but lack temporal stability. By engineering the ga

Etienne Gauthier, Francis Bach, Michael I. Jordan

We introduce $\textit{Backward Conformal Prediction}$, a method that guarantees conformal coverage while providing flexible control over the size of prediction sets. Unlike standard conformal prediction, which fixes the coverage level and allows the conformal set size to vary, our approach defines a rule that constrains how prediction set sizes behave based

Jon M. Miller, Ehud Behar, Hisamitsu Awaki, Ann Hornschemeier

Up to 40% of galaxies in the local universe host a low-luminosity active galactic nucleus (LLAGN), making it vital to understand this mode of black hole accretion. However, the presence or absence of Seyfert-like geometries - an accretion disk close to the black hole, an optical broad line region (BLR), and a molecular torus - remains uncertain owing to the

Yong Si, Junyi Fan, Li Sun, Shuheng Chen

Traumatic Brain Injury (TBI) is a major contributor to mortality among older adults, with geriatric patients facing disproportionately high risk due to age-related physiological vulnerability and comorbidities. Early and accurate prediction of mortality is essential for guiding clinical decision-making and optimizing ICU resource allocation. In this study, w

Abhinav Rajvanshi, Pritish Sahu, Tixiao Shan, Karan Sikka

Adaptive collaboration is critical to a team of autonomous robots to perform complicated navigation tasks in large-scale unknown environments. An effective collaboration strategy should be determined and adapted according to each robot's skills and current status to successfully achieve the shared goal. We present SayCoNav, a new approach that leverages larg

Jiong Qiu, Rhiannon Fleming

Imaging observations of the solar lower atmosphere by the Atmosphere Imaging Assembly (AIA) have been mostly used as the context, and their quantitative information has been much less explored. The chromosphere responds rapidly to energy release by magnetic reconnection during flares. Furthermore, a flare is a collection of multiple energy release events tha

Adrian Clingher, Andreas Malmendier, Tony Shaska

The paper discusses geometric and computational aspects associated with $(n,n)$-isogenies for principally polarized Abelian surfaces and related Kummer surfaces. We start by reviewing the comprehensive Theta function framework for classifying genus-two curves, their principally polarized Jacobians, as well as for establishing explicit quartic normal forms fo