Eric

Co-Higgs Bundle Classification Enables Poisson 3-Fold Structure Analysis on Hopf Surfaces

This paper extends prior classification work on rank-2 co-Higgs bundles over Hopf surfaces to construct and analyze associated Poisson 3-folds through their symplectic leaf structure. The work bridges algebraic geometry and symplectic geometry by leveraging bundle data to characterize the underlying geometric singularities of Poisson manifolds. This construction provides a concrete method for studying Poisson structures on complex varieties derived from holomorphic bundle geometry.

Local Reduction Algorithm Achieves 90%+ Feasibility for Uncertain Microgrid Optimization

This paper addresses optimal sizing and power scheduling in microgrids with demand, generation, and price uncertainties by proposing two problem formulations: one as a mixed-integer linear program with binary variables and big-M constraints, and another as a continuous nonlinear program using smooth reformulations of logical constraints. A novel local reduction algorithm is extended to solve both formulations, demonstrating feasibility rates exceeding 90% across 100,000-sample Monte Carlo simulations. The approach balances computational tractability with handling multiple uncertainty sources in energy storage system dispatch.

Compositional Grading Enables Ferroelectric Scaling in Ultrathin AlScN Films

Compositional grading within AlN-AlScN heterostructures mitigates the thickness-scaling limitations of ferroelectric AlScN by distributing structural and polarization discontinuities across the film, reducing defect formation and electric field concentration. A 20 nm graded structure demonstrates 21% higher breakdown field, 10% enhanced remanent polarization, and 40× higher resistivity than homogeneous AlScN, while functional ferroelectric switching is achieved in 5 nm stacks with only 2 nm of active Al₀.₆₄Sc₀.₃₆N. This approach establishes compositional engineering as a scalable defect-mitigation strategy for CMOS-compatible non-volatile memory applications.

Conditional Spinodal Decomposition Enables Dendrite-Resistant Li-Mg Anodes

Li-Mg alloy anodes for lithium metal batteries undergo spinodal decomposition that creates ordered B2 phases alongside η-BCC phases, forming interconnected chemical gradients that accelerate Li diffusion. This microstructural arrangement provides fast pathways for lithium transport while suppressing dendrite formation at high current densities, achieving improved battery performance using abundant, inexpensive magnesium rather than relying on rare materials. The mechanism represents a previously unreported phase behavior that fundamentally improves both the energy density and cyclability of next-generation batteries.

Protostellar Jets as Regulators of Star Formation Efficiency in Young Clusters

Protostellar jets reduce star formation efficiency in centrally concentrated molecular clouds from 19-33% down to 12-16%, operating as an underappreciated early feedback mechanism that reshapes cluster structure. Beyond efficiency suppression, jets introduce temporal burstiness to star formation and produce more dynamically unbound, substructured stellar systems that better match observations of real young clusters, suggesting jets play a critical regulatory role even in initially concentrated cloud environments.

Well-posedness of Kohn-Sham DFT Models for Encapsulated 2D Materials with Yukawa-type Interactions

This paper investigates Kohn-Sham Density Functional Theory (DFT) models for two-dimensional materials encapsulated within a three-dimensional environment between conducting electrodes. The Dirichlet boundary conditions imposed by the electrodes effectively screen the Coulomb interaction, reducing it to a short-ranged Yukawa type. The study demonstrates the well-posedness of these nonlinear Kohn-Sham DFT models for both periodic (e.g., graphene) and quasi-periodic (e.g., twisted bilayer graphene) materials in this configuration.

Kondo Effect Drives Quantum Criticality in Ferromagnetic CeRh$_6$Ge$_4$

CeRh$_6$Ge$_4$ exhibits ferromagnetic quantum criticality, a phenomenon typically avoided in clean itinerant Ce compounds. Spectroscopic investigations reveal that the hybridization of Ce 4f electrons with conduction electrons, driven by the Kondo effect, is key to this behavior. The observed temperature-dependent linear dichroism suggests an orbital-dependent Kondo effect, indicating a multiorbital ground state that needs to be considered in models of pressure-induced criticality.

Topology of H-Structures and Intrinsic Torsion Energy on Lie Groups

This paper investigates the topology of isometric classes of tensorial H-structures on closed, connected Lie groups, particularly focusing on their relationship with Riemannian metrics and intrinsic torsion energy. The authors establish a homotopy equivalence between the space of H-structures and Riemannian metrics, and analyze the scale-degenerate nature of the intrinsic torsion energy functional. Finite-time singularities are reinterpreted as concentrations in specific homotopy classes with zero energy infima.

How to Use Imperfect Advice Against No-Regret Learners in Repeated Games

This paper introduces a pseudo-metric to quantify the usefulness of advice (e.g., simulators or payoff matrix predictions) for an agent playing a two-player repeated game against a no-regret learner. When advice has correctness guarantees, the agent can leverage simulators to compute approximate Stackelberg strategies with reduced interaction complexity. Without guarantees, the agent faces a fundamental trade-off: they cannot simultaneously ensure near-Stackelberg performance when advice is accurate and no-regret behavior when advice is inaccurate, though they can weakly dominate certain correlated equilibria.

CEM-GMsFEM: Efficient Solution for High-Contrast Heterogeneous PDEs

This review details the Constrained Energy Minimizing Generalized Multiscale Finite Element Method (CEM-GMsFEM) for solving elliptic PDEs with high contrast heterogeneous coefficients. The method constructs multiscale basis functions using spectral auxiliary spaces and an oversampling strategy, which ensures localized computations and exponential error decay. Rigorous error estimates and numerical simulations confirm the method's optimal convergence and robustness, with documented applications in various fields.

Radio-Frequency Powering Systems for Accelerators: A Review of Technologies and Principles

This review provides a foundational overview of radio-frequency (RF) powering systems for accelerators, encompassing vacuum tube technologies (tetrodes, klystrons, IOTs) and transistors. It details essential principles of RF power transmission, including combining/splitting systems and transmission lines, and specifically addresses the role of RF power couplers. The content is derived from a contribution to the CERN Accelerator School.

RL-Driven Dynamic Parameter Tuning for High-Speed BFT Consensus

Modern BFT protocols expose many tunable parameters but rely on static or heuristic configurations, causing significant performance degradation under dynamic workloads and network conditions. AutoPilot is a reinforcement learning framework that continuously monitors runtime conditions and adjusts protocol parameters online in a decentralized manner, with built-in robustness against adversarial data pollution. Implemented on top of the Autobahn BFT protocol, it reduces end-to-end latency by 49.8% versus defaults and outperforms random exploration by 73.3% across dynamic environments.

CASCO Simulation and Experimental Validation of Pulsating Heat Pipes: Nucleation Barriers and Stopover Regimes

This work investigates the steady-state operation of pulsating heat pipes (PHPs) using the CASCO simulation software, version 4. Experimental validation was conducted on two prototypes to analyze the impact of surface properties and stopovers. An optimal nucleation barrier was identified, which minimizes thermal resistance at a specific evaporator power, leading to a continuous functioning regime with pressure waves propagating throughout the PHP. The stopover regime, characterized by alternating rapid pressure growth and slower decay, is also examined, with its duration shown to decrease with increased heating load due to faster liquid film shrinkage.

Meta-Learning with Attentive Latent Neural Processes Enables Scalable Photometric Reverberation Mapping for LSST

The authors present a meta-learning framework using Attentive Latent Neural Processes (ALNPs) combined with Self-Organizing Maps and Mixture Density Models to perform unsupervised photometric reverberation mapping of AGN. The framework clusters light curves by topology, learns latent representations encoding both transfer functions and SMBH parameters, and improves light-curve reconstruction by 60-70% and transfer function recovery by ~35% over baseline regressors. It generalizes from simulated to real ZTF data, making it well-suited for the massive AGN datasets (1000-4000 per sq. deg) expected from LSST.

Elastoinertial Effects Govern Dynamic Nonlinearity in Soft Hair Beds

This work characterizes the transient rheological response of elastic hair beds to oscillatory shear flows. The coupled hair-fluid interaction yields a nonlinear dynamic response dependent on frequency and amplitude, featuring a crossover between stress-lagging and stress-leading regimes influenced by viscosity, inertia, and elasticity. A rigid-beam model qualitatively captures this behavior, with implications for understanding mechanotransduction and biological signaling stability.

Numerical Study of Oscillatory Magnetic Reconnection Driven by Alternating Currents in a Laboratory Plasma

This numerical experiment investigates oscillatory magnetic reconnection in a laboratory plasma using the MPI-AMRVAC framework. The study details the dynamics of current sheet formation and evolution, driven by alternating current-induced magnetoacoustic waves perturbing a magnetic null. Key findings include the temporal reorientation of current sheets, the influence of the Hall effect on plasma flow, and the correlation between alternating current amplitude and reconnection characteristics.

Adaptive multiscale model reduction for linear elasticity equation in perforated domains

Area-preserving crystalline curvature flow in two dimensions

Impact of mechanical constraints on tokamak design and implications for high field power plants

Mass-Varying Neutrinos from an Inverse Symmetron

Lecture notes: Introduction to the Off-shell Double Copy Program

Learning to model pediatric asthma exacerbation from multiple risk factors: a case study in coastal Virginia

SECUREVENT: Hybrid AI/ML Security Monitoring for Distributed Event-Based Systems

Discovering Agents for Discovery: The Case for DNS

Revisiting ram pressure stripping in Wolf-Lundmark-Melotte: No evidence for stripped HI with LGLBS

Symmetric ribbon numbers of low-complexity knots

Approximating Hartree-Fock theory via an efficiently local reformulation

Carrier Localization in Pnictogen-Based Chalcohalides from Defect-Bound Hot Polarons

REACT: A Conditioning Framework for User-Adaptive sEMG Hand Pose Estimation

Examining Agents' Bias Amplification versus Suppression in Multi-Agent Systems

Energy and Scaling Limits of Phase-Change Memory

Geometric Origin of Macroscopic Alignment in Granular Flows

Consistent Identification of Top-$K$ Nodes in Noisy Networks

Dual-Use AI Face Swap Apps Are Mostly Unsafe: A Systematic Safety Audit

Muon in Vision Transformers: Optimizer-Recipe Interactions and Gradient Spectra

Calderon-type commutators and chamber lifting in the Dunkl setting

Cullen and Woodall numbers in Padovan and Perrin sequences

NIMROD-to-IMAS workflow for extended-magnetohydrodynamic data with reusable datasets and implications for IMAS schema development

OptiQU: Coordinated Multi-Level Voltage and Reactive Power Control for Enhanced Voltage Quality and Secure Grid Operation

A quasi-neutral electromagnetic hybrid model with drift-kinetic electrons and fully kinetic ions

TriSweep: A Four-Drone Swarm Framework for Electromagnetic Side-Channel Analysis

A universal framework to identify eccentric binary mergers: GW200105 case study

Discrete Wavelet Transform for Serial X-ray Crystallography Image Segmentation

Performance of low vision individuals when selecting a target with head-pointing in virtual reality

A discussion of two new proofs of Fefferman's Fourier extension theorem in the plane

Unveiling Magnetic Frustration via the Elastocaloric Effect

GQA-μP: The maximal parameterization update for grouped query attention

ITHICA: Intra-Thread Instruction Checking Approach for Defect-Induced Silent Data Corruptions

MyoChallenge 2025: A New Benchmark for Human Athletic Intelligence

Large Language Models for AI-Assisted Radiotherapy Scheduling: A Feasibility Study Under Realistic Operational Constraints