Searcharxiv⌕ Search

arXiv subjects

Jianhua Wang

Publications and source records attributed to Jianhua Wang.

At least 19 recordsLinked to original sources
arXiv2026

DBcover: A White-box SQL Test Generation Framework for Coverage Improvement

Relational Database Management Systems (RDBMSs) are the backbone of modern data-intensive applications, making reliability and robustness critical. However, achieving high coverage in RDBMS testing remains challenging because of large codebases and complex execution logic. Traditional fuzzing relies on random SQL generation and cannot capture the correspondence between SQL inputs and internal execution paths, while symbolic execution suffers from prohibitive cost and scalability limitations. We propose DBcover, an LLM-driven white-box SQL test generation framework based on contextual reasoning. DBcover uses lightweight dynamic analysis to extract SQL-to-path correspondence and call graphs as global context, and collects source-level information around target functions as local context. These contexts are organized in a unified knowledge graph for efficient retrieval and reuse. DBcover then performs two-phase test generation: it first selects a semantically relevant seed whose execution path is close to the uncovered target, and then guides the LLM with global and local context to generate SQL test cases that trigger previously uncovered code regions. Experiments show that DBcover achieves 80.1% and 82.3% coverage on PostgreSQL and MySQL, and is also effective on the enterprise RDBMS KingbaseES, demonstrating its practical applicability to closed-source systems.

cs.DB↗
arXiv2026

Integrated Energy System Optimal Operation in Coal District With Hydrogen Heavy Trucks

The coal industry contributes significantly to the social economy, but the emission of greenhouse gases puts huge pressure on the environment during mining, transportation, and power generation. In the integrated energy system (IES), current research on power-to-gas (P2G) technology mainly focuses on injecting hydrogen generated from renewable-energy-powered electrolysis into natural gas pipelines. However, hydrogen injection may cause irreversible hydrogen embrittlement of pipelines. In this paper, sufficient hydrogen is produced through both P2G technology and coal-to-hydrogen (C2H) based on coal gasification, considering the typical scenario where coal-producing regions are rich in both coal and renewable energy resources. To transport mined coal, hydrogen heavy-duty trucks provide a promising solution by consuming hydrogen in a timely manner, thereby avoiding potentially dangerous hydrogen injection into pipelines and reducing the need for relatively expensive hydrogen storage. An optimal scheduling model for an electric-gas IES is proposed based on second-order cone programming (SOCP). The proposed model innovatively studies a closed industrial loop including coal mining, hydrogen production, coal transportation by hydrogen trucks, and integrated energy systems to enhance renewable energy utilization and coordinate multiple energy carriers. Finally, a case study based on the IEEE 30-bus power system and the Belgium 24-node natural gas network is conducted. The results show that the proposed hydrogen production technology reduces typical daily operating costs by 7.7%. Under China's carbon emissions trading system, the operating costs of hydrogen heavy-duty trucks are reduced by 0.95% and 4.68% compared with electric vehicles and diesel trucks, respectively. Under Europe's stricter carbon emissions trading system, ...

eess.SY↗
arXiv2026

Ferroelectric Band Twinning from Pair-State Symmetry

Ferroelectric switching provides a nonvolatile way to control electronic structures, but a general symmetry rule connecting the full Bloch bands of two switchable polarization states is still lacking. Here, we introduce ferroelectric band twinning, a pair-state relation in which the bands of two opposite-polarization states are mapped onto each other by a non-inversion state-exchange symmetry. Using dichromatic groups, we derive the band-twinning rule and identify 11 ferroelectric band-twinning point-group classes. Screening the Ferroelectric Materials Database yields 16 candidate compounds, of which the two lattice-metric-preserving candidates, bulk gamma-Ag3SI and BaAl2O4, are selected for first-principles validation. For gamma-Ag3SI, we further show that the same pair-state symmetry controls the transformation of shift-current tensor components under polarization reversal. These results establish ferroelectric band twinning as a general symmetry framework for nonvolatile control of momentum-dependent electronic structures in ferroelectrics.

cond-mat.mtrl-sci↗
arXiv2026

Null-Space Constrained Low-Rank Adaptation for Response-Specified Large Language Model Unlearning

Large language model unlearning aims to suppress designated undesirable knowledge while preserving benign capabilities. Many unlearning objectives focus on suppressing undesired answers, while recent target-guided variants specify replacement behavior but still leave update locality largely unconstrained. This paper introduces \emph{Null-Space Constrained Response-Specified Unlearning} (NSRU), a projection-constrained low-rank framework for controlled LLM unlearning. NSRU uses an explicitly structured safe target response to specify the desired behavior for each forget query, while suppressing the original undesired content. To localize adaptation, NSRU estimates per-module retain subspaces from benign hidden representations and uses an orthogonal-projected low-rank parameterization to confine LoRA updates to the null space of the retain subspace. The resulting objective jointly optimizes safe-target learning, undesired-response suppression, and retention preservation under this constrained parameterization. We provide a local first-order analysis showing that the projected update reduces retain-side perturbations while preserving editable directions for shaping forget-query behavior. Experiments on TOFU show that NSRU effectively suppresses extractable forget-set knowledge while improving retain QA performance, model utility, and safe-target alignment over representative baselines. On WMDP, NSRU keeps hazardous-domain accuracy near the random-choice region while preserving broad and domain-adjacent MMLU utility. Ablation studies support the complementary roles of safe-target supervision, undesired-response suppression, retention loss, and null-space projected updates, while sensitivity and robustness analyses indicate stable behavior across the tested hyperparameter and prompt variations.

cs.AI↗
arXiv2026

Robust Driving Control for Autonomous Vehicles: An Intelligent General-sum Constrained Adversarial Reinforcement Learning Approach

Deep reinforcement learning (DRL) has demonstrated remarkable success in developing autonomous driving policies. However, its vulnerability to adversarial attacks remains a critical barrier to real-world deployment. Although existing robust methods have achieved success, they still suffer from three key issues: (i) these methods are trained against myopic adversarial attacks, limiting their abilities to respond to more strategic threats, (ii) they have trouble causing truly safety-critical events (e.g., collisions), but instead often result in minor consequences, and (iii) these methods can introduce learning instability and policy drift during training due to the lack of robust constraints. To address these issues, we propose Intelligent General-sum Constrained Adversarial Reinforcement Learning (IGCARL), a novel robust autonomous driving approach that consists of a strategic targeted adversary and a robust driving agent. The strategic targeted adversary is designed to leverage the temporal decision-making capabilities of DRL to execute strategically coordinated multi-step attacks. In addition, it explicitly focuses on inducing safety-critical events by adopting a general-sum objective. The robust driving agent learns by interacting with the adversary to develop a robust autonomous driving policy against adversarial attacks. To ensure stable learning in adversarial environments and to mitigate policy drift caused by attacks, the agent is optimized under a constrained formulation. Extensive experiments show that IGCARL improves the success rate by at least 27.9% over state-of-the-art methods, demonstrating superior robustness to adversarial attacks and enhancing the safety and reliability of DRL-based autonomous driving.

cs.LG↗
arXiv2026

One-dimensional electronics with edge states in two-dimensional altermagnets

The coupling between real-space inhomogeneities coordinates and spin (r-s) provides an alternative route to achieve efficient spin manipulation in spintronics beyond the conventional momentum-spin (k-s) coupling paradigm. Here we demonstrate an unexpected manifestation of one-dimensional (1D) r-s coupling in two-dimensional (2D) altermagnetic second-order topological insulators, where the spin-split floating edge states -- energetically isolated within the bulk band gap -- emerge and exhibit both Neel-vector-dependent and electrically tunable behaviors. The 1D edge-spin r-s coupling ensures carrier transport to be exclusively carried by the edge states with quantized spin conductance, giving rise to an unconventional edge tunnel magnetoresistance (edge-TMR) effect that can be switched On or Off. As a proof of concept, we computationally design an edge-TMR device based on Cr_2Se_2O monolayer to demonstrate its edge transportation and controllability via the Néel order or electric field. Our findings propose a general prototype altermagnetic device for next-generation low-dimensional spintronics.

cond-mat.mes-hall↗
arXiv2026

Pure Spin Photocurrent in Altermagnetic Photovoltaic Battery

Altermagnets, featuring momentum-dependent spin splitting without net magnetization, provide a promising platform for spintronic functionalities beyond conventional ferromagnets and antiferromagnets. Here, we propose an altermagnetic spin photovoltaic battery consisting of a nonmagnetic semiconducting layer sandwiched between two altermagnetic electrodes. Using first-principles quantum-transport simulations, we show that a V2Te2O/ZnSe/V2Te2O junction supports a pure spin photocurrent for opposite Néel vectors in the two altermagnetic electrodes, with spin-up and spin-down photocurrents equal in magnitude and opposite in sign. The effect persists under both linearly and circularly polarized light and remains tunable with photon energy and polarization angle. Our results establish a realistic route toward light-driven pure spin-current generation in altermagnetic junctions.

cond-mat.mes-hall↗
arXiv2026

New NanoSIMS Multielement Isotope Data Reveal CO Novae As Key Sources Of 13C-rich Presolar Silicon Carbide Grains

We present new multielement NanoSIMS isotopic measurements (C, N, Si, Mg-Al, Ti, and Ni) for four putative nova SiC grains and 79 AB SiC grains from the Murchison meteorite to reassess their stellar origins. High-resolution imaging and a revised Mg/Al relative sensitivity factor for SiC yield substantially improved 26Al/27Al ratios and the most reliable multielement characterization to date for 13C-rich presolar SiC grains. To interpret these data, we computed an expanded suite of hydrodynamic CO, ONe, and recurrent nova models spanning a range of white-dwarf masses and pre-enrichment parameters. When all isotopic systems are considered together (C, N, Mg-Al, Si, Ti, and Ni), the CO nova models provide the closest and most self-consistent match to both the putative nova grains and the subset of AB grains lacking s-process signatures. CO novae of low- to intermediate-mass naturally reproduce the observed 14N/15N-26Al/27Al trend, the Si isotope compositions of AB grains which dominantly reflect Galactic chemical evolution (GCE), and the mild Si isotope shifts in putative nova grains relative to the GCE trend defined by AB grains. In contrast, ONe and recurrent nova models fail multiple isotopic constraints simultaneously. These results demonstrate that low- to intermediate-mass CO novae (0.6-1.0 Msun) are the most plausible stellar sources of 13C-rich SiC dust lacking s-process signatures (1-2\% of all presolar SiC), and they establish a multielement, model-anchored framework for quantifying nova contributions to the dust reservoir in the interstellar medium.

astro-ph.SR↗
arXiv2026

Constraints for Nuclear Astrophysics from an Unusual Presolar Silicate-Oxide Aggregate Grain Found in Primitive Ordinary Chondrite Meteorite Hills 00526

We report O, Mg-Al, Si, Ca, and Ti isotopic data for an unusual presolar oxide/silicate aggregate grain, M526-69, previously reported in the primitive ordinary chondrite Meteorite Hills 00526. The $\approx 1μ$m aggregate consists of a Mg- and Ca-rich silicate, a Al-rich oxide, and a tiny TiO$_2$ grain. A large $^{18}$O depletion and high inferred $^{26}$Al/$^{27}$Al classifies M526-69 as a Group 2 grain. Both low-mass (LM) and intermediate-mass (IM) asymptotic giant branch (AGB) stars are considered viable candidate parent stars of Group 2 grains based on their O isotopes and inferred $^{26}$Al/$^{27}$Al ratios. The lack of a large $^{30}$Si excess in M526-69 strongly supports an LM-AGB origin for it and other Group 2 grains. The stable Mg, Ca, and Ti isotopes all reflect the initial composition of the parent star, set by galactic chemical evolution (GCE) processes. Presolar O-rich grains provide a better measure of the GCE trends for Ti isotopes than presolar SiC grains as the latter are also affected by neutron capture reactions in the parent stars. Most of the Mg, Ca, and Ti isotopic ratios in M526-69 are consistent with its parent star having metallicity lower than solar. However, small excesses in stable non-radiogenic $^{26}$Mg, $^{46}$Ti, and $^{44}$Ca do not fit this pattern and instead point to heterogeneous GCE processes, though quantitative modeling is needed to test this hypothesis. Multi-phase presolar grains are extremely valuable for nuclear astrophysics as they can both provide isotopic compositions for multiple elements that must be matched at a single time and place in a single star.

astro-ph.SR↗
arXiv2026

Static and Dynamic Graph Alignment Network for Temporal Video Grounding

Temporal Video Grounding (TVG) aims to localize temporal moments in an untrimmed video that semantically correspond to given natural language queries. Recently, Graph Convolutional Networks (GCN) have been widely adopted in TVG to model temporal relations among video clips and enhance contextual reasoning by constructing clip-level graphs. Despite their effectiveness, existing GCN-based TVG methods encounter three critical bottlenecks: 1) Most methods construct graph nodes using either static or dynamic features alone, resulting in incomplete visual representation and overlooking complementary semantics, 2) Most methods construct temporal graphs in a query-agnostic manner, leading to inefficient feature interaction within the temporal graph representation, and 3) Most methods often suffer from a single-granularity semantic matching, while direct training on complex temporal localization task may lead to slow convergence and suboptimal precision. To address these challenges, we propose Static and Dynamic Graph Alignment Network (SDGAN). First, SDGAN jointly exploits static and dynamic visual features to construct two complementary temporal graphs and performs Position-wise Nodes Alignment, enabling more expressive and robust visual representation. Second, SDGAN introduces Query-Clip Contrastive Learning and Adaptive Graph Modeling to explicitly align visual clips with their corresponding textual queries, yielding query-aware visual representations. Third, SDGAN incorporates multi-granularity temporal proposals within Progressive Easy-to-Hard Training Strategy, effectively bridging coarse-grained semantic localization and fine-grained temporal boundary refinement. Extensive experiments on three benchmark datasets demonstrate that SDGAN achieves superior performance across complex TVG scenarios. Codes and datasets are available at https://github.com/ZhanJieHu/SDGAN.

cs.CV↗
arXiv2026

Including Node Textual Metadata in Laplacian-constrained Gaussian Graphical Models

This paper addresses graph learning in Gaussian Graphical Models (GGMs). In this context, data matrices often come with auxiliary metadata (e.g., textual descriptions associated with each node) that is usually ignored in traditional graph estimation processes. To fill this gap, we propose a graph learning approach based on Laplacian-constrained GGMs that jointly leverages the node signals and such metadata. The resulting formulation yields an optimization problem, for which we develop an efficient majorization-minimization (MM) algorithm with closed-form updates at each iteration. Experimental results on a real-world financial dataset demonstrate that the proposed method significantly improves graph clustering performance compared to state-of-the-art approaches that use either signals or metadata alone, thus illustrating the interest of fusing both sources of information.

stat.ML↗
arXiv2026

TIP: Resisting Gradient Inversion via Targeted Interpretable Perturbation in Federated Learning

Federated Learning (FL) facilitates collaborative model training while preserving data locality; however, the exchange of gradients renders the system vulnerable to Gradient Inversion Attacks (GIAs), allowing adversaries to reconstruct private training data with high fidelity. Existing defenses, such as Differential Privacy (DP), typically employ indiscriminate noise injection across all parameters, which severely degrades model utility and convergence stability. To address those limitation, we proposes Targeted Interpretable Perturbation (TIP), a novel defense framework that integrates model interpretability with frequency domain analysis. Unlike conventional methods that treat parameters uniformly, TIP introduces a dual-targeting strategy. First, leveraging Gradient-weighted Class Activation Mapping (Grad-CAM) to quantify channel sensitivity, we dynamically identify critical convolution channels that encode primary semantic features. Second, we transform these selected kernels into the frequency domain via the Discrete Fourier Transform and selectively inject calibrated perturbations into the high-frequency spectrum. By selectively perturbing high-frequency components, TIP effectively destroys the fine-grained details necessary for image reconstruction while preserving the low-frequency information crucial for model accuracy. Extensive experiments on benchmark datasets demonstrate that TIP renders reconstructed images visually unrecognizable against state-of-the-art GIAs, while maintaining global model accuracy comparable to non-private baselines, significantly outperforming existing DP-based defenses in the privacy-utility trade-off and interpretability. Code is available in https://github.com/2766733506/asldkfjssdf_arxiv

cs.LG↗
arXiv2026

Sparse Threats, Focused Defense: Criticality-Aware Robust Reinforcement Learning for Safe Autonomous Driving

Reinforcement learning (RL) has shown considerable potential in autonomous driving (AD), yet its vulnerability to perturbations remains a critical barrier to real-world deployment. As a primary countermeasure, adversarial training improves policy robustness by training the AD agent in the presence of an adversary that deliberately introduces perturbations. Existing approaches typically model the interaction as a zero-sum game with continuous attacks. However, such designs overlook the inherent asymmetry between the agent and the adversary and then fail to reflect the sparsity of safety-critical risks, rendering the achieved robustness inadequate for practical AD scenarios. To address these limitations, we introduce criticality-aware robust RL (CARRL), a novel adversarial training approach for handling sparse, safety-critical risks in autonomous driving. CARRL consists of two interacting components: a risk exposure adversary (REA) and a risk-targeted robust agent (RTRA). We model the interaction between the REA and RTRA as a general-sum game, allowing the REA to focus on exposing safety-critical failures (e.g., collisions) while the RTRA learns to balance safety with driving efficiency. The REA employs a decoupled optimization mechanism to better identify and exploit sparse safety-critical moments under a constrained budget. However, such focused attacks inevitably result in a scarcity of adversarial data. The RTRA copes with this scarcity by jointly leveraging benign and adversarial experiences via a dual replay buffer and enforces policy consistency under perturbations to stabilize behavior. Experimental results demonstrate that our approach reduces the collision rate by at least 22.66\% across all cases compared to state-of-the-art baseline methods.

cs.LG↗
arXiv2025

LAMIC: Layout-Aware Multi-Image Composition via Scalability of Multimodal Diffusion Transformer

In controllable image synthesis, generating coherent and consistent images from multiple references with spatial layout awareness remains an open challenge. We present LAMIC, a Layout-Aware Multi-Image Composition framework that, for the first time, extends single-reference diffusion models to multi-reference scenarios in a training-free manner. Built upon the MMDiT model, LAMIC introduces two plug-and-play attention mechanisms: 1) Group Isolation Attention (GIA) to enhance entity disentanglement; and 2) Region-Modulated Attention (RMA) to enable layout-aware generation. To comprehensively evaluate model capabilities, we further introduce three metrics: 1) Inclusion Ratio (IN-R) and Fill Ratio (FI-R) for assessing layout control; and 2) Background Similarity (BG-S) for measuring background consistency. Extensive experiments show that LAMIC achieves state-of-the-art performance across most major metrics: it consistently outperforms existing multi-reference baselines in ID-S, BG-S, IN-R and AVG scores across all settings, and achieves the best DPG in complex composition tasks. These results demonstrate LAMIC's superior abilities in identity keeping, background preservation, layout control, and prompt-following, all achieved without any training or fine-tuning, showcasing strong zero-shot generalization ability. By inheriting the strengths of advanced single-reference models and enabling seamless extension to multi-image scenarios, LAMIC establishes a new training-free paradigm for controllable multi-image composition. As foundation models continue to evolve, LAMIC's performance is expected to scale accordingly. Our implementation is available at: https://github.com/Suchenl/LAMIC.

cs.CV↗
arXiv2025

Experiments reveal extreme water generation during planet formation

The most abundant type of planet discovered in the Galaxy has no analogue in our Solar System and is believed to consist of a rocky interior with an overlying thick H2 dominated envelope. Models have predicted that the reaction between the atmospheric hydrogen and the underlying magma ocean can lead to the production of significant amounts of water. The models suffer however from the current lack of experimental data on the reaction between hydrogen and silicate melt at high pressures and temperatures. Here we present novel experimental results designed to investigate this interaction. Laser heating diamond anvil cell experiments were conducted between 16 and 60 GPa at temperatures above 4000 K. We find that copious amounts of hydrogen dissolve into the silicate melt with a large dependence on temperature rather than pressure. We also find that the reduction of iron oxide leads to the production of significant amounts of water along with the formation of iron-enriched blebs. Altogether, the results predict that the typical processes attending planet formation will result in significant water production with repercussions for the chemistry and structure of the planetary interior as well as the atmosphere.

astro-ph.EP↗
arXiv2025

Robust High-Resolution Multi-Organ Diffusion MRI Using Synthetic-Data-Tuned Prompt Learning

Clinical adoption of multi-shot diffusion-weighted magnetic resonance imaging (multi-shot DWI) for body-wide tumor diagnostics is limited by severe motion-induced phase artifacts from respiration, peristalsis, and so on, compounded by multi-organ, multi-slice, multi-direction and multi-b-value complexities. Here, we introduce a reconstruction framework, LoSP-Prompt, that overcomes these challenges through physics-informed modeling and synthetic-data-driven prompt learning. We model inter-shot phase variations as a high-order Locally Smooth Phase (LoSP), integrated into a low-rank Hankel matrix reconstruction. Crucially, the algorithm's rank parameter is automatically set via prompt learning trained exclusively on synthetic abdominal DWI data emulating physiological motion. Validated across 10,000+ clinical images (43 subjects, 4 scanner models, 5 centers), LoSP-Prompt: (1) Achieved twice the spatial resolution of clinical single-shot DWI, enhancing liver lesion conspicuity; (2) Generalized to seven diverse anatomical regions (liver, kidney, sacroiliac, pelvis, knee, spinal cord, brain) with a single model; (3) Outperformed state-of-the-art methods in image quality, artifact suppression, and noise reduction (11 radiologists' evaluations on a 5-point scale, $p<0.05$), achieving 4-5 points (excellent) on kidney DWI, 4 points (good to excellent) on liver, sacroiliac and spinal cord DWI, and 3-4 points (good) on knee and tumor brain. The approach eliminates navigator signals and realistic data supervision, providing an interpretable, robust solution for high-resolution multi-organ multi-shot DWI. Its scanner-agnostic performance signifies transformative potential for precision oncology.

cs.CV↗
arXiv2025

Electric-Field Control of Terahertz Response via Spin-Corner-Layer Coupling in Altermagnetic Bilayers

Electric field control of electron charge and spin degrees of freedom is fundamental to modern semiconductor and spintronic devices. Yet controlling electromagnetic waves with an electric field, particularly in the terahertz (THz) band, remains a challenge. Here, we propose a spin-corner-layer coupling (SCLC) mechanism in second-order topological altermagnetic bilayers. By using an electric field to influence electrons between different layers, the SCLC mechanism enables simultaneous control over corner and spin degrees of freedom, thereby allowing electric-field tuning of the absorption, emission intensity, and even polarization of THz waves. Taking bilayer NiZrI$_6$ nanodisks as a prototype, we demonstrate that an ultralow electrostatic field can switch both the spin and the layer polarizations of corner states. This dual switching modulates transition dipole moments and oscillator strengths between different corner states, thereby enabling the manipulation of THz waves. This study establishes a mechanism for the electric-field control of spin and THz waves through SCLC, yielding important implications for the advancement of THz spintronics.

cond-mat.mtrl-sci↗
arXiv2025

Unconventional tunnel magnetoresistance scaling with altermagnets

In conventional magnetic tunnel junctions (MTJs), the tunnel magnetoresistance (TMR) typically increases with barrier thickness as electron transmission in the antiparallel configuration decays faster than that of the parallel configuration. In this work, we reveal an anomalous scaling effect in altermagnetic tunnel junctions (AMTJs), where the TMR decreases anomalously with an increasing barrier thickness. The anomalous scaling originates from the overlapping spin-split branches forming a transmission path that cannot be suppressed in the antiparallel state. Such phenomenon is explained by a double-barrier model and is further demonstrated using ab initio quantum transport simulations in 2D V2Te2O/Cr2Se2O/V2Te2O and V2Te2O/ZnSe/V2Te2O AMTJs. Our work identifies a peculiar unexpected transport characteristic of AMTJ, providing a fundamental limit on AMTJ device design and illustrating the potential optimal design of AMTJ at the ultrascaled monolayer limit.

cond-mat.mes-hall↗