arXiv subjects

Chenxi Hu

Publications and source records attributed to Chenxi Hu.

At least 19 recordsLinked to original sources

arXiv2026

Frequency-Calibrated Decision-Focused Learning for BESS Dispatch Under Renewable Uncertainty

Battery energy storage system (BESS) flexibility depends on state of charge and intertemporal operating constraints. Renewable forecast errors of similar magnitudes can lead to different BESS dispatch results. An error may alter early charging or discharging decisions and reduce the flexibility available for subsequent regulation, while another may have little effect on the resulting schedule. To account for these unequal consequences, this paper proposes Frequency-Calibrated Decision-Focused Learning (FC-DFL), which trains renewable predictors using the realized operating costs of their induced BESS schedules. Gaussian smoothing and solver-based zeroth-order estimation provide training feedback without requiring differentiating through the non-differentiable mixed-integer BESS dispatch problem. The same decision feedback is further combined with prediction sensitivity to develop a decision-oriented XAI method that identifies the input features and forecast periods most consequential to BESS operation. Case studies using German wind and solar data show that FC-DFL reduces the integrated absolute frequency-deviation metric by 6.05% and decision regret by 23.92% relative to accuracy-based predict-then-optimize method. The XAI analysis further demonstrates that the identified high-impact forecast periods produce 2.84 times larger average absolute changes in realized operating cost than low-impact periods. These findings highlight the importance of incorporating downstream decision-making outcomes into renewable forecasting and demonstrate the value of FC-DFL in improving BESS regulation and providing operationally meaningful forecast explanations.

eess.SY↗

arXiv2026

Multi-Timescale Deliverability of Shared Hybrid Storage for AI Data Center Campuses

Artificial intelligence data center (AIDC) campuses contain sustained power variations and subsecond power fluctuations, requiring both sufficient stored energy and fast response. A shared hybrid energy storage system (HESS) can mitigate aggregate AIDC power deviations, but interval-based schedules cannot capture the delay between commanded and delivered power at the campus point of common coupling (PCC). To this end, this paper proposes a multi-timescale scheduling model for mitigating AIDC power fluctuations with the shared HESS. First, it models electrical pooling at the campus bus and constrains the shared HESS to a single PCC injection. Second, a two-layer model passes five-minute scheduled power, energy state, and reserve capacity to a causal fast response model that distinguishes commanded power from delivered power. Case studies using the National Laboratory of the Rockies (NLR) GenAI Power Profiles Dataset show reductions in both PCC and fast integral absolute error (IAE), while retaining most of the improvement available under ideal response.

eess.SY↗

arXiv2026

Data Center Load Accommodation under QoS and Checkpoint-Induced Frequency Security Constraints

The rapid growth of artificial intelligence workloads is accelerating the expansion and grid interconnection of data centers (DCs). Existing data center load accommodation (DCLA) assessments primarily focus on grid and resource adequacy and often imply that greater workload flexibility always increases accommodable capacity. However, workload flexibility has a dual effect. Workload deferral and temporal shifting can alleviate network congestion, whereas synchronized checkpointing of training workloads can cause large and rapid load reductions, threatening frequency stability. Therefore, DCLA cannot be determined solely by grid adequacy. This paper develops an optimal DCLA method that jointly accounts for workload quality of service (QoS), checkpoint-induced frequency security (FS), and grid adequacy. First, a QoS-aware model characterizes workload-dependent flexibility while ensuring service feasibility. Second, checkpoint-induced FS constraints are developed for periodic and event-triggered power variations, with key frequency-security limits reformulated in algebraic form. Third, a two-stage lexicographic model is developed to maximize DCLA and minimizes the required energy storage (ES) deployment, and then optimizes system operation under the given DC and ES siting and sizing decisions. Case studies show that the proposed method avoids a 40.3% overestimation of DCLA compared with the grid-adequacy benchmark. QoS is the primary limiting factor, while checkpoint-induced frequency-security constraints become more restrictive as training workloads increase. These findings highlight that reliable DCLA should account for both its grid-support value and the associated QoS and FS risks.

eess.SY↗

arXiv2026

Converter-Aware Coordination of Computing Flexibility in Behind-the-Meter Data Centers

Data centers are emerging as large electricity consumers whose demand can be reshaped through workload scheduling. However, computing flexibility in behind-the-meter facilities does not automatically translate into electrically realizable power flexibility. Ignoring converter and network operating limits may cause converter overloads, protection trips, and interruptions to power supply and computing services. This paper proposes a converter-aware intraday scheduling methodology. First, a task-level mixed-integer linear formulation preserves synchronous training configurations and completion requirements, while Minkowski-sum aggregation and Shapley--Folkman convexification yield a linear inference-pool representation without unit-level binary variables. The latter admits a conditional relative relaxation-gap bound that decreases inversely with pool size. Second, a four-node electrical model links computing demand, on-site injections, and grid exchange to converter and network limits through lifted voltage variables and second-order cone relaxation. Finally, a mixed-integer model coordinates workloads, photovoltaic generation, energy storage, and grid purchases. Case studies show that electrical limits reshape scheduling and renewable utilization, while on-site resources reduce grid dependence. Under the studied conditions, fixed training cannot support full inference service within voltage limits, whereas coordinated scheduling achieves negligible inference shortage and passes verification against the original alternating-current model. These findings demonstrate the role of computing--electrical coordination in realizing the operational value of workload flexibility.

eess.SY↗

arXiv2026

From Research Goals to Scientific Closure: Evaluating Agents Through Evidence-Supported Physics Research

Can AI agents organize scientific investigations that lead to well-supported conclusions? To evaluate this capability, we introduce PhysClosureBench, a benchmark of 50 expert-verified, goal-specified research tasks spanning strong-field, attosecond, and neighboring physics areas. Each task specifies the research objective and essential scientific constraints, while leaving the agent to organize the investigation and determine how to generate and validate the evidence needed to answer the question. A 20-criterion Atomic Score measures local scientific accomplishment, while Research Closure (RC0--RC4) assesses how fully the investigation resolves the specified research question based on whether its indispensable evidence requirements and applicable scope conditions are satisfied. We evaluate seven frontier LLMs, where Claude-Opus-5 achieves the highest mean Atomic Score (6.29/10) but reaches RC4 in only 32.7% of attempts. Across the benchmark, 36.2% of attempts scoring at least 8/10 still fall short of RC4, while all seven models reach RC4 on substantially fewer tasks in all three runs than in at least one. Case-level analysis further illustrates how misinterpretation of generated results and unresolved physical applicability limits can lead to incomplete closure. Together, these results reveal a persistent gap between AI agents' local scientific accomplishments and their ability to achieve complete research closure.

cs.AI↗

arXiv2026

NOVA-CIM: Noise- and Correlation-Tolerant Stochastic Interfaces for Analog Compute-in-Memory

Analog compute-in-memory (CIM) enables energy-efficient model acceleration, but its reliance on ADC-based readout, which directly quantizes noisy column currents, makes inference accuracy highly sensitive to analog read noise, active-row scaling, and ADC precision. In this paper, we present NOVA-CIM, a noise- and correlation-tolerant stochastic interface for analog CIM by replacing multi-bit ADC readout with random-reference 1-bit sensing and reconstructing results through lightweight counting. By converting column currents into comparison probabilities, this probability-domain readout averages zero-mean dynamic read noise over stochastic samples while reducing dependence on high-resolution ADCs. We provide a unified robustness analysis showing that dynamic read noise is suppressed through temporal averaging and that spatial input-bitstream correlation increases instantaneous current variance rather than introducing first-order MAC bias. MAC-level experiments and end-to-end evaluation on ViT-Base validate the analysis: under read noise, Top-1 accuracy remains 84.48% near the 84.51% bfloat16 (BF16) baseline; under stochastic number generator (SNG) reuse, MAC bias stays near zero while root-mean-square error (RMSE) and stochastic cross-correlation (SCC) grow as predicted.

cs.AR↗

arXiv2026

Event-Structured Physics-Informed Neural Networks for Differentiable Critical Clearing Boundaries

Transient-stability assessment determines whether a power system can recover after a disturbance and is therefore essential to preventing generator trips and cascading outages. A key metric is the critical clearing time (CCT), which specifies the maximum time available to clear a fault before synchronism is lost. Reliable CCT estimation is challenging because complicated fault-clearing dynamics require repeated simulations over many fault severities and clearing times. We propose an event-structured physics-informed neural network (ES-PINN) that aligns its representation with the pre-fault, fault-on, and post-clearing swing dynamics and enforces exact state chaining across event interfaces. A smooth trajectory-induced stability margin defines a differentiable approximation of the CCT boundary, enabling accurate boundary extraction, local sensitivity analysis, and optional direct CCT prediction through a distilled readout. We further prove a local residual-to-trajectory-to-CCT error estimate, in which exact event chaining eliminates separate state-interface defect terms. Experiments on IEEE 9-, 14-, and 30-bus systems show that ES-PINN consistently improves held-out trajectory and stability-boundary accuracy over matched neural-surrogate baselines across mechanical and electrical contingencies with multiple clearing configurations. Additional full-network DAE validation, multi-fault experiments, and runtime analyses further demonstrate the effectiveness and computational efficiency of the proposed framework.

cs.LG↗

arXiv2026

GEAR: Reconstruction of Classical Paintings via Geometry Grounding and Appearance Restitution

Classical paintings preserve rich spatial, cultural, and historical content, making their reconstruction as explorable 3D scenes valuable for digital preservation, immersive exhibition, and cultural engagement. Yet, unlike photographs, they often depict scenes in a single-view, stylized manner, with weak perspective, lighting, and depth cues. Existing 3D reconstruction methods are largely built on natural-image priors, making it difficult to recover geometrically plausible and visually faithful 3D representations from such inputs. To address this challenge, we introduce Classical Painting-to-3D (CP3D), a new task that aims to recover a 3D representation from a single classical painting while jointly ensuring geometric plausibility, appearance fidelity to the source artwork, and plausible novel-view synthesis. We further propose GeAR, a training-free two-stage framework for Geometry Grounding and Appearance Restitution. GeAR first converts the input painting into a geometry-grounded representation with more coherent shading and illumination cues, improving the stability of 3D Gaussian reconstruction. It then restores artwork-faithful appearance across views under spatial constraints and multi-view consistency, recovering the painterly textures and details weakened during grounding. In addition, we construct HeriArch, a curated benchmark of 10,160 high-resolution classical artworks for systematic evaluation of CP3D. Extensive experiments and user studies show that GeAR consistently outperforms strong baselines in geometric plausibility, appearance fidelity, and human preference. Code and dataset will be released publicly.

cs.MM↗

arXiv2026

A Unified Generative Framework for Scalable Chemical Reaction Network Exploration

Chemical reaction networks (CRNs) are crucial for understanding reaction mechanisms and guiding chemical synthesis, yet the computational exploration remains limited by the combinatorial growth of chemical space, the reliability of reaction path screening, and the cost of evaluating thermodynamic and kinetic properties. Here, we present ByteCRN, an end-to-end framework for computational CRN exploration that combines chemically informed reaction enumeration with generative transition state modeling. A key component of our framework is a generative rectified flow architecture for both transition state generation and reaction validation, where it maps reactant-product pairs to candidate transition state structures and verifies connectivity by mapping back to reactants and products. This unified generative strategy replaces the most expensive steps of conventional computational workflows, namely iterative transition state search and intrinsic reaction coordinate validation, within a complete CRN construction pipeline. ByteCRN delivers a 10--100-fold acceleration over traditional workflows while maintaining high predictive fidelity for individual reactions. At the network scale, it effectively prunes $\sim$70-90% of the enumerated reactions, streamlining the exploration of complex reaction space. Its utility is illustrated through the discovery of novel pathways involving cyanoacetaldehyde and the successful modeling of the challenging $γ$-ketohydroperoxide network, demonstrating a practical, scalable approach to autonomous chemical exploration.

physics.chem-ph↗

arXiv2026

MatterSim-MT: A multi-task foundation model for in silico materials characterization

Accurate property characterization is a major bottleneck in materials design. While first-principles methods and task-specific machine-learning models have driven important progress, they remain fundamentally limited in scalability and generalizability across the vast space of structures and properties relevant to real-world materials design. We present MatterSim-MT, a multi-task foundation model for in silico materials simulation and property characterization. The model is pretrained on over 35 million first-principles-labeled structures covering 89 elements, temperatures up to 5000 K and pressures up to 1000 GPa, and is fine-tuned on various properties including Bader charges, magnetic moments, Born effective charges, and dielectric matrices. Out of the box, MatterSim-MT not only serves as a foundation model for predicting material structure, dynamics and thermodynamics, its multi-task architecture also enables a wide range of complex simulations that cannot be captured by potential energy surfaces alone. For example, we demonstrate pressure-dependent LO-TO phonon splitting in SiC with close agreement with experiment, electric hysteresis in ferroelectric BaTiO3, and the cationic-to-anionic redox transition during delithiation of a Li-rich cathode material. Finally, we show that MatterSim-MT scales well with more data and parameters, can be efficiently fine-tuned to higher levels of theory, and can be efficiently extended to new systems via active learning. Overall, we believe this approach provides a scalable route to accurate in silico materials characterization.

cond-mat.mtrl-sci↗

arXiv2026

Resilient Load Forecasting under Climate Change: Adaptive Conditional Neural Processes for Few-Shot Extreme Load Forecasting

Extreme weather can substantially change electricity consumption behavior, causing load curves to exhibit sharp spikes and pronounced volatility. If forecasts are inaccurate during those periods, power systems are more likely to face supply shortfalls or localized overloads, forcing emergency actions such as load shedding and increasing the risk of service disruptions and public-safety impacts. This problem is inherently difficult because extreme events can trigger abrupt regime shifts in load patterns, while relevant extreme samples are rare and irregular, making reliable learning and calibration challenging. We propose AdaCNP, a probabilistic forecasting model for data-scarce condition. AdaCNP learns similarity in a shared embedding space. For each target data, it evaluates how relevant each historical context segment is to the current condition and reweights the context information accordingly. This design highlights the most informative historical evidence even when extreme samples are rare. It enables few-shot adaptation to previously unseen extreme patterns. AdaCNP also produces predictive distributions for risk-aware decision-making without expensive fine-tuning on the target domain. We evaluate AdaCNP on real-world power-system load data and compare it against a range of representative baselines. The results show that AdaCNP is more robust during extreme periods, reducing the mean squared error by 22\% relative to the strongest baseline while achieving the lowest negative log-likelihood, indicating more reliable probabilistic outputs. These findings suggest that AdaCNP can effectively mitigate the combined impact of abrupt distribution shifts and scarce extreme samples, providing a more trustworthy forecasting for resilient power system operation under extreme events.

cs.LG↗

arXiv2025

Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge

Cardiovascular health is vital to human well-being, and cardiac magnetic resonance (CMR) imaging is considered the {clinical reference standard} for diagnosing cardiovascular disease. However, its adoption is hindered by long scan times, complex contrasts, and inconsistent quality. While deep learning methods perform well on specific CMR imaging {sequences}, they often fail to generalize across modalities and sampling schemes. The lack of benchmarks for high-quality, fast CMR image reconstruction further limits technology comparison and adoption. The CMRxRecon2024 challenge, attracting over 200 teams from 18 countries, addressed these issues with two tasks: generalization to unseen {modalities} and robustness to diverse undersampling patterns. We introduced the largest public multi-{modality} CMR raw dataset, an open benchmarking platform, and shared code. Analysis of the best-performing solutions revealed that prompt-based adaptation and enhanced physics-driven consistency enabled strong cross-scenario performance. These findings establish principles for generalizable reconstruction models and advance clinically translatable AI in cardiovascular imaging.

eess.IV↗

arXiv2025

C3D-AD: Toward Continual 3D Anomaly Detection via Kernel Attention with Learnable Advisor

3D Anomaly Detection (AD) has shown great potential in detecting anomalies or defects of high-precision industrial products. However, existing methods are typically trained in a class-specific manner and also lack the capability of learning from emerging classes. In this study, we proposed a continual learning framework named Continual 3D Anomaly Detection (C3D-AD), which can not only learn generalized representations for multi-class point clouds but also handle new classes emerging over time.Specifically, in the feature extraction module, to extract generalized local features from diverse product types of different tasks efficiently, Kernel Attention with random feature Layer (KAL) is introduced, which normalizes the feature space. Then, to reconstruct data correctly and continually, an efficient Kernel Attention with learnable Advisor (KAA) mechanism is proposed, which learns the information from new categories while discarding redundant old information within both the encoder and decoder. Finally, to keep the representation consistency over tasks, a Reconstruction with Parameter Perturbation (RPP) module is proposed by designing a representation rehearsal loss function, which ensures that the model remembers previous category information and returns category-adaptive representation.Extensive experiments on three public datasets demonstrate the effectiveness of the proposed method, achieving an average performance of 66.4%, 83.1%, and 63.4% AUROC on Real3D-AD, Anomaly-ShapeNet, and MulSen-AD, respectively.

cs.CV↗

arXiv2025

Optimizing Metal-Organic Chemical Vapor Deposition for Ultrawide Band-Gap MgSiN2 Thin Films

Orthorhombic II-IV nitride semiconductors offer an expanded and more tunable material set with unique properties, while maintaining close compatibility with the wurtzite crystal structure of the III-nitrides. In particular, MgSiN2, a II-IV nitride closely lattice matched to GaN and AlN has a band gap suitable for photonic applications in the UV-C wavelength region. MgSiN2 is also a promising candidate to exhibit ferroelectricity, which has only been observed in very few nitride materials. This study builds on our previous work on the metal-organic chemical vapor deposition (MOCVD) of MgSiN2 thin films grown on GaN-on-sapphire and c-plane sapphire substrates by exploring higher growth temperature windows, resulting in higher crystalline quality and improved interfaces. Correlations between the growth conditions (Mg:Si precursor molar flow rate ratio, reactor pressure, and growth temperatures from 900C to 960C) and the resultant film quality are investigated for films grown on GaN-on-sapphire. High-resolution transmission electron microscopy (HR-TEM) reveals high-quality orthorhombic single-crystal MgSiN2, confirming successful epitaxial growth on GaN. Optical transmittance measurements indicate the direct band gap is 6.34-6.36 eV and indirect band gap is 5.77-5.81 eV, affirming the realization of an ultrawide-band gap II-IV nitride semiconductor that is structurally compatible with existing III-nitride device platforms.

cond-mat.mtrl-sci↗

arXiv2025

Metal-organic chemical vapor deposition of MgGeN2 films on GaN and sapphire

MgGeN2 films were synthesized using metal-organic chemical vapor deposition on GaN/c-sapphire templates and c-plane sapphire substrates. Energy-dispersive X-ray spectroscopy was used to estimate the cation composition ratios. To mitigate magnesium evaporation, the films were grown at pyrometer temperature 745 °C with a wafer rotation speed of 1000 rpm. Growth rates were determined by fitting energy-dispersive X-ray spectroscopy spectra to film thicknesses using NIST DTSA-II software. The thickness estimates determined by this method were consistent with scanning transmission electron microscopy measurements done for selected samples. Scanning electron microscopy images revealed faceted surfaces indicative of a tendency toward three-dimensional growth. X-ray diffraction spectra confirmed that the films were highly crystalline and exhibited preferential orientation in alignment with the substrate. Atomic force microscopy measurements show that film thicknesses are consistent across samples grown on both GaN templates and sapphire substrates, with typical roughnesses around 10 nm. Transmittance spectra of films grown on double-side-polished sapphire substrates yielded band gaps of 4.28 +- 0.06 eV for samples exhibiting close-to-ideal stoichiometry. Comparison of the measured spectra with ab initio calculations are in good agreement both near the band gap and at higher energies where excitation is into higher-lying bands. These findings provide insight into the growth and characterization of MgGeN2, contributing to the development of this material for potential applications in optoelectronics and power electronics.

cond-mat.mtrl-sci↗

arXiv2025

Metal-organic chemical vapor deposition of MgSiN$_{2}$ thin films

Orthorhombic-structured II-IV nitrides provide a promising opportunity to expand the material platform while maintaining compatibility with the wurtzite crystal structure of the traditional III-nitride material system. Among them, MgSiN$_{2}$ stands out due to its close compatibility with GaN and AlN and its theoretically predicted ultrawide direct band gap of 6.28 eV. In this work, the growth of MgSiN$_{2}$ thin films on GaN-on-sapphire and c-plane sapphire substrates was investigated using metal-organic chemical vapor deposition (MOCVD). MOCVD growth conditions were correlated with film quality and crystallinity for samples grown on GaN-on-sapphire substrates. The effects of Mg:Si precursor molar flow rate ratios and growth pressure at two different temperatures, 745$^{\circ}$C and 850$^{\circ}$C, were studied comprehensively. High-resolution scanning transmission electron microscopy (STEM) imaging confirmed the formation of high-quality, single-crystal MgSiN$_{2}$ films. Optical band gap extraction from transmittance measurements yielded direct band gap values ranging from 6.13 eV to 6.27 eV for samples grown under various conditions, confirming the realization of an ultrawide-band gap, III-nitride-compatible, II-IV-nitride material.

cond-mat.mtrl-sci↗

arXiv2024

Improve Myocardial Strain Estimation based on Deformable Groupwise Registration with a Locally Low-Rank Dissimilarity Metric

Background: Current mainstream cardiovascular magnetic resonance-feature tracking (CMR-FT) methods, including optical flow and pairwise registration, often suffer from the drift effect caused by accumulative tracking errors. Here, we developed a CMR-FT method based on deformable groupwise registration with a locally low-rank (LLR) dissimilarity metric to improve myocardial tracking and strain estimation accuracy. Methods: The proposed method, Groupwise-LLR, performs feature tracking by iteratively updating the entire displacement field across all cardiac phases to minimize the sum of the patchwise signal ranks of the deformed movie. The method was compared with alternative CMR-FT methods including the Farneback optical flow, a sequentially pairwise registration method, and a global low rankness-based groupwise registration method via a simulated dataset (n = 20), a public cine data set (n = 100), and an in-house tagging-MRI patient dataset (n = 16). The proposed method was also compared with two general groupwise registration methods, nD+t B-Splines and pTVreg, in simulations and in vivo tracking. Results: On the simulated dataset, Groupwise-LLR achieved the lowest point tracking errors and voxelwise/global strain errors. On the public dataset, Groupwise-LLR achieved the lowest contour tracking errors, reduced the drift effect in late-diastole, and preserved similar inter-observer reproducibility as the alternative methods. On the patient dataset, Groupwise-LLR correlated better with tagging-MRI for radial strains than the other CMR-FT methods in multiple myocardial segments and levels. Conclusions: The proposed Groupwise-LLR reduces the drift effect and provides more accurate myocardial tracking and strain estimation than the alternative methods. The method may thus facilitate a more accurate estimation of myocardial strains for clinical assessments of cardiac function.

eess.IV↗

arXiv2024

MatterSim: A Deep Learning Atomistic Model Across Elements, Temperatures and Pressures

Accurate and fast prediction of materials properties is central to the digital transformation of materials design. However, the vast design space and diverse operating conditions pose significant challenges for accurately modeling arbitrary material candidates and forecasting their properties. We present MatterSim, a deep learning model actively learned from large-scale first-principles computations, for efficient atomistic simulations at first-principles level and accurate prediction of broad material properties across the periodic table, spanning temperatures from 0 to 5000 K and pressures up to 1000 GPa. Out-of-the-box, the model serves as a machine learning force field, and shows remarkable capabilities not only in predicting ground-state material structures and energetics, but also in simulating their behavior under realistic temperatures and pressures, signifying an up to ten-fold enhancement in precision compared to the prior best-in-class. This enables MatterSim to compute materials' lattice dynamics, mechanical and thermodynamic properties, and beyond, to an accuracy comparable with first-principles methods. Specifically, MatterSim predicts Gibbs free energies for a wide range of inorganic solids with near-first-principles accuracy and achieves a 15 meV/atom resolution for temperatures up to 1000K compared with experiments. This opens an opportunity to predict experimental phase diagrams of materials at minimal computational cost. Moreover, MatterSim also serves as a platform for continuous learning and customization by integrating domain-specific data. The model can be fine-tuned for atomistic simulations at a desired level of theory or for direct structure-to-property predictions, achieving high data efficiency with a reduction in data requirements by up to 97%.

cond-mat.mtrl-sci↗