Accurate global Subseasonal-to-Seasonal (S2S) climate forecasting is critical for disaster preparedness and resource management, yet it remains challenging due to chaotic atmospheric dynamics. Existing models predominantly treat atmospheric fields as isotropic images, conflating the distinct physical processes of zonal wave propagation and meridional transport, and leading to suboptimal modeling of anisotropic dynamics. In this paper, we propose the Symmetric Orthogonal Operator Network (SOON) for global S2S climate forecasting. It couples: (1) an Anisotropic Embedding strategy that tokenizes the global grid into latitudinal rings, preserving the integrity of zonal periodic structures; and (2) a stack of SOON Blocks that models the alternating interaction of Zonal and Meridional Operators via a symmetric decomposition, structurally mitigating discretization errors inherent in long-term integration. Extensive experiments on the Earth Reanalysis 5 dataset demonstrate that SOON establishes a new state-of-the-art, significantly outperforming existing methods in both forecasting accuracy and computational efficiency.
There is currently a major trend to hairpin-winding motors for small and medium drives with increased power, specifically more torque density in the automotive industry. Practically all large players in the field either already use this winding technology or have announced doing so soon. However, hairpins, bar windings, and other segmented winding techniques are not purely a material and production issue. Instead their application to small drives influences all aspects of the design of machines, which are currently explored and studied by the industry. These range from not obvious gaps in the theory, parameter studies for maxima of efficiency, possible as well as advantageous winding schemes, thermal design, and ways to control ac losses to specific materials and process difficulties. Despite the apparent novelty of the trend, however, designers could revisit a widely forgotten knowledge base of more than 100 years for many of those questions. This old knowledge base and the understanding that many recently presented concepts have been developed earlier may speed up the technological development and appear to be a key to further innovation. Instead, many problems need to be solved
We present an exact rate analysis for a secret key that can be shared among two parties employing a linear quantum repeater chain. One of our main motivations is to address the question whether simply placing quantum memories along a quantum communication channel can be beneficial in a realistic setting. The underlying model assumes deterministic entanglement swapping of single-spin quantum memories and it excludes probabilistic entanglement distillation, and thus two-way classical communication, on higher nesting levels. Within this framework, we identify the essential properties of any optimal repeater scheme: entanglement distribution in parallel, entanglement swapping as soon and parallel quantum storage as little as possible. While these features are obvious or trivial for the simplest repeater with one middle station, for more stations they cannot always be combined. We propose an optimal scheme including channel loss and memory dephasing, proving its optimality for the case of two stations and conjecturing it for the general case. In an even more realistic setting, we consider additional tools and parameters such as memory cut-offs, multiplexing, initial state and swapping g
We study the spectral gaps of parallel and simulated tempering chains targeting multimodal Gibbs measures. In particular, we consider chains constructed from Metropolis random walks that preserve the Gibbs distributions at a sequence of harmonically spaced temperatures. We prove that their spectral gaps admit polynomial lower bounds of order $11$ and $12$ in terms of the low target temperature. The analysis applies to a broad class of potentials, beyond mixture models, without requiring explicit structural information on the energy landscape. The main idea is to decompose the state space and construct a Lyapunov function based on a suitably perturbed potential, which allows us to establish lower bounds on the local spectral gaps.
We develop a generalized non-Hermitian Hamiltonian formalism for guided resonances in photonic crystal slabs, derived directly from Maxwell's equations through a systematic guided-mode expansion. By expanding the electromagnetic fields over the complete mode basis of an unpatterned slab and systematically integrating out radiative Fabry--Pérot channels, we obtain the analytical operator structure of the Hamiltonian, which treats guided-mode coupling and radiation losses on equal footing. The resulting Hamiltonian provides explicit expressions for both dispersive and radiative coupling terms in terms of modal overlap integrals and Fourier components of the permittivity modulation. For specific geometries, the Hamiltonian coefficients can be extracted from full-wave simulations enabling accurate modeling without phenomenological assumptions. As a case study, we investigate hexagonal lattices with both preserved and broken $C_6$ symmetry, demonstrating predictive agreement for complex band structures, near-field distributions, and far-field polarization patterns. In particular, the formalism reproduces symmetry-protected bound states in the continuum (BICs) at the $Γ$ point, accidenta
Pairs of entangled photons are crucial for photonic quantum technologies. The demand for integrability and multi-functionality suggests 'flat' platforms - ultrathin layers and metasurfaces - as sources of photon pairs. With the success in demonstrating spontaneous parametric down-conversion (SPDC) from such sources, an alternative process to generate photon pairs, spontaneous four-wave mixing (SFWM), also starts to attract interest. In materials with nonzero second-order nonlinear susceptibility $χ^{(2)}$, SFWM can generate photon pairs both directly, through the third-order nonlinear susceptibility $χ^{(3)}$, and in a cascaded way, through second harmonic generation (SHG) followed by SPDC. Usually, the cascaded process is more efficient. Here, we show that in a thin layer, direct SFWM dominates, because the wavevector mismatch for SFWM is much smaller than for SHG or SPDC. To demonstrate it, we implement the photon pair generation via SFWM in a second-order nonlinear material - a thin layer of lithium niobate (LN). The existence of both second- and third-order nonlinear processes offers broader opportunities for quantum state engineering.
Short-period massive binary stars are predicted to undergo chemically homogeneous evolution (CHE), making them prime candidates for producing binary black holes (BBHs) that may merge within the age of the Universe. Most of these binaries have a tertiary companion, and here we explore how a nearby third body possibly influences this evolutionary channel. Our analysis combines analytic treatments of triple dynamics with insights from detailed stellar evolution models, focusing on the role of the von Zeipel-Lidov-Kozai mechanism while also accounting for tidal and general relativistic apsidal precession. We examine the dynamics of triples at three critical evolutionary stages: the zero-age main sequence, shortly after the main sequence, and at the time of BBH formation. We find that, for triples having outer orbital periods less than 70 d(120 d), the inner binary can merge during(or after) the main-sequence stage, leading to a hydrogen-rich(helium-rich) stellar merger. If a stellar merger is avoided, the inner binary may eventually form a BBH. In mildly hierarchical triples, with outer periods of around 100 d, the tertiary component can trigger a rapid merger of the BBH on timescales
We quantitatively study the mixing rate of randomly shifted alternating shears on the torus. This flow was introduced by Pierrehumbert '94, and was recently shown to be exponentially mixing. In this work, we quantify the dependence of the exponential mixing rate on the flow amplitude. Our approach is based on constructing an explicit Lyapunov function and a coupling trajectory for the associated two-point Markov chain, together with an application of the quantitative Harris theorem.
Learning to reject provide a learning paradigm which allows for our models to abstain from making predictions. One way to learn the rejector is to learn an ideal marginal distribution (w.r.t. the input domain) - which characterizes a hypothetical best marginal distribution - and compares it to the true marginal distribution via a density ratio. In this paper, we consider learning a joint ideal distribution over both inputs and labels; and develop a link between rejection and thresholding different statistical divergences. We further find that when one considers a variant of the log-loss, the rejector obtained by considering the joint ideal distribution corresponds to the thresholding of the skewed Bhattacharyya divergence between class-probabilities. This is in contrast to the marginal case - that is equivalent to a typical characterization of optimal rejection, Chow's Rule - which corresponds to a thresholding of the Kullback-Leibler divergence. In general, we find that rejecting via a Bhattacharyya divergence is less aggressive than Chow's Rule.
We present the formalism of hadron-to-two-hadron transition generalized parton distributions (GPDs) for spinless hadron case. Definitions of the twist-2 unpolarized and polarized $π\toππ$ transition GPDs are introduced with a particular choice of kinematic variables that characterize the produced two-pion system. In the vicinity of $ρ(770)$, we work out the two-pion decay angular distributions of the $e^-π\to e^-γππ$ cross section, incorporating both the Bethe-Heitler and deeply virtual Compton scattering processes. Each cross section exhibits a distinctive angular distribution, which is sensitive to the polarization states of the produced $ρ(770)$ resonance. In addition, we construct the double partial-wave expansion of $π\toππ$ transition GPDs in the two-pion decay angles as a generalization of GPDs for transition from a pion to a spin-$\ell$ resonance state. With the help of the Omnès representation, we constrain the $π\toππ$ transition GPDs in terms of the low energy $ππ$ scattering phase shift and build a phenomenological model for these GPDs.
We classify real-analytic $\mathrm{SL}(n,\mathbb{R})$-actions on closed manifolds of dimension m for $3\leq n\leq m\leq2n-3$, which extends Fisher--Melnick's work for $\mathrm{SL}(n,\mathbb{R})$-actions on closed n-manifolds. Additionally, we classify smooth $\mathrm{SL}(n,\mathbb{R})$-actions on closed m-manifolds that are fixed-point free. As a corollary, we obtain the density or non-density of structural stability of fixed-point free $\mathrm{SL}(n,\mathbb{R})$-actions.
Type 1 Diabetes is a chronic autoimmune condition in which the immune system attacks and destroys insulin-producing beta cells in the pancreas, resulting in little to no insulin production. Insulin helps glucose in your blood enter your muscle, fat, and liver cells so they can use it for energy or store it for later use. If insulin is insufficient, it causes sugar to build up in the blood and leads to serious health problems. People with Type 1 Diabetes need synthetic insulin every day. In diabetes management, continuous glucose monitoring is an important feature that provides near real-time blood glucose data. It is useful in deciding the synthetic insulin dose. In this research work, we used machine learning tools, deep neural networks, deep reinforcement learning, and voting and stacking regressors to predict blood glucose levels at 30-min time intervals using the latest DiaTrend dataset. Predicting blood glucose levels is useful in better diabetes management systems. The trained models were compared using several evaluation metrics. Our evaluation results demonstrate the performance of various models across different glycemic conditions for blood glucose prediction. The source
Quantum information theory explores the limits of manipulating quantum states. While auxiliary systems often enhance information processing, a systematic explanation for their power has been lacking. This thesis addresses this gap by investigating the underlying sources of strength in using auxiliary systems. We then apply these insights to practical problems in quantum computing and devise an algorithmic paradigm leveraging auxiliary systems. The first part examines catalysts -- auxiliary systems that remain unaltered -- and identifies three advantages: a memory effect, the ability to fine-tune catalyst states, and their role as seed states for resource distribution. The second part presents a strategy for solving recursive problems in quantum algorithms by employing auxiliary states as memories, achieving an exponential reduction in circuit depth at the cost of increased width. The findings in this thesis would facilitate future research into fundamental problems like resource interconversion and practical ones like optimal quantum circuit synthesis.
Deep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, an Extreme Learning Machine (ELM) framework for DeepONets that leverages the backpropagation-free nature of ELM. By reformulating DeepONet training as a least-squares problem for newly introduced parameters, the ELM-DeepONet approach significantly reduces training complexity. Validation on benchmark problems, including nonlinear ODEs and PDEs, demonstrates that the proposed method not only achieves superior accuracy but also drastically reduces computational costs. This work offers a scalable and efficient alternative for operator learning in scientific computing.
Current and future surveys rely on machine learning classification to obtain large and complete samples of transients. Many of these algorithms are restricted by training samples that contain a limited number of spectroscopically confirmed events. Here, we present the first real-time application of Active Learning to optimise spectroscopic follow-up with the goal of improving training sets of early type Ia supernovae (SNe Ia) classifiers. Using a photometric classifier for early SN Ia, we apply an Active Learning strategy for follow-up optimisation using the real-time FINK broker processing of the ZTF public stream. We perform follow-up observations at the ANU 2.3m telescope in Australia and obtain 92 spectroscopic classified events that are incorporated in our training set. We show that our follow-up strategy yields a training set that, with 25% less spectra, improves classification metrics when compared to publicly reported spectra. Our strategy selects in average fainter events and, not only supernovae types, but also microlensing events and flaring stars which are usually not incorporated on training sets. Our results confirm the effectiveness of active learning strategies to c
Visual Question Answering (VQA) has recently emerged as a potential research domain, captivating the interest of many in the field of artificial intelligence and computer vision. Despite the prevalence of approaches in English, there is a notable lack of systems specifically developed for certain languages, particularly Vietnamese. This study aims to bridge this gap by conducting comprehensive experiments on the Vietnamese Visual Question Answering (ViVQA) dataset, demonstrating the effectiveness of our proposed model. In response to community interest, we have developed a model that enhances image representation capabilities, thereby improving overall performance in the ViVQA system. Specifically, our model integrates the Bootstrapping Language-Image Pre-training with frozen unimodal models (BLIP-2) and the convolutional neural network EfficientNet to extract and process both local and global features from images. This integration leverages the strengths of transformer-based architectures for capturing comprehensive contextual information and convolutional networks for detailed local features. By freezing the parameters of these pre-trained models, we significantly reduce the comp
We introduce HyperCLOVA X, a family of large language models (LLMs) tailored to the Korean language and culture, along with competitive capabilities in English, math, and coding. HyperCLOVA X was trained on a balanced mix of Korean, English, and code data, followed by instruction-tuning with high-quality human-annotated datasets while abiding by strict safety guidelines reflecting our commitment to responsible AI. The model is evaluated across various benchmarks, including comprehensive reasoning, knowledge, commonsense, factuality, coding, math, chatting, instruction-following, and harmlessness, in both Korean and English. HyperCLOVA X exhibits strong reasoning capabilities in Korean backed by a deep understanding of the language and cultural nuances. Further analysis of the inherent bilingual nature and its extension to multilingualism highlights the model's cross-lingual proficiency and strong generalization ability to untargeted languages, including machine translation between several language pairs and cross-lingual inference tasks. We believe that HyperCLOVA X can provide helpful guidance for regions or countries in developing their sovereign LLMs.
Enhancing light emission from perovskite nanocrystal (NC) films is essential in light-emitting devices, as their conventional stacks often restrict the escape of emitted light. This work addresses this challenge by employing a TiO$_2$ grating to enhance light extraction and shape the emission of CsPbBr$_3$ nanocrystal films. Angle-resolved photoluminescence (PL) demonstrated a tenfold increase in emission intensity by coupling the Bloch resonances of the grating with the spontaneous emission of the perovskite NCs. Fluorescence lifetime imaging microscopy (FLIM) provided micrometer-resolution mapping of both PL intensity and lifetime across a large area, revealing a decrease in PL lifetime from 8.2 ns for NC films on glass to 6.1 ns on the TiO$_2$ grating. Back focal plane (BFP) spectroscopy confirmed how the Bloch resonances transformed the unpolarized, spatially incoherent emission of NCs into polarized and directed light. These findings provide further insights into the interactions between dielectric nanostructures and perovskite NC films, offering possible pathways for designing better performing perovskite optoelectronic devices.
Transformer-based large-scale pre-trained models achieve great success. Fine-tuning is the standard practice for leveraging these models in downstream tasks. Among the fine-tuning methods, adapter-tuning provides a parameter-efficient fine-tuning by introducing lightweight trainable modules while keeping most pre-trained parameters frozen. However, existing adapter-tuning methods still impose substantial resource usage. Through our investigation, we show that each adapter unequally contributes to both task performance and resource usage. Motivated by this insight, we propose Selective Adapter FrEezing (SAFE), which gradually freezes less important adapters early to reduce unnecessary resource usage while maintaining performance. In our experiments, SAFE reduces memory usage, computation amount, and training time by 42.85\%, 34.59\%, and 11.82\%, respectively, while achieving comparable or better task performance compared to the baseline. We also demonstrate that SAFE induces regularization effect, thereby smoothing the loss landscape, which enables the model to generalize better by avoiding sharp minima.