arXiv preprints from January 1, 2026 through September 15, 2026 — 06:54:37 EST
Transition-metal nanoclusters exhibit structural and electronic properties that depend on their size, often making them superior to bulk materials for heterogeneous catalysis. However, their performance can be limited by sulfur poisoning. Here, we use dispersion-corrected density functional theory (DFT) and physics-informed machine...
We study generalization in an overparameterized continual linear regression setting, where a model is trained with L2 (isotropic) regularization across a sequence of tasks. We derive a closed-form expression for the expected generalization loss in the high-dimensional regime that holds for arbitrary linear teachers. We demonstrate...
Large Language Models (LLMs) such as ChatGPT promise revolutionary capabilities for Sixth-Generation (6G) wireless networks but their massive computational requirements and tendency to generate technically incorrect information create deployment barriers. In this work, we introduce MAINTAINED: autonomous artificial intelligence agent...
We propose to use the self-assembly ability of a block copolymer combined with compression-expansion cycles to obtain CdSe quantum dots (QDs) structures of different morphology. The methodology proposed consists in transferring onto mica mixed Langmuir monolayers of QDs and the polymer poly (styrene-co-maleic anhydride) partial 2...
We define a new escape game in graphs that we call Nemesis. The game is played on a graph having a subset of vertices labeled as exits and the goal of one of the two players, called the fugitive, is to reach one of these exit vertices. The second player, i.e. the fugitive adversary, is called the Nemesis. Her goal is to trap the...
Robust reversible watermarking in encrypted images (RRWEI) faces an inherent challenge in simultaneously achieving robustness, reversibility, and content privacy under severely constrained embedding capacity. Existing RRWEI schemes often exhibit limited robustness against noise, lossy compression, and cropping attacks due to...
Social media imagery provides a low-latency source of situational information during natural and human-induced disasters, enabling rapid damage assessment and response. While Visual Question Answering (VQA) has shown strong performance in general-purpose domains, its suitability for the complex and safety-critical reasoning required...
Wi-Fi networks increasingly suffer from performance degradation caused by contention-based channel access, dense deployments, and largely self-managed operation among mutually interfering access points (APs). In this paper, we propose a Digital Twin (DT) framework that captures the essential spatial and temporal characteristics of...
Despite recent progress in 3D Gaussian-based head avatar modeling, efficiently generating high fidelity avatars remains a challenge. Current methods typically rely on extensive multi-view capture setups or monocular videos with per-identity optimization during inference, limiting their scalability and ease of use on unseen subjects....
Although Multimodal Large Language Models (MLLMs) demonstrate strong omni-modal perception, their ability to forecast future events from audio-visual cues remains largely unexplored, as existing benchmarks focus mainly on retrospective understanding. To bridge this gap, we introduce FutureOmni, the first benchmark designed to evaluate...
Users should not be systemically disadvantaged by the language they use for interacting with LLMs; i.e. users across languages should get responses of similar quality irrespective of language used. In this work, we create a set of real-world open-ended questions based on our analysis of the WildChat dataset and use it to evaluate...
Fluid turn-taking remains a key challenge in human-robot interaction. Self-supervised speech representations (S3Rs) have driven many advances, but it remains unclear whether S3R-based turn-taking models rely on prosodic cues, lexical cues or both. We introduce a vocoder-based approach to control prosody and lexical cues in speech more...
We estimate the national social cost of carbon using a recent meta-analysis of the total impact of climate change and a standard integrated assessment model. The average social cost of carbon closely follows per capita income, the national social cost of carbon the size of the population. The national social cost of carbon measures...
The fact that moisture propagates in soils at a finite speed is confirmed by natural everyday experience as well as by controlled laboratory tests. In this text, we rigorously derive analytical upper bounds for the speed of moisture front propagation under gravity for the solution to the Richards equation with compactly supported...
To address the issues of high operational costs and low energy efficiency (EE) caused by the dense deployment of small base stations (s-BSs) in 5G ultra-dense networks (UDNs), this paper first constructs a multi-objective mathematical optimization model targeting maximizing EE and minimizing the number of active BSs. The model...
Diffusion model, the state-of-the-art generative machine learning architecture, has shown promising results airfoil inverse designs. In this study, we implemented and trained a series of diffusion models on three different airfoil geometry data encoding formats -- principal component weights, ordered $x$-$y$ coordinates, and 2D signed...
In this study, we revisit the well-known cosmic microwave background (CMB) anomaly referred to as Hemispherical Power Asymmetry (HPA), using CMB temperature maps from the Planck mission public release 4 (PR4) and the WMAP nine-year data release. Employing the Local Variance Estimator (LVE) method, we systematically reexamine the...
We present a unified overview of recent progress in the study of QCD radiation in heavy-quark jets, focusing on the dead-cone effect. Using precision data from LEP at $\sqrt{s}=91.2$~GeV, we demonstrate strong momentum-space suppression in charm and bottom quark jets, supported by Monte Carlo simulations with \textsc{Pythia}8, and...
Modulating retroreflectors (MRRs) offer a promising pathway to low-complexity and energy efficient asymmetric optical inter-satellite link (OISL) for small spacecrafts, such as CubeSats. In this paper, we develop a unified statistical channel model for an on off keying modulated, retroreflector-enabled OISL. The model captures both...
Continuous attractor networks (CANs) are a well-established class of models for representing low-dimensional continuous variables such as head direction, spatial position, and phase. In canonical spatial domains, transitions along the attractor manifold are driven by continuous displacement signals, such as angular velocity-provided...
Thin films of colloidal semiconductor CdSe quantum-dots (QDs) and a styrene/maleic anhydride copolymer were prepared by the Langmuir-Blodgett technique and their photoluminescence was characterized by confocal fluorescence lifetime microscopy. In the films, the photoluminescence dynamics are strongly affected by excitation energy...
We investigate dimensional constraints arising from representation theory when abstract graph edges possess internal degrees of freedom but lack geometric properties. We prove that such internal degrees of freedom can only encode directional information, necessitating quantum states in $\mathbb{C}^2$ (qubits) as the minimal...
Multiple-input multiple-output (MIMO) systems require efficient and accurate channel estimation with low pilot overhead to unlock their full potential for high spectral and energy efficiency. While deep generative models have emerged as a powerful foundation for the channel estimation task, the existing approaches using...
Concept erasure aims to suppress sensitive content in diffusion models, but recent studies show that erased concepts can still be reawakened, revealing vulnerabilities in erasure methods. Existing reawakening methods mainly rely on prompt-level optimization to manipulate sampling trajectories, neglecting other generative factors,...
As edge devices gain stronger computing power, deploying high-performance DNN models on untrusted hardware has become a practical approach to cut inference latency and protect user data privacy. Given high model training costs and user experience requirements, balancing model privacy and low runtime overhead is critical. TEEs offer a...