arXiv papers from May 2026

Boris Horvat, Alen Orbanić, Iztok Kavkler

Horvat et al. (J. Math. Chem., 2014) showed that the rank of the $n$-th Hadamard power $D^{(n)}$ of a Euclidean distance matrix satisfies $\operatorname{rank}D^{(n)} \le R_d^n$, and proved that the inequality is strict whenever an annihilating polynomial exists. The converse - that the absence of annihilating polynomials forces $\operatorname{rank}D^{(n)} =

Xinze Zhang

Dense volumetric reconstruction of cloud microphysical fields from sparse ground-based instruments remains an open problem, largely because the available measurements are heterogeneous in both modality and spatial coverage. We present AtmoFuseNet, a framework that fuses multi-view sky camera imagery with millimeter-wave cloud radar and ceilometer observation

David Rico Menéndez

This technical note surveys latency contributors that matter when deploying edge-enabled applications and proximity services in 5G. Rather than proposing a new mechanism, we focus on building a reproducible latency catalogue grounded on: (i) theoretical targets and ranges reported in standards, and (ii) representative empirical measurements from the literatu

Song-Duo Ma, Chu-Yun Chen, Bang-An Li, Pin-Yu Chen

Large Language Models (LLMs) are reshaping recommender systems by enabling more semantic, generative, and interactive recommendation pipelines. However, this shift also introduces new fairness challenges, as biases may arise from pretrained knowledge, prompts, generated explanations, decoding strategies, and feedback loops. This survey provides a systematic

Marwan Dhuheir, Thang X. Vu, Symeon Chatzinotas

Industrial 6G networks require ultra-reliable, low-latency, and energy-efficient connectivity in dynamic and blockage-prone environments, where conventional terrestrial deployments often fail to ensure stable coverage. Hence, in this paper, we propose a RIS-enabled Open-RAN framework for integrated terrestrial/non-terrestrial (TN/NTN) industrial 6G networks,

Ratun Rahman

Emerging 6G and edge-intelligent networks require effective and balanced routing algorithms among varied and spatially distributed devices. Existing federated routing systems often prioritize aggregate latency or throughput above fairness and the underlying geometric structure of network topologies. This paper describes Geo-FairFed, a geometric fairness-awar

John Regan, Jon Stevens, Brian Martin

This paper presents an empirical analysis of AbbVie Intelligence's measurable impact on employee work activities across 192 distinct occupations in 2024 and 2025. Drawing on 598,744 de-identified AI conversations classified according to the O*NET Intermediate Work Activity (IWA) taxonomy, we compute occupation-level AI Applicability Scores that quantify

Mingi Kang

The shift from Convolutional Neural Networks to Transformers has reshaped computer vision, yet these two architectural families are typically viewed as fundamentally distinct. Convolutional Neural Networks are defined by spatially local convolution operations, while Transformers rely on global self-attention. We argue that convolution and self-attention, des

Xinze Zhang

Video anomaly detection in surveillance settings must balance detection accuracy against real-time throughput, a tension that existing methods address either through stronger feature extractors or more efficient architectures, but rarely both. We present VigilFormer, a unified framework that combines deformable spatio-temporal attention with causal temporal

Durga Sandeep Saluru

We study multiple-choice video question answering on the ImplicitQA benchmark, where the correct answer is never explicitly shown but must be inferred from off-screen events, line-of-sight cues, causal structure, and cross-shot spatial layout. On this benchmark a single frontier video LLM already operates near its accuracy ceiling, and we observe that conven

Dénes Toth, George Ambroladze, Edwin Sundberg, Ali Beikmohammadi

Urban traffic signal control at IoT-instrumented intersections must remain effective under sensor occlusion, weather attenuation, and nonstationary demand. Conventional controllers degrade under these conditions, and learned policies remain difficult to audit. To address these challenges, we propose an active inference controller for a four-arm signalized in

Yutao Deng, Jianjun Gao, Weichen Wang

The multiperiod mean-variance (MV) portfolio optimization serves as a vital expansion of Markowitz's static MV portfolio selection framework. Just like its static counterpart, the multiperiod MV portfolio remains susceptible to estimation errors. We propose a reference-regulated multiperiod mean-variance (RRMV) framework that penalizes deviations from a

Jung-Hoon Cho, Cathy Wu

Transit network design depends not only on the optimization algorithm but also on who shows up to the public hearing. Current practice often collects one-directional comments from self-selected attendees, leaving participant mix as an uncontrolled source of outcome variation. We present AGORA, a framework that holds the network, demand, and solver fixed whil

Guisong Liu, Pengfei Wei, Jainsong Zhang, Martin Dresler

Mobile sleep staging serves as a foundational infrastructure for in-home sleep monitoring and closed-loop modulation. But existing sequential models such as RNNs and Transformers are computationally expensive for mobile deployment. In this paper, we propose Random Attention (RA), a lightweight temporal modeling module based on fixed random projections, which

Axel Laborieux, Christos Sourmpis, Juan Gabriel Kostelec, Qinghai Guo

The Softmax Attention operation in Transformer language models has a quadratic complexity in the sequence length and a growing state size in the form of KV cache, which becomes a bottleneck in long context scenarios. To overcome this limitation, alternative architectures with linear complexity and finite state size have been introduced, such as State-Space M

Yunhao Zhang, Ruiying Qi, Jiale Zheng, Jianfeng Zhang

While Next-Token Prediction (NTP) has unified LLM pretraining, its adaptation to unbounded, continuous time series (TS) remains open. To bridge the gap, we introduce UniTok, a universal tokenizer that transforms TS into discrete tokens, and UniTok-FM, a foundation model pretrained via NTP on these tokens. UniTok-FM is a general-purpose foundation model that

Xinze Zhang

Non-alcoholic fatty liver disease (NAFLD) affects roughly 25% of global adults, posing substantial hepatic and cardiovascular risks. Yet, population-level screening tools remain inadequate. We present Method, a machine-learning framework for NAFLD risk prediction coupling gradient-boosted decision trees with conformal prediction to yield calibrated, distribu

Yingxuan Zhuang, Jingxiao Yang, Miao Pan, Cheng Tan

MLLMs frequently hallucinate objects inconsistent with visual inputs. This issue is typically attributed to the over-reliance on language priors, which can override the visual context. Recent training-free decoding strategies address this by penalizing language priors. However, these methods overlook the dual nature of language priors, where they can be both

Orfeu Bertolami

We argue that an airborne proton accelerator is an interesting tool for weather control. Following the findings of the CLOUD experiment at CERN, one expects that a beam of protons, likewise cosmic rays and other aerosols, can enhance the formation of low-altitude clouds, allow for tailor made cooling of overheated areas and induce the precipitation of high-a

Lecheng Yan, Yichong Zhang, Ben Pan, Xiaoyu Zheng

Editing a long-form video from heterogeneous footage requires more than selecting clips: an agent must preserve narrative intent across material preparation, timeline construction, post-production, and revision while leaving enough evidence to diagnose failures. We present \textbf{Crayotter}, an open-source multimodal multi-agent system for prompt-driven vid

Xiongri Shen, Zhenxi Song, Jiaqi wang, Yi Zhong

Multimodal neuroimaging fusion of functional MRI (fMRI) and diffusion tensor imaging (DTI) provides complementary information for cognitive impairment analysis, but remains challenged by heterogeneous feature spaces and misaligned representations. We propose \textit{NeuroAlign}, a hierarchical framework for structured multimodal fusion. It introduces (1) \te

Jin Hyung Cho

The exact ground state of the two-dimensional Hubbard model is critical for understanding cuprate superconductivity. Previous numerical studies on narrow cylinders found insulating, static stripes that inherently suppress superconductivity. Here, using constrained-path auxiliary-field quantum Monte Carlo on isotropic lattices up to $24 \times 24$ sites, we s

Spoorthi M, Suja Palaniswamy

Accurate classification of nuclei subtypes in histopathology images is critical for downstream tasks including tumor grading, immune infiltrate quantification, and prognosis prediction. Existing approaches rely on either convolutional or transformer-based encoders in isolation, limiting their ability to simultaneously capture fine-grained local texture and l

Jared Fernandez, Clara Na, Yonatan Bisk, Constantine Samaras

Proper accounting of the energy requirements and environmental impact of artificial intelligence (AI) systems is necessary for researchers, developers, policy makers, and users to assess the barriers to building systems at scale. With the growing complexity of pipelines and underlying infrastructure needed to develop and deploy AI systems, previous approache

Huy Nghiem, Sy-Tuyen Ho, Sarah Wiegreffe, Hal Daumé

Emergent misalignment (EM) occurs when narrow finetuning causes a model to behave dangerously outside the finetuning task. Standard training signals can miss this shift, making reliable detection costly if it depends on repeated behavioral evaluation. We ask whether emergent misalignment can instead be detected from internal representations during finetuning

Mehmet Ergezer

We describe a reproducible computational framework for upper-bound searches on r_3(N), the maximum size of a 3-term-arithmetic-progression-free subset of [1,N]. The framework combines a verified lower-bound witness, endpoint forcing, depth-d witness-variable splitting, OEIS A003002 window-cardinality pruning, and recursive refinement of timed-out subproblems

Shuoyao Wang, Weisheng Xie, Mingze Gong, Suzhi Bi

Deep learning-based joint source-channel coding has recently demonstrated strong potential for semantic communication (SemComm). However, most existing approaches focus on optimizing visual-fidelity metrics, which can lead to reduced perceptual quality. Generative model-based SemComm leverages rich prior knowledge from large-scale pre-training to enhance per

Yotam Svoray

In this note, we obtain bounds for the $F$-pure threshold of isolated hypersurface singularities over an algebraically closed field of positive characteristic in terms of classical singularity invariants, notably the Milnor and Tjurina numbers. For curve singularities, we show that the $F$-pure threshold admits bounds, and often explicit computations, in ter

Mingze Gong, Jia Yan, Shuoyao Wang

Aerial semantic relay communications (SRC) employs an unmanned aerial vehicle (UAV) equipped with a semantic encoder as a relay, which not only extends the data acquisition coverage of the base station (BS) from resource-limited sensing device (SD) but also enhances communication efficiency through semantic feature transmission over the UAV-BS link. Existing

Patrick BarAvi

The point-charge approximation is one of the most successful idealizations in molecular biophysics, but it becomes strained in strong fields, confined geometries, and crowded aqueous environments. We develop a minimal Extended Structural Dynamics (ESD) model in which charged entities are treated as finite, deformable objects with an internal breathing mode r

Dipanjan Mitra

Wideband radio interferometers increasingly rely on analog quadrature hybrids to synthesize circular polarization from linear feeds. These systems are typically calibrated under the assumption that instrumental polarization leakage can be represented as a static complex offset, independent of parallactic angle. In this work, we demonstrate that this assumpti

James H. Atwater, David Lambert, Yuri Rostovtsev

As has been shown by multiple authors in recent decades, it is possible to reformulate various portions of the standard model over the ring of complex quaternions. In this paper, we utilize a complex quaternion spin representation of the spacetime algebra to derive the magnetic moments of standard model fermions and the $W^\pm$ boson. The moments calculated

Valentin Bergeron, Karolina Gorna

Enforcing invariants in safety-critical systems is increasingly urgent as AI-generated code becomes widespread. Unfortunately, the runtimes required to support high-level specification languages are too large for most embedded targets. In this article, we show how formally verified firmware is achievable today. We built Encore!, a bare-metal Continuation Pas

Thomas Izgin, Andreas Meister, Isaac Azure

In this study, five distinct COVID-19 models developed in different countries, each designed to reflect the prevailing epidemiological condition at the time of formulation, are examined. The models are reformulated while still maintaining their original structure, using their common transmissions from one compartment to the other. Modified Patankar-Runge-Kut

Alejandro Frank, Laurence A. Jacobs

We introduce temporal matrix scale invariance (tMSI), a mathematical structure for the two-time correlation kernel of a multivariate observable. A kernel $C(t,t')$ satisfies tMSI of order $α$ if $C(kt, kt') = k^{-α}C(t,t')$ for all $k>0$; this condition holds near a tipping point, where the divergence of the coherence time produces temporal scale

Slava G. Turyshev

We present a physics-based framework that runs from fundamental interactions and constants to biospheres, using a sequence of quantitative nonequilibrium thresholds ("gates"). Each gate is an inequality in measurable variables-free-energy flux, reaction-transport rates, replication fidelity, coding capacity, ecological closure, and climate feedback g

Blaž Bertalanič, Carolina Fortuna

Inference-time multi-agent LLM scaling lacks a shared unit: counting nominal agents conflates cost with independent evidence. We derive a two-parameter scaling law $R(N) = N_\text{eff}/N = 1/(1+c(N-1)N^{-β})$ where the regime exponent $β$ classifies any configuration into one of three asymptotic regimes -- hard-ceiling at $1/c$ ($β= 0$), sublinear at $N^β/c$

Donghwan Lee

Periodic target updates in Q-learning and soft target updates in actor-critic methods are empirically well established stabilization mechanisms, but their precise theoretical explanation is still incomplete. This paper gives a rigorous and exact analysis of these mechanisms for Q-learning with linear function approximation (linear Q-learning) using the exact

Eliot Krzysztof Jones, Mateusz Dziemian, Matt Fredrikson, J Zico Kolter

Agentic scaffolds have dramatically improved LLM performance on complex, long-horizon tasks, yielding both broad benefits and amplified risks in domains like cybersecurity. Existing benchmarks for AI agents in cybersecurity focus mainly on measuring proficiency--how effectively agents can complete offensive security tasks--but neglect a critical question: wh

Chengliang Liu, Liangbo Ning, Yujuan Ding, Wenqi Fan

Retrieval-Augmented Generation (RAG)-enhanced LLM systems, while powerful, introduce substantial inference costs due to the inclusion of an extra multi-stage pipeline that dynamically retrieves and synthesizes information from external knowledge sources. This high operational cost exposes a critical vulnerability to Inference Cost Attacks (ICAs). However, ex

Chenshuang Zhang, Kyeong Seon Kim, Chengxin Liu, Tae-Hyun Oh

Despite the success of audio-visual large-language models (LLMs), they can produce plausible but ungrounded outputs, termed hallucination. Existing benchmarks focus on environmental sounds (e.g., dog barking) to indicate event occurrence. In contrast, human speech carries fundamentally different, rich semantics and temporal structures, yet it remains unexplo

Yifan Wang

Interactive driving exposes a failure mode that is easy to miss in rule-aware autonomous-driving stacks: a hard-rule margin can be negative for an ego candidate even though a small lawful accommodation by a non-priority agent would restore feasibility. Existing rulebooks, shields, and reachability filters are strong at vetoing unsafe actions, while predictio

Huanli Gong, Zhipeng Wei, Yu Fu, Haz Sameen Shahgir

Multi-turn jailbreak attacks pose a growing threat to large language model (LLM) safety because they exploit feedback from auxiliary judge models to iteratively refine prompts toward harmful goals. Existing defenses largely detect or block unsafe content at individual turns or at the final response, leaving the judge-driven refinement loop intact and allowin

Spoorthi M, Suja Palaniswamy

We identify and resolve a previously unreported failure mode in TensoRF when applied to X-ray attenuation fields: the default density shift of -10, originally introduced for RGB scene reconstruction, suppresses density gradients and prevents sparse-view medical reconstruction regardless of learning rate or regularization strategy. Setting the density shift t

Yuejiao Wang, Zihao Ji, Pengfei Cai, Xu Li

Recent advances in neural song generation have enabled high-quality synthesis from lyrics and global textual prompts. However, most systems fail to model temporally varying attributes of songs, severely limiting fine-grained control over musical structure and dynamics. To address this, we propose SegTune, a Diffusion Transformer-based framework enabling stru

Kevin Hernández, Marcos Orellana-Iraheta, William Larín-Escobar

We derive and analyze the exact solutions of the inverted Dirac-Moshinsky oscillator (IDMO) in $(1+1)$ dimensions, obtained from the standard model via the substitution $p \to p + imωβx$. The upper spinor component satisfies a Weber equation with complex spectral parameter $λ= (E^2-m^2)/(2mω)+i/2$, whose solutions are parabolic cylinder functions $D_ν(ξ)$ wi

Nikhil Kothari, Saksham Samdani, Ritam Mallick, Praveen Gupta

Semantic retrieval in e-commerce must handle short, noisy, and colloquial queries over large product catalogs with fine-grained attribute distinctions. We present a Siamese LLM dual-encoder trained through a two-stage pipeline: contrastive learning with a false-negative margin mask to prevent penalization of near-duplicate products, followed by Relative Odds

Siyi Chen, Weiming Zhuang, Jingtao Li, Lingjuan Lv

Unified vision-language models (VLMs) integrate visual understanding and visual generation within a single autoregressive backbone, but their joint training is computationally expensive and largely overlooked from an efficiency perspective. In this work, we study the feasibility and limits of token-reduction-based acceleration for unified VLM training. Throu

Bole Ma, Jan Eitzinger, Harald Köstler, Gerhard Wellein

Frontier LLMs increasingly decide what a query attends to with a sparse-attention indexer that picks a few KV-cache blocks per query: attention's unit is now a small, reusable chunk. Agentic workloads hammer it: many sub-agents query one large codebase, reusing the same blocks. When that corpus outgrows one GPU it is partitioned across instances, so a qu

Silvio Franz, Roberto Mulet

The Longest Increasing Subsequence problem - a classic combinatorial challenge with deep connections to statistical mechanics- exhibits a rich thermodynamic landscape. Introducing a temperature we identify two distinct energy scales: A Schottky-like crossover at T_cross and a condensation transition at T_cond, below which the number of maximum-length configu

Narakorn Kaewkhao, Suparat Marit, Phongpichit Channuie

The Rosen Lagrangian framework promotes the cosmological constant to a scale-factor-dependent quantity, $Λ(a)=Λ_{0}a^λ$, thereby providing a dynamical dark energy scenario for $λ\neq 0$. In the special case $λ=0$, the model naturally reduces to the standard $Λ$CDM cosmology. Within this framework, the conformal Killing equations are employed to determine the

Ethan Akin

We provide a simple construction which realizes the Birkhoff center depth at an arbitrary ordinal level and relate it to the Cantor-Bendixson depth.

Yaxuan Kong, Qingren Yao, Yuqi Nie, Yichen Li

Time series data inform critical decisions across many real-world domains. While large language model (LLM) agents can analyze data through natural language and tools, it remains unclear whether they can conduct reliable time series analysis across multi-turn conversations. Existing benchmarks focus on single-step tasks such as forecasting and anomaly detect

Vincent Koc, Patrick Erichsen, Jacob Tomlinson, Agustin Rivera

Agent skills extend AI agents with reusable instructions, tools, scripts, references, and workflows, establishing a security boundary distinct from both model safety and traditional package-malware detection. ClawHub Security Signals is a sanitized dataset of 67,453 latest public OpenClaw skill versions. Each row pairs redacted SKILL.md content and sanitized

Hao Liang, Zhixuan Ge, Soumendu Majee, Joanna Li

Reconstructing a photorealistic 3D face avatar from a single unconstrained photograph is challenging: feed-forward 3D Gaussian Splatting (3DGS) models degrade on out-of-distribution inputs, while pretrained diffusion models produce high-fidelity images but lack multi-view consistency. We observe that these paradigms are fundamentally complementary: explicit

Shihan Kanungo

In this expository paper, we develop the basic ideas underlying Grothendieck groups and to illustrate their appearance across algebra, topology, representation theory, and homological algebra. Motivated by the universal construction associated to a commutative monoid, we define the Grothendieck groups abelian categories and rings. Along the way we study seve

Chun-Hsien Hsu

We geometrize the Poisson summation formula for the zero locus of a split quadratic form in an even number of variables over number fields. We do so by making explicit the relationship between Schwartz spaces on quadrics defined in two different ways: via Braverman-Kazhdan spaces and via theta lifts.

Nagarjuna Kanamarlapudi, Praveen K

We present a controlled experiment evaluating 12 multi-agent LLM collaboration topologies for software architecture design. Using a $2\times2\times2$ factorial design (Authority $\times$ Roles $\times$ Dynamics), we conducted 520 experimental runs across 8 design tasks of varying complexity, with 5 repetitions each. Designs were evaluated on a 12-dimensional

Vahidin Jeleskovic

Regression models and Vector Autoregressive Models (VARs) play crucial roles in econometrics by allowing the analysis of multiple variables simultaneously. Despite their utility, these models face challenges like underfitting and overfitting, especially when determining the optimal model specification, which can lead to significant computational costs. To ad

Bappa Muktar, Justin Moskolaï Ngossaha, Adama Nouboukpo

Vehicular Ad Hoc Networks (VANETs) are a key component of Intelligent Transportation Systems, enabling cooperative communication among vehicles and between vehicles and roadside infrastructure. However, their highly dynamic topology makes them vulnerable to network fragmentation, particularly in highway scenarios, low-density traffic conditions, localized ac

Ming Xu, Iman Shames

In this article, we establish the global convergence properties of the FilterDDP algorithm, which extends the discrete-time differential dynamic programming (DDP) algorithm of Mayne and Jacobson [\emph{International Journal of Control}, 3, (1966), pp. 85-95] to handle nonlinear constraints over states and controls, in addition to the dynamics. FilterDDP adop

André Oliveira Pinheiro

We develop a holographic framework for continuous higher-form symmetries and their low-energy effective descriptions, based on bulk path integrals, holographic renormalisation and boundary-condition-changing deformations. We show how approximate higher-form symmetries associated with a defect current can be realised holographically through massive antisymmet

Ali Alavi

We describe our submission to the VRR Challenge @ CVPR 2026, built on the \emph{ImplicitQA} / \emph{VRR-QA} benchmark~\cite{implicitqa}: multiple-choice video question answering in which answers are deliberately \emph{not} observable in any single frame and must be inferred from spatial layout, motion, depth, viewpoint, causality, and social context across d

Qiyuan Cheng, Shunzi Guo

We study the curve diffusion flow for open planar curves whose endpoints are constrained to lie on two fixed straight lines that intersect at an angle $θ(\in(0,π)) $. For every such angle, we prove that under suitable initial conditions the flow exists globally in time. Moreover, we show that the evolving curve converges - exponentially and in the smooth top

Wei-Tzu Lee, Keisuke Kamahori, Baris Kasikci

Long-form automatic speech recognition (ASR) requires both high accuracy and low latency, but existing systems force a trade-off between the two. Chunk-based pipelines process audio in parallel windows for low latency, but lose cross-chunk context and need brittle heuristics to align speakers and timestamps at boundaries. Long-context ASR models resolve ever

Francielle Vargas, João Robiatti, Diego Alves, Lucas Pascotti Valem

Ensuring factuality and interpretability in RAG remains an open and urgent problem. We introduce Contrastive Evidence Rationale Attention (CERA), the first retrieval framework to employ subjectivity-based hard negative selection and inject an evidential inductive bias into contrastive learning through an auxiliary attention alignment loss. CERA fine-tunes a

Yingzi Ma, Xiaogeng Liu, Yawen Zheng, Chaowei Xiao

With the rapid advancements in text-to-image diffusion models, generative video models (T2V models) like Sora can now produce short synthetic videos from a text prompt or an initial image. However, synthetic video generation -- especially when guided by an initial image -- often poses risks, including the potential creation of illegal, politically sensitive,

Ehsan Roohi

Cold-to-hot heat transfer in rarefied cavities is usually treated as a signature of Fourier-law failure. Here it is used to ask whether a correct anti-Fourier heat-flux field certifies the flux-side fourth-order closure state. In a two-dimensional monatomic flow, the heat-flux hierarchy observes the divergence of the composite R26-level tensor \(A_{ij}=R^{\c

Hongfei Du, Jiacheng Shi, Sidi Lu, Gang Zhou

Integrating large language models (LLMs) into text-to-speech (TTS) systems has improved speech expressiveness, yet interpretable emotional control remains challenging. Existing approaches primarily rely on external conditioning or global activation steering, offering limited insight into the internal representations underlying emotional control. In this work

Heedou Kim, Mogan Gim, Donghee Choi, Hoonick Lee

The rapid evolution of online scams, driven by transnational networks and mass produced social engineering scenarios, has exposed the speed limitations of conventional detection, necessitating tighter interagency coordination. While LLMs show promise in scam identification, their role in accelerating integrated response frameworks remains underexplored. We p

Anthony Haitao Zou, Yizhou Jake Cai, Ting Fung Ma

In this paper, we propose a generic optimization approach for challenging objective functions that finds applications in various statistical problems. We focus on objective functions with two parameter blocks of one amenable to analytic optimization, and another that is irregular or computationally expensive. To address this setting, we propose the Voronoi-E

Pablo A. Monroy-D'Croz, Rafael Ramirez-Melendez, Julian Cespedes-Guevara

This paper presents a minimalist brain-computer Musical Interface (BCMI) that functions as a real-time affective sonification system, translating prefrontal EEG activity into adaptive music. Emotional valence is estimated from frontal alpha asymmetry (AF7/AF8) and mapped to musical features such as mode, tempo, rhythmic density, and pitch register through a

Nataraj Agaram Sundar, Tejas Morabia

High-stakes production document-generation systems require language models to be adaptive, evidence-grounded, and auditable. We present HOPM, a hierarchical online prompt mutation framework evaluated on a real marketplace dispute-evidence workflow. HOPM treats prompts as online policies: a family/version router selects a prompt, deterministic guardrails attr

Griffin Pitts, Ashish Aggarwal

This full research paper investigates how engineering students' course-related beliefs relate to exam performance in a flipped introductory programming course. Understanding factors that influence student learning and performance has long been a focus of computing education research. While prior studies have identified psychological and contextually rele

Payel Mukhopadhyay, Stefan S. Nixon, Romain Watteaux, Michael McCabe

Whether physics foundation models can be usefully deployed on laboratory experiments remains an open question for scientific machine learning (ML). We test this question on the Rayleigh-Taylor instability (RTI), a ubiquitous and demanding fluid instability seen from tabletop flows to supernova explosions, in which small perturbations at a density interface g

Mahdi Bakhtiyarzadeh, Hadi Bayrami Asl Tekanlou, Jafar Razmara

The development of automatic term extraction has become increasingly important in modern technology. Automatic term extraction can be found in virtually every search engine that is currently available to users. Recent advancements have provided promising results for the extraction of automatic terms; however, accurate labeling is difficult because of several

JR Huml, Jonathan Wenger, John P. Cunningham

Due to their explicit priors and ability to model uncertainty, Bayesian methods have played a major role in dynamical latent variable modeling of single-cell neural recordings. However, modern-sized datasets have made overparameterized deep networks the preferred methods of choice due to their predictive power and favorable computational scaling. While many

Bernat Espigule

We investigate the topology of connectedness loci, denoted as $M_n$, for a one-parameter family of collinear affine iterated function systems featuring equally spaced translations. These loci are arithmetically equivalent to the closures of roots of monic polynomials whose non-leading coefficients fall within a prescribed finite interval of integers. Our mai

Luciana Basualdo Bonatto, Marcy Robertson

We construct an operadic model for the higher-genus Teichmüller tower. More precisely, we define a modular operad $\mathbf{S}$ in groupoids built from mapping class groups, with compositions and contractions encoding gluing operations on surfaces. We prove a presentation theorem for maps out of $\mathbf{S}$, showing that they are determined by a small number

Nikola Banić, Neven Elezović

Histogram uniformity testing is a common statistical task usually performed using Pearson's chi-square test. This paper proposes a new test based on the discrete total variation that is easy to compute and, for comb-like (alternating) deviations, achieves up to 67% higher statistical power than Pearson's chi-square test, making it a complement to sta

Ahmed Elhady, Eneko Agirre, Mikel Artetxe

Despite expanding their multilingual coverage, the advanced reasoning capabilities of LLMs remain largely confined to a few high-resource languages like English. To address this, we propose an unsupervised Reinforcement Learning (RL) approach to enhance multilingual reasoning by enforcing cross-lingual self-consistency: the principle that a model should prod

Santosh Kumar, Mukesh Kumar, Eunso Shin, Bassem Tossoun

Photonic integrated circuits have emerged as a powerful platform for high speed communication, sensing, and information processing due to their large bandwidth, low latency, and inherent parallelism. However, the absence of efficient, scalable, and non-volatile memory elements remains a fundamental limitation for realizing fully programmable and adaptive pho

Mingzhong Sun, Teresa Yeo, Armando Solar-Lezama, Tan Zhi-Xuan

Studies of human reasoning have shown that people are typically stronger at evaluating reasoning than producing it from scratch. In contrast, large reasoning models (LRMs) are trained to excel at producing long chains of reasoning to solve complex problems. How then do LRMs perform at evaluating reasons? We investigate this with the Valid-Answer-Invalid-Reas

Brenda Nogueira, Gisela A. Gonzalez-Montiel, Nitesh V. Chawla, Nuno Moniz

Developing effective anticancer therapeutics remains challenging due to tumor heterogeneity and the absence of well-defined molecular targets across cancer subtypes. Generative models conditioned on cancer genotypes offer a promising avenue for personalized drug discovery, yet existing approaches lack explicit optimization for simultaneous sensitivity, synth

Michael Taenzer

Multi-pitch estimation (MPE) typically predicts which pitches are active in a mixture, but not which instrument or source produced them. This paper investigates a lightweight slot-attention framework for multi-instrument MPE (MI-MPE), where a mixture CQT is mapped to an unordered set of source-like pitch maps. The model uses permutation-invariant Hungarian m

Subhadip Dey, Hee Oh, Konstantinos Tsouvalas

Roblin's theorem asserts that, in rank one, coamenable normal subgroups have the same critical exponent as the ambient group. We investigate the higher-rank analogue of this rigidity phenomenon. In higher rank, growth is directional, and there is no single analogue of Roblin's theorem. Instead, the answer splits into three complementary phenomena. Fi

Hojune Kim, Timothy Chen, Jiankai Sun, Lars W. Osterberg

Training vision-language-action (VLA) policies for humanoid loco-manipulation is constrained by the high cost and complexity of collecting human teleoperation demonstrations. VLA policies fine-tuned in simulators have, until now, failed to transfer effectively in humanoid loco-manipulation tasks. We present LEGS (Loco-manipulation via Embodied Gaussian Splat

Mohammad Ali Javidian

Bayesian optimization is a popular way to optimize expensive systems, where every experiment, simulation, or intervention costs time or money. In its standard form, it treats the variables we control as plain inputs to a black box and cannot tell apart mere correlation from a real cause and effect. Causal Bayesian optimization closes part of this gap by usin

Hamidreza Hasani Balyani, Seyed Pouyan Mousavi Davoudi, Alireza Amiri-Margavi, Amin Gholami Davodi

Large language models are increasingly deployed as advisors whose objective is not aligned with the user's: recommenders optimize for engagement, sales assistants for purchases, negotiation agents for concessions. Whether such advisors stay truthful when honesty conflicts with their own payoff is a core alignment-evaluation question. We turn the canonica

Tucker Hathaway, Daniel Chew

Adversarial feature extraction and blocking jamming threaten tactical CPM links. This paper presents a unitary spreading-based Transmission Security (TRANSEC) enhancement to obscure physical-layer signatures and improve anti-jamming (AJ) resilience. The enhancement can be used to augment existing techniques. The enhancement preserves the constant-envelope (0

Ish Kumar Jain, Rohith Reddy Vennam, Dinesh Bharadia

The next generation of 6G networks aims to utilize ultra-wideband spectrum and massive antenna arrays to serve multiple users with both control and data channels at low latency and high efficiency. However, phased arrays at mmWave and mid-bands are fundamentally constrained to a single beam or suffer sharp beamforming loss when split across directions, limit

Jordi Guàrdia i Rúbies, John W. Jones, Kevin Keating, Sebastian Pauli

We give an algorithm for choosing a distinguished defining polynomial for a p-adic field extension. This algorithm formed an important ingredient in the recent expansion of the database of p-adic fields within the L-functions and modular forms database.

V. S. Raghu Parupudi, Harsha Ponnada, Aditi Kaushal, S. Shria Parupudi

Reference-free evaluation of large language model (LLM) creativity relies on perplexity, entropy, and top-1 margin. We show that a much stronger signal lives one step earlier in the pipeline: in how sampling temperature \emph{reshapes} the model's token distribution before the next token is drawn. On Llama-3.1-8B-Instruct generations of 500 open-ended cr

Denis Lebold, Hendrik Wöhrle

The increasing computational complexity of deep neural network inference poses significant challenges for efficient hardware acceleration on embedded platforms, particularly with respect to resource consumption and scalability. This work presents OpenEye, a scalable and sparsity-aware FPGA-based hardware accelerator designed to efficiently execute common neu

Souvik Bhowmick, Sekhar Ghosh, Vishvesh Kumar, R. Lakshmi

The main aim of this paper is to establish the Hölder continuity and the Harnack inequality for weak solutions to Dirichlet problems associated with superposition operators of mixed fractional order, thereby complementing our previous work \cite{BGKL2026}. To achieve this, we extend the De Giorgi--Nash--Moser theory to the framework of superposition operator

Gayrat Toshpulatov

The Boltzmann equation describing the transport of electrons in semiconductor devices with an external electrostatic potential is considered when the spatial variable is in a torus and the wave vector is in the Brillouin zone. We prove the exponential time decay of solutions towards the global equilibrium in a weighted $L^2$ space. Our result holds for wide

Adam Coffman, Yifei Pan

For real multivariate polynomials $P$ and $Q$ both vanishing at a point, if the zero set of $Q$ is contained in the zero set of $P$, then there exists a rational function of the form $P^{p}/Q^{q}$ which is locally bounded and such that its extension that vanishes on the zero set of $Q$ is discontinuous. The proof uses inequalities of Lojasiewicz.

Fiona Y. Wang, Markus J. Buehler

Scientific discovery is not only answer generation but revision of the representational regime in which evidence, artifacts, operations, and verifiers are typed. We develop a category-theoretic account of agentic discovery for materials science. In a fixed regime b with schema category S_b, the system state is a copresheaf I_t: S_b -> Set, and provenance is

Triet M. Le

A central difficulty in training Joint-Embedding Predictive Architectures (JEPAs) is preventing representation collapse. LeJEPA addresses this by enforcing an isotropic Gaussian target on the embeddings via Sketched Isotropic Gaussian Regularization (SIGReg). This target is in tension with the manifold hypothesis, which expects embeddings to concentrate on a

Raj Patel, David Amebley, Taye Akinrele, Shaswata Mitra

Network intrusion detection is a core component of modern cybersecurity infrastructure, yet the deep learning models that dominate the field are computationally demanding, motivating interest in lightweight alternatives suited to edge and neuromorphic deployment. Spiking Neural Networks (SNNs) are therefore a natural candidate, but their design space, spanni