arXiv subjects

Sanjeev Kumar

Publications and source records attributed to Sanjeev Kumar.

At least 19 recordsLinked to original sources

arXiv2026

A novel Grand-Potential Phase-Field Lattice-Boltzmann model for multi-phase solidification with convection

Melt convection plays a critical role in microstructure evolution during alloy solidification, yet accurately capturing its interaction with moving solid-liquid interfaces remains a significant computational challenge, particularly in multi-phase, multi-component systems. In this work, we develop a computational framework that couples a Grand-Potential phase-field model with the Lattice Boltzmann method (LBM) to simulate convection-driven solidification within a unified and thermodynamically consistent formulation. The proposed approach rigorously enforces no-slip conditions at evolving solid-liquid interfaces, while fluid transport is solved using the standard single-relaxation-time Bhatnagar-Gross-Krook collision operator. The framework provides an efficient and robust methodology for resolving the coupled evolution of interfaces, solute transport, and fluid flow. The versatility of the proposed framework is demonstrated through simulations of dendritic and eutectic solidification under natural convection. The results show that convection significantly modifies solute segregation, destabilizes growth fronts, and induces oscillatory growth modes in both systems. These examples illustrate the capability of the proposed method to capture complex flow-induced morphological instabilities and provide new insights into the mechanisms governing convection-driven microstructure evolution in alloy solidification.

cond-mat.mtrl-sci↗

arXiv2026

When Tokenizers Fail: Byte-Level Chunking for Zero-Shot Transfer to Low-Resource Languages

Subword tokenization hinders low-resource language processing by imposing frequency patterns from dominant languages onto script-sharing variants. Byte-level models bypass this issue by processing raw UTF-8 characters, yet they create a granularity mismatch for word-level tasks in non-Latin scripts. Hierarchical byte-level architectures address this mismatch by grouping bytes into word-aligned chunks. However, these architectures require massive training data and suffer from representational misalignment when paired with frozen subword-based language models. In this paper, we propose an adapted hierarchical network framework that bridges this modality gap without extensive training. Our method initializes byte embeddings directly from the subword representations of a frozen base model. We apply a chunk alignment loss to project dynamically grouped byte chunks toward precomputed subword targets, and interleave lightweight part-of-speech (POS) supervision to guide boundary detection. Experiments across six languages demonstrate that our tokenizer-free approach improves performance for word-level morphological tasks, yielding up to a 13.3% improvement on POS tagging.

cs.CL↗

arXiv2026

Evolution of terahertz third harmonic response across rare-earth nickelate phase-diagram

High harmonic generation (HHG) is a sensitive probe for investigating electronic structures and dynamics of materials and a source for attosecond pulses. In particular, HHG with terahertz (THz) light can enable probing of nonlinear responses in correlated materials arising from low-energy many-body interactions. However, THz HHG studies have so far largely focused on topological materials and superconductors, leaving out other potential material systems which could also become efficient THz HHG sources. Here, we report THz third harmonic generation (THG) in rare-earth nickelates -- a prototype material for exploring the Mott insulator-metal transition and related technological applications. We find that the THG amplitude is highly sensitive to the strengths of electronic and magnetic phases of nickelates. In films with sharp phase-transitions, the local maximum and minimum in the temperature-dependent THG amplitude coincide with insulator-metal and magnetic transition temperatures, respectively. While in films with weaker transitions, these features shift toward lower temperatures or even monotonous THG enhancement is observed down to low temperatures. We developed a generalized theory for THz harmonic generation in negative charge-transfer insulators and outlined strategies to enhance the THz nonlinearities further. Our study broadens the scope of THz HHG studies and related applications to strongly correlated materials.

cond-mat.str-el↗

arXiv2026

Topological spin-texture transitions in van der Waals magnets revealed by X-ray Fourier transform holography

Nontrivial topological spin-textures, such as skyrmions, merons, bimerons, and skyrmioniums, are envisioned as robust building blocks for future memory and logic devices. Controllable transformations between these states require a quantum-mechanical description of electronic degrees of freedom and atomic-scale insight beyond existing phenomenological models. Here, we report an atomic-scale investigation of topological phase transitions and their protection in the two-dimensional van der Waals ferromagnet Fe$_3$GeTe$_2$ (FGT) using a combined experimental-theoretical approach. Synchrotron-based Fourier transform holography directly images labyrinth domains, isolated skyrmions, mixed labyrinth-skyrmion phases, and skyrmion bags with high spatial resolution. We compare these observations to simulations based on an electronic lattice Hamiltonian that captures both metallicity and relativistic spin-orbit coupling in FGT. By systematically exploring a broad range of temperatures and magnetic fields, we map the mechanisms governing topological transitions and their stability. This sequential-integrated experimental-theoretical framework advances understanding of spin-texture interactions and enables precise control of external tuning parameters. Our results establish a platform for creating, stabilizing, and manipulating topological states, paving the way for engineered spin-texture transitions in next-generation spintronic technologies.

cond-mat.mtrl-sci↗

arXiv2026

Supervised machine learning of compressible flow past a rotating cylinder

High-fidelity numerical simulations of compressible flow past a rapidly rotating cylinder are used to investigate the evolution of aerodynamic loads and flow instability over a wide range of Reynolds numbers (Re = 1000 to 6000). The study reveals a transition from periodic vortex shedding to complex multi-mode oscillatory states, with a critical bifurcation identified near Re = 5650. Spectral analysis of lift and drag signals shows the emergence and interaction of multiple dominant frequencies, accompanied by amplitude modulation and nonlinear mode coupling in the post-bifurcation regime. To model these highly nonlinear dependencies, data-driven approaches are systematically explored using a database of 101 high-fidelity simulations (1 million core hours). Polynomial regression provides baseline fits but fails to capture localized fluctuations near bifurcation. Bayesian regression frameworks employing B-spline and Gaussian radial basis functions improve flexibility and uncertainty quantification, with spline-based models demonstrating superior performance in capturing piecewise nonlinear trends. Artificial neural networks (ANNs) are then developed as high-capacity surrogate models, achieving excellent predictive accuracy for maximum lift coefficient and instability onset time, while maintaining reasonable fidelity for the more challenging drag coefficient. Beyond regression, the ANN is further evaluated as a generative model to reconstruct flow behavior at unseen Re. A hierarchical refinement strategy is introduced, and results show that when trained on high-fidelity data, ANN-based models can serve as efficient and reliable surrogates for complex fluid dynamics problems.

physics.flu-dyn↗

arXiv2026

Impact of non-equilibrium radiation in a high-enthalpy inductively coupled plasma wind tunnel

High-power inductively coupled plasma (ICP) wind tunnels are widely used to reproduce high-enthalpy environments relevant to atmospheric entry and hypersonic testing. Despite their importance, radiative heat transfer in ICP facilities is commonly neglected or modeled using simplified optically thin assumptions, and the impact of non-equilibrium radiation on plasma dynamics remains poorly quantified. In this work, a loosely coupled, multi-physics framework is developed to systematically investigate radiative cooling effects in the 350 kW Plasmatron X facility at the University of Illinois Urbana-Champaign. The approach self-consistently couples a magnetohydrodynamic plasma framework with a spectral radiative transport solver, eliminating the need for optically thin or empirical models. Simulations are performed for nitrogen and air plasmas over a wide range of operating pressures (1-101 kPa) and powers (100-350 kW). The results reveal a strong pressure dependence of radiative losses, with radiation contributing negligibly at low pressures, but becoming a dominant energy sink at elevated pressures. At atmospheric pressure, radiative losses account for up to approximately 32% and 22% of the input power for nitrogen and air plasmas, respectively, leading to substantial reductions in core plasma temperatures. Nitrogen plasmas consistently exhibit higher radiative losses than air as a result of increased concentrations of radiatively active species and higher electron number densities. Pressure-power maps of radiative heat loss relative to input power are constructed to quantify combined operating effects and to provide guidance for facility operation and modeling fidelity. Finally, an assessment of self-absorption demonstrates that the Plasmatron X torch operates predominantly in an optically thin regime, even at the highest power and pressure conditions considered.

physics.plasm-ph↗

arXiv2026

Partial Kondo Screening Solves the Mystery of Rare Earth Tetraborides

We invoke a new mechanism to account for multiple magnetization plateaus observed in rare-earth tetraborides. Using a combination of hybrid and semiclassical Monte Carlo simulations of the Kondo lattice model (KLM) on the Shastry-Sutherland lattice (SSL), we find robust magnetization plateaus at fractions 1/6, 2/9, 1/4, 1/3, 1/2, 2/3 and 3/4 of the saturation magnetization. We find that most of the plateau states are partially Kondo screened and emerge from the field-tuning of a complex three-way competition between the kinetic energy, the Kondo coupling, and the magnetic frustration. Most remarkably, the unusual magneto-transport reported in ErB$_4$ and TmB$_4$ admits an unexpectedly simple explanation within our mechanism. This work not only provides an elegant and simple solution to the long-standing puzzle of metamagnetism and anomalous magnetotransport in RB$_4$, but also introduces a novel mechanism to predict and discover new correlated phases in frustrated Kondo lattices.

cond-mat.str-el↗

arXiv2026

Valence Bond Glass and Glassy Spin Liquid in Disordered Frustrated Magnets

The absence of conventional magnetic order together with anomalous low-temperature magnetic heat capacity is often interpreted as evidence for quantum spin liquid ground states in frustrated magnets. Using a recently developed semiclassical Monte Carlo approach, we show that similar thermodynamic signatures arise in the highly frustrated regime of the disordered spin-1/2 J1-J2 Heisenberg model on the square lattice. By analyzing the freezing parameters, the distribution of spin-spin correlations, and the specific heat, we identify the ground state as a valence-bond glass that melts into a glassy spin liquid at finite temperatures. We show that the low-temperature specific-heat anomaly originates from collective singlet excitations, and consequently it is insensitive to external magnetic fields. This leads to a robust experimental signature of the valence bond glass phase and a completely new interpretation of the thermodynamic data on disordered spin-liquid candidate materials.

cond-mat.str-el↗

arXiv2026

Spin qubit gates via phonon buses in electron nanowires

Scalable architectures for quantum computing using semiconductor quantum dots require interactions between qubits beyond adjacent quantum dots. Here, we propose using nanowires of electrons to mediate the interaction between two quantum dots. Virtual phonons in the linear chain of electrons can mediate an interaction that gives rise to effective spin-spin coupling of the electrons in distant quantum dots. We find coupling strengths of more than 30 MHz for experimentally realisable parameters in GaAs quantum dots.

cond-mat.mes-hall↗

arXiv2026

Stochastic Evolution of Galactic Star Formation with Halo Coupling, AGN Quenching and Hopf Bifurcation Dynamics

We present a computational framework for galactic evolution based on a coupled stochastic nonlinear oscillator, implemented with the \textbf{Stochastic Hopf Engine}. Gas density ($G$) and star formation rate ($S$) co-evolve through a supercritical Hopf bifurcation, capturing the transition from quiescent stability to merger-driven starbursts. Scatter in dark matter halo properties, modeled as multiplicative noise via the \textbf{Euler--Maruyama method}, broadens the bifurcation into a regime where noise-induced bursts occur below the deterministic threshold. Simulations reveal a periodic signature, the \textbf{Galactic Heartbeat}, emerging as a deterministic limit cycle validated by the \textbf{data3} resonance peak in the star-formation spectrum. A radial reduction yields an effective \textbf{Fokker--Planck equation} for burst amplitude; its stationary solution matches numerical PDFs, providing statistical closure. Including differential shear $\Omega(r)$ and spatially varying bifurcation fields reproduces spiral morphologies and AGN-driven quenching. Driving the growth parameter sub-critical ($r_{agn} < 0$) yields ``Red and Dead'' cores via attractor collapse. Dark matter halo scatter suppresses mean star formation while enhancing intermittency, offering a minimal yet interpretable framework linking local feedback and global potentials to macroscopic galactic evolution.

astro-ph.GA↗

arXiv2026

Self-correction phase transition in the dissipative toric code

We analyze a time-continuous version of a cellular automaton decoder for the toric code in the form of a Lindblad master equation. In this setting, a self-correcting quantum memory becomes a thermodynamical phase of the steady state, which manifests itself through the steady state being topologically ordered. We compute the steady state phase diagram, finding a competition between the error correction rate and the update rate for the classical field of the cellular automaton. Strikingly, we find that self-correction of errors is possible even in situations where conventional quantum error correction does not have a finite threshold.

quant-ph↗

arXiv2026

Evaluating Extremely Low-Resource Machine Translation: A Comparative Study of ChrF++ and BLEU Metrics

Evaluating machine translation (MT) quality in extremely low-resource language (ELRL) scenarios poses unique challenges, as widely used metrics such as BLEU, effective in high-resource settings, often misrepresent quality in data-scarce contexts. This work presents a comparative analysis of BLEU, an n-gram-based metric, and ChrF++, a character-based metric, for MT evaluation in ELRL settings. We examine how each metric responds to translation artifacts, including hallucinations, repetition, source-text copying, and diacritic (\textit{matra}) variations across three ELRLs: Magahi, Bhojpuri, and Chhattisgarhi, with a focus on outputs from large language models (LLMs) and neural MT (NMT) systems. While recent work often relies solely on ChrF++, our findings show that BLEU, despite its lower absolute scores, provides complementary lexical-precision insights that improve interpretability.

cs.CL↗

arXiv2026

From Coils to Surface Recession: Multiphysics Simulation of Ablation in ICP Wind Tunnels

This work presents a multi-solver, coupled computational framework for predicting the thermo-chemical material response of thermal protection systems in inductively coupled plasma (ICP) wind tunnels. The framework integrates a high-fidelity Navier-Stokes plasma solver, an electromagnetic field solver, and a discontinuous-Galerkin material response solver using a partitioned coupling strategy. This enables an ab initio, end-to-end simulation of the 350 kW Plasmatron X facility at the University of Illinois Urbana-Champaign (UIUC), including plasma generation, electromagnetic heating, near-wall thermochemistry, and time-accurate material ablation. The model captures key ICP physics such as vortex-mode recirculation, Joule-heating-driven plasma formation, and Lorentz-force-induced flow confinement, and accurately predicts the transition from subsonic to supersonic jet behavior at low pressures. Validation against cold-wall calorimetry shows that predicted stagnation-point cold-wall heat fluxes fall well within experimental uncertainty, while coupled ablation simulations accurately reproduce measured stagnation temperature histories and recession rates with errors below 12% and 10%, respectively. Remaining discrepancies are attributed to uncertainties in power-coupling efficiency, equilibrium ablation modeling, and material property datasets. Sensitivity analyses reveal that a 10% variation in system efficiency can induce changes of up to 11% in steady- state surface temperature and 17% in recession rate, whereas an equivalent variation in material thermal conductivity results in comparatively minor deviations of about 1.5% and 0.5%, respectively. Overall, the framework demonstrates strong predictive capability for ICP wind tunnel environments and provides a foundation for improved design, interpretation, and planning of hypersonic material testing campaigns.

physics.plasm-ph↗

arXiv2026

Room-Temperature Terahertz Photoconductivity Polarity Switching in High Entropy Nickelates with Implications for Photonic Synapses

High entropy oxides (HEO) hold the potential to revolutionize the conventional material paradigms by leveraging high order of chemical disorder that induces highly desirable exotic phases for advanced applications. Here, we devise a methodology to enhance the efficiency of an artificial photonic synapse using a high entropy rare earth nickelate. Combined with epitaxial strain, we show that high entropy can further manipulate the phase of these locally disordered materials. Using time-averaged and time resolved Terahertz (THz) spectroscopy as dynamic probe, for the first time we show a rare combination of i) crystal axis dependent insulator to metal THz electronic phase transition and ii) coexistence of negative and positive THz photoconductivity at room temperature. Detailed analysis within theoretical models, including density functional theory (DFT)-based band structure calculations, suggest origin of these properties as disproportionate ordering of oxygen vacancies. Based on these findings, a conceptual THz-based artificial photonic synapse is proposed. This work underlines the pivotal role of HEO in advancing diverse THz functionalities, representing a critical step toward futuristic applications like THz-based high-speed computing and communication with an emphasis in THz frequency domain.

cond-mat.mtrl-sci↗

arXiv2025

Thermal Evolution of Skyrmions in Synthetic Ferrimagnets of Co/Gd Heterostructure for Topological Spintronic Applications

Synthetic ferrimagnetic (SFiM) multilayers offer a versatile platform for hosting skyrmions with tunable magnetic properties, combining the advantages of ferromagnets and antiferromagnets. Unlike synthetic antiferromagnets, SFiMs retain a finite magnetization that allows direct observation of magnetic textures while still benefiting from reduced dipolar fields and a suppressed skyrmion Hall effect. However, a systematic investigation of their temperature and field dependent magnetization evolution, including the labyrinthine-to-skyrmion transition in Co/Gd-based SFiMs, remains less explored. Here, we demonstrate the stabilization of 70 nm-radius skyrmions at room temperature and reveal how the Co and Gd sublattices influence the temperature-dependent net magnetization. Further, we develop a microscopic spin model for SFiM incorporating the relevant magnetic interactions, which reproduces the experimental observations and captures the temperature-dependent magnetic phase evolution. This framework highlights the interplay of fundamental interactions controlling skyrmion stability in SFiM and provides a pathway for engineering heterostructures for topological spintronic applications.

cond-mat.mtrl-sci↗

arXiv2025

Harnessing Multifractality to Enhance Thermal Stability in Mixed-Phase Vanadium Oxide Thin Films

Vanadium oxide thin films exhibit temperature-driven electronic transitions desirable for sensing and microelectronic applications, yet their performance is often limited by thermal hysteresis. This study demonstrates that electronic stability is governed not simply by roughness or crystallinity but by a unique combination of surface morphological complexity and thermal hysteresis, revealed across films deposited with varying working pressure using Direct Current/Radio Frequency magnetron sputtering. Specifically, the film grown at 15 mTorr shows a distinct convergence of highest morphological vertical complexity and lowest thermal hysteresis, exhibiting nearly reversible transport with activation energies ranging from 0.26 to 0.28 eV and negative temperature coefficients of resistance between -0.0337 and -0.035 K-1. While conventional roughness metrics and mono-fractal parameters do not capture this behavior, multifractal detrended fluctuation analysis uncovers a pronounced peak in multifractality strength, which correlates inversely with thermal hysteresis. This highlights multifractality strength as a predictive descriptor of electronic stability, identifying a multiscale structural signature that enhances stress accommodation during thermal cycling. These results define an optimal deposition window and provide a morphology-guided pathway for developing thermally robust mixed-phase vanadium oxide films.

cond-mat.mtrl-sci↗

arXiv2025

MHR: Momentum Human Rig

We present MHR, a parametric human body model that combines the decoupled skeleton/shape paradigm of ATLAS with a flexible, modern rig and pose corrective system inspired by the Momentum library. Our model enables expressive, anatomically plausible human animation, supporting non-linear pose correctives, and is designed for robust integration in AR/VR and graphics pipelines.

cs.GR↗

arXiv2025

Role of varying Reynolds number for flow past a rotating cylinder at high rotation rate

The present study reports comprehensive bifurcation analysis of flow past a rotating cylinder at a fixed rotation rate by varying free-stream Reynolds number ($Re_{\infty}$) from 1000-6000 in intervals of 50. Two-dimensional compressible Navier-Stokes equations are solved using dispersion relation preserving numerical methods over 101 test cases, amounting to $10^6$ core hours of computing. The dataset produced from high-fidelity simulations serve as useful benchmarking tools for testing compressible flow solvers, estimating unsteady force distribution and vorticity dynamics. For moderate $Re_{\infty}$, rotation induces circulation that reduces pressure drag with increasing $Re_{\infty}$. For higher $Re_{\infty}$, boundary layer becomes thinner with suppressed flow separation, but effect of rotation saturates. Thus, benefits of increasing $Re_{\infty}$ taper off and pressure recovery stalls. The bifurcation analysis reveals a critical $Re_{\infty}$ of 5650 beyond which global behavior of Magnus-Robins effect changes significantly. Supercritical flow is receptive to time-dependent instabilities and structures in wake of the cylinder become dynamically unstable. Even small changes in $Re_{\infty}$ leads to different instantaneous force distributions and sharp fluctuations in lift and drag calculations. Stronger, coherent vortices in the wake generate consistent, high-energy periodic signals, contributing to strong Fourier amplitudes in spectra. An artificial neural network (ANN) is trained using simulation datasets to serve as fast, inexpensive alternatives for calculating lift, drag, and onset time of instability. The ANN reduces time required for simulation by 99.9\%, enabling dense parametric sweeps. Maximum accuracy achieved for the ANN is between 90-99\% for the parameters examined.

physics.flu-dyn↗