arXiv Science⌕ Search

arXiv subjects

Hui Li

Publications and source records attributed to Hui Li.

At least 73 records · Page 4Linked to original sources
arXiv2026

FinixDoc: Rethinking Financial Document Parsing Beyond Saturated Benchmarks

Financial document parsing requires accuracy, structural consistency, and verifiability that current benchmarks often fail to reflect. We present FinixDoc, an end-to-end agentic parsing system for real-world financial documents, with FinixDoc-VL, a 4B-scale vision-language model built on Qwen3-VL-4B, as its core parser. To characterize the gap between benchmark and deployment performance, we introduce a Document Parsing Capability Matrix organized along two practical axes: visual quality and document scale. Guided by this matrix, FinixDoc-VL is trained with a domain-adapted recipe combining homoglyph-aware contrastive learning and multi-stage reinforcement learning with composite domain-specific rewards. To better leverage our accumulated advantage in low-quality financial-document data and support large-scale, high-quality data production, we further build a human-in-the-loop Data Factory pipeline with confidence-aware expert review. For evaluation, we construct FinixDocBench, a financial-domain evaluation suite covering digital-native, camera-captured, ultra-large-page, and internal-workflow scenarios, with a compliance-reviewed subset released alongside this technical report. On its main subsets, FinixDoc-VL achieves the highest overall score (81.43) among evaluated baselines, outperforming the next-best open-source model by 5.13 points, with the largest gains on internal financial workflows (FinixInner: 84.08 vs. 78.73).

cs.AI↗
arXiv2026

Giant Surface-driven Nonlinear Hall Effect in BiTeCl at Room Temperature

The nonlinear Hall effect (NLHE) provides a pathway to generate a Hall response in time-reversal-symmetric yet inversion-symmetry-broken systems. NLHE can rectify an alternating current into a transverse direct voltage, making it attractive for radio-frequency rectification, energy harvesting, and terahertz detection, applications for which device miniaturization remains a central pursuit. In this context, the inherent inversion symmetry breaking at surfaces is particularly appealing: because symmetry is necessarily broken at the surface of any crystal, irrespective of whether its bulk is centrosymmetric, surface-driven nonlinear responses lift the stringent constraint on bulk symmetry and open a route toward compact device architectures. Here we report the observation of a giant, surface-driven second-order nonlinear Hall effect in the Rashba-type polar semiconductor BiTeCl at room temperature. The determined second-order nonlinear Hall susceptibility at 300 K reaches 1.68 $μ$mV$^{-1}$, which is 80 times larger than that of the best previously reported surface-dominated systems. We attribute this giant response to the synergistic interplay between BiTeCl's polar crystal structure and its rich surface states: the polar stacking renders the top and bottom surfaces inequivalent, so that the nonlinear response originates from a single surface without compensation from the other. Symmetry and scaling analyses suggest that both skew-scattering and side-jump mechanisms contribute to the observed effect. Our findings not only identify BiTeCl as a promising platform for future applications utilizing the NLHE, but also establish the asymmetry between the opposite surfaces of a polar crystal as a general design principle for discovering surface-driven materials with larger nonlinear Hall responses.

cond-mat.mtrl-sci↗
arXiv2026

Quantum critical behavior in chains of hindered dipolar planar rotors

We study the ground-state properties of linear chains of dipolar planar rotors hindered by a six-fold on-site potential, a model motivated by water molecules confined in the hexagonal cavities of beryl. Using density matrix renormalization group (DMRG) calculations, we locate the quantum phase transition between the disordered and ferroelectrically ordered phases using the von Neumann entanglement entropy and the Binder ratio of the polarization. A sweep of the six-fold pinning strength shows that increasing hindrance shifts the critical dipolar coupling gc to smaller values. These results suggest that crystal-field hindrance can promote, rather than suppress, dipolar ordering. This work has implications for ferroelectricity and quantum-device tuning in confined molecular rotors

cond-mat.mes-hall↗
arXiv2026

Remarkable Enhancement of High Harmonic Generation from Superhard Material under High Pressure

High harmonic generation (HHG) in solids offers a pathway to develop compact extreme ultraviolet (EUV) sources crucial for attosecond science and advanced spectroscopy. Here, we demonstrate theoretically that high pressure dramatically enhances HHG in superhard hexagonal tungsten nitride (h-WN6). Compared with solid-state systems at ambient pressure, the reshaped electronic environment under high pressure leads to a unique band-gap widening in h-WN6, which raises the material's damage threshold, allowing the use of stronger laser fields and enabling access to higher-energy bands. This pressure-induced band-gap widening offers a promising strategy to overcome the cutoff limitation of solid-state EUV light sources.

physics.optics↗
arXiv2026

The THESAN-ZOOM project: clumpiness of high-redshift galaxies and its connection to bursty star formation

Recent JWST observations have revealed diverse high-redshift galaxy morphologies, including a population with irregular and clumpy structures. The physical origin of these structures, and the extent to which observational biases shape their appearance, remain uncertain. We present a power-spectrum-based method for quantifying galaxy clumpiness across spatial scales, using the radiation-hydrodynamic simulation suite THESAN-ZOOM, which employs a state-of-the-art galaxy formation model that resolves the multiphase interstellar medium (ISM). Although the total stellar mass distributions in THESAN-ZOOM galaxies are usually smooth, clumpy structures appear in the H$α$, far-ultraviolet (FUV), and optical light distributions. Tracers sensitive to shorter-timescale star formation exhibit more pronounced small-scale structure ($\sim10^{2}$--$10^{3}{\rm pc}$). The corresponding projected light spectra follow $P(k)\propto k^{-1}$ to $k^{-2}$, with progressively shallower slopes for tracers sensitive to more recent star formation, reflecting enhanced small-scale power and greater spatial intermittency in young stellar populations. This behaviour is consistent with a highly compressible, shock-dominated ISM in which stellar feedback and outflows reorganise dense gas into filamentary and clumpy structures. We also find that galaxy clumpiness depends on the treatment of stellar feedback. Weaker early stellar feedback enhances small-scale power in both the mass and light distributions. Clumpiness also varies strongly over the bursty star formation cycle, implying that observed samples may be biased towards galaxies caught in phases of elevated star formation. Galaxy clumpiness, therefore, could provide a complementary probe of the bursty star formation in the early Universe.

astro-ph.GA↗
arXiv2026

Benchmarking Automated Security Patch Backporting: How Far Are We?

Automated security patch backporting is critical for mitigating N-day vulnerabilities. Recent tools report success rates above 80% on their respective datasets. However, these evaluations are often confined to homogeneous environments, such as one repository or specific project versions. Consequently, it remains unclear how well these tools generalize beyond their originally targeted scenarios. We present Porting Benchmark, a curated dataset of 1,234 security patch backporting cases spanning cross-version, cross-branch, and cross-repository scenarios, paired with a common evaluation framework. Using this benchmark, we evaluate five tools spanning program analysis, LLM prompting, and LLM agents under aligned settings. Our results show that aligned evaluation changes the apparent performance landscape: PortGPT and TSBPort remain comparatively strong on the Replication Dataset, while FixMorph and Mystique degrade substantially under the common protocol. Performance degrades sharply on structurally complex patches: the best commit-level success rate falls from 85.2% on Type-I patches to 24.0% on Type-IV. We identify four root-cause categories (missing target API awareness, cross-version semantic mismatch, non-local dependency propagation failure, and patch construction or localization failure) and derive concrete directions for next-generation tool design. On a 45-case dynamically validated subset with verified test cases and constructed POCs, we further observe that reference-based benchmark scores do not fully capture real-world remediation: exact match sharply under-credits harder target adaptations, while executable validation reveals residual integration failures in the target that static reference agreement misses. Executable-feedback refinement provides limited but measurable recovery on the hardest executable cases.

cs.SE↗
arXiv2026

Breaking the mutual exclusivity between metallicity and ferroelectricity in a non-polar covalent semiconductor via orbital selective doping

The mutual exclusion of ferroelectricity and metallic conductivity is a long-standing tenet because itinerant electrons screen long-range Coulomb forces that stabilize the bulk polar order. Here, we break this paradigm by heavily doping a non-polar covalent semiconductor of cubic silicon carbide (3C-SiC) with nitrogen. This introduces heavy electron doping, inducing metallicity and driving a structural transition from the non-polar F-43m to the polar R3m symmetry via the pseudo-Jahn-Teller effect. Remarkably, we provide direct, atomic-scale visualization of about 180° polarization reversal under an external voltage bias in a ferroelectric metal. The strongly directional character of antibonding orbitals occupied by conduction electrons prevents them from screening the local Si-C polarization, resulting in the coexistence of metallicity and ferroelectricity. Ferroelectric tunnel junctions demonstrate nonvolatile memory properties with a well-defined high-resistance state (HRS) and low-resistance state (LRS), an ultrahigh response speed (~50 ns), an ultralow operating voltage (1 V), an endurance exceeding 85927 cycles, and a projected retention time of 100 years. Our results provide a novel strategy for pioneering ferroelectricity in a metal, a new ferroelectric metal platform for exploring exotic properties, and a ferroelectric device with high performance that meets the requirements for low consumption and high-speed non-volatile devices.

cond-mat.mtrl-sci↗
arXiv2026

Efficient Privacy-Preserving Range Filtered Approximate Nearest Neighbor Search

Range-filtered approximate nearest neighbor search (RFANNS) is an important primitive for vector databases; it retrieves vectors that are similar to a query and satisfy a numerical range predicate, but existing RFANNS indexes expose vectors, attributes, and queries in plaintext. This assumption is unsuitable for outsourced vector databases, where sensitive data and queries must be protected from an honest-but-curious cloud server. To the best of our knowledge, this is the first study that systematically formulates and evaluates privacy-preserving RFANNS over outsourced encrypted vector databases. Our approach separates range localization from encrypted vector search: an authorized user maps the query range to a compact set of nodes in a local N-ary attribute tree, and the server searches only the corresponding proximity graph sub-indices over encrypted vectors. To reduce expensive encrypted comparisons, we use a filter-and-refine pipeline that first retrieves coarse candidates with approximate distance-comparison-preserving encryption and then reranks a small candidate set with exact distance-comparison encryption. We then analyze the computation, storage, communication, and leakage of the protocol. Experiments on four widely used vector datasets show that our method improves the QPS-Recall trade-off over representative secure adaptations of existing RFANNS approaches, scaling effectively to large datasets.

cs.DB↗
arXiv2026

FROG: Efficient Range-Filtering Approximate Nearest Neighbor Search on GPUs

Range-filtering approximate nearest neighbor search (RFANNS) is a fundamental operation in modern vector databases. Given a query vector $q$ and a numerical range predicate, RFANNS returns the $k$-approximate nearest neighbors ($k$-ANN) of the query $q$ among the objects whose attributes satisfy the range predicate. However, existing RFANNS methods are not well suited to high-throughput GPU execution. CPU indexes offer limited parallel scalability, generic GPU filtering is highly selectivity-dependent, and GPU indexes built from locally optimized subgraphs can incur long search trajectories and redundant distance computations. To address these limitations, we present FROG, a GPU-oriented RFANNS index that replaces multiple locally optimal substructure building with a globally aware, vertex-centric design. It organizes diverse expansion neighbor candidates for each vertex in a GPU-friendly structure and rapidly identifies the expansion neighbors used for computation at query time. Moreover, GPU-oriented algorithms and implementations are developed for both index construction and query processing. Experiments on six datasets show that FROG improves mixed-selectivity query throughput by 14.7--37.7$\times$ over 44-core CPU baselines and 4.5--7.6$\times$ over the strongest GPU baseline. It also accelerates index construction by 2.4--14.8$\times$ over the GPU baseline.

cs.DB↗
arXiv2026

Flexibility of the isometric immersion system in arbitrary dimension and codimension and the energy scaling of prestrained thin films

We prove that, for a given $C^{r,β}$-regular Riemann metric posed on a $d$-dimensional domain, every short immersion into the Euclidean space $\mathbb{R}^{d+k}$, can be uniformly approximated by exact isometric immersions of regularity $C^{1,α}$ for any $α<α_0=\min\{\frac{r+β}{2}, \frac{1}{1+d(d+1)/k}\}$. Our theorem recovers several previously known results as special cases. The novelty thereof lies in providing a unified flexibility statement for arbitrary dimensions $d$ and codimensions $k$, while also treating the so far uncharted range $k\in (1, \frac{d(d+1)}{2}-d+1)\setminus \{d\}$, where no corresponding general result was previously available. Our threshold flexibility exponent $α_0$ agrees with that previously obtained for the closely related Monge-Ampère system. As an application, we prove a new estimate in the quantitative immersability of thin prestrained films, setting the scaling exponent of the infimum of non-Euclidean energies in presence of an arbitrary prestrain metric, and in the limit of the film's vanishing thickness, at $\frac{4α_0}{α_0+1}$.

math.AP↗
arXiv2026

Amplitude analysis for charmed meson decays at BESIII

Amplitude analysis bridges the gap between experimental measurements of multibody charmed-meson decays and theoretical predictions of intermediate two-body processes. This work presents a comprehensive overview of the amplitude-analysis methodology employed by the BESIII Collaboration, emphasizing practical implementation. We detail the construction of the probability density function and likelihood function for unbinned maximum-likelihood fits. This encompasses Monte Carlo integration techniques for normalization, the incorporation of detection efficiency and resolution effects, and multidimensional background modeling utilizing XGBoost classifiers. Furthermore, we describe the amplitude formalism for both hadronic and semileptonic decays, incorporating standard resonance-propagator parametrizations. Key analytical aspects, including the evaluation of fit fractions, the generation of kinematic projections, and the estimation of statistical uncertainties, are also discussed.

hep-ex↗
arXiv2026

Plasmoid-Mediated 2D Magnetic Reconnection in Partially Ionized Plasmas

Magnetic reconnection in partially ionized plasmas is an important channel for energy release. While the plasmoid instability is well characterized in 2D fully ionized plasmas, its behavior in the presence of neutral-dominated plasma remains poorly understood in the nonlinear, high-Lundquist-number ($S = 10^5$) regime. We present high-resolution ($16384 \times 4096$ cells) two-dimensional two-fluid (ion $+$ neutral) simulations of Harris-sheet reconnection with upstream plasma beta $β= 2$, comparing fully ionized and partially ionized (ionization fraction $ξ= 10^{-1}$ and $10^{-2}$) regimes. Neutral-ion decoupling accelerates the linear tearing stage and alters the plasmoid hierarchy: the large-scale ``monster'' plasmoid that dominates the fully ionized case is suppressed, and the sheet instead fragments into a dense chain of sub-scale plasmoids. Below the neutral-ion decoupling scale $\ell_{\rm dec}$, ions concentrate into the plasmoids, reaching peak overdensities $ρ_i/ρ_{i,0} \approx 10$ ($ξ= 10^{-1}$) and $3-5\times10^{3}$ ($ξ= 10^{-2}$), while the neutrals remain comparatively smooth. This local pile-up raises the ionization fraction and recouples the two fluids within the plasmoids. Measured from the out-of-plane electric field at the reconnection sites, the reconnection rate in the $ξ= 10^{-2}$ case achieves $R_{\rm rec}\approx0.01$, whereas the $ξ= 10^{-1}$ case rises to a rate $\approx0.02$ and further $0.035$ when apparent coalescence occurs. In the $ξ= 10^{-2}$ case, the ambipolar drift drives a rapid ion inflow $\sim0.5\,v_{A,0}$ into the layer at the same reconnection sites, far above the neutral inflow velocity $\sim0.1\,v_{A,0}$. Here, $v_{A,0}$ is the upstream total Alfvén speed.

astro-ph.HE↗
arXiv2026

Physics Consistency and Latent Dynamics in Spatiotemporal Physics Field Generation

Data-driven models for spatiotemporal physical field generation, such as flow and acoustic fields, often deviate from governing equations and lack interpretability in latent temporal dynamics. To address these challenges, we propose HMT-PF, a hybrid Mamba-Transformer architecture for physical field generation. The framework incorporates a query-based gradient computation mechanism and a physics-informed fine-tuning strategy to enhance physical consistency. Analysis of the latent space reveals that the initial latent state vector evolves as an autonomous dynamical system under the Mamba backbone. Principal component analysis (PCA) indicates that a small number of dominant modes in initial latent state vector govern the key evolution patterns of the physical field, while a Jacobian-based temporal sensitivity analysis characterizes the intrinsic dynamical structure and stability of the latent evolution. Experiments across five benchmark datasets demonstrate strong performance, and physics-informed fine-tuning further reduces physical residuals, highlighting the effectiveness of the proposed latent-level fusion strategy. An empirical scaling law between prediction error and physical residual is identified, revealing a consistent exponential relationship in the low-error regime. Based on this observation, an dual-metric framework is proposed to jointly evaluate numerical accuracy and physical realism.

cs.LG↗
arXiv2026

Homojunction-induced thermopower enhancement in polymer films

It has been more than twenty years since conductive polymers began to receive attention as an emerging thermoelectric material. However, the trade-off between electrical conductivity (σ) and thermopower (S) has proven to be a major challenge that has obstructed their use in actual devices. Here we report the discovery that the thermopower of the p- and n-type legs of organic thermogenerators can be substantially enhanced, without significant deterioration of σ, by constructing an in-plane segmented structure consisting of a homojunction with different doping levels on either side. In such segmented layers, the S is abnormally higher than the average value of the constituent parts when applying a forward temperature gradient (heating the heavily doped counterpart), while it is lower upon a reverse temperature gradient. Typically, for a two-stage segmented film of p-type PDPP-Se, an abnormally large S of 210 uV K-1 and σ of 2.5*10^4 S m-1 are obtained, resulting in a large power factor (PF) of 1100 uW m-1 K-2 and a record ZT of 1.36 at room temperature. The enhanced thermopower is attributed to an additional voltage developed at the homojunction under heating as explained by kinetic Monte Carlo simulations. This finding provides a breakthrough approach to the modulation of thermoelectric transport properties of conductive polymers.

cond-mat.mtrl-sci↗
arXiv2026

Knowledge-Geometry Decoupling: Refreshable Pretrained Transfer for Streaming Recommendation

Industrial recommenders increasingly adopt the pretrain-then-transfer paradigm, yet behavioral distribution drift raises two questions: what to learn from behavior sequences, and how to transfer the learned knowledge while the pretrained model is continually refreshed. To resolve them, we propose Knowledge-Geometry Decoupling (KGD). For what to learn, conventional next-token prediction treats adjacency as dependency and may encode spurious transitions across unrelated sessions. We introduce Behavioral Multi-Token Prediction (BMTP) to retain only collaboratively or semantically related future items as supervision, yielding cleaner and more transferable behavioral knowledge. For how to transfer, pretrained knowledge and task-specific geometry impose conflicting optimization demands on shared parameters. To handle it, KGD assigns them to separate parameter sets: a refreshable encoder owns behavioral knowledge, while a task learner reads contextualized encoder states through read-only cross-attention and writes task-specific geometry through Anchored Calibration Residual (ACR) orthogonal to the pretrained embedding. The decoupled ownership enables continual knowledge refresh without task-gradient interference or invalidating downstream adaptation. KGD improves over strong pretrain-transfer baselines by 4-12% on eight public benchmarks and sustains its advantage over a 90-day production stream where baselines show no gains. KGD has been fully deployed in Shopee. In a live A/B test on Shopee Homepage Search, it increases GMV per user by 1.75% and advertising revenue by 1.53%, demonstrating its high practical value. We provide the core implementation of KGD at https://github.com/FuCongResearchSquad/KGD4REC.

cs.IR↗
arXiv2026

PackingGPT: 3D Packing Agent for Real Furniture in Last-Mile Delivery

3D bin packing rectangular items into standardised containers to maximise space utilisation under geometric shipping automation. Loading a furniture purchase into a personal vehicle is the same task, but under more complex conditions that standard container loading algorithms ignore. This paper addresses the physically stable placement under these realistic conditions with heterogeneous boxes (e.g. varying dimensions and weights) and occupied containers (e.g. groceries). This paper provides a real-world benchmark dataset and baseline model for the Heterogeneous furniture-in-vehicle packing task. The dataset uses real furniture company flat-pack packaging data covering a large number of catalogue products via family-level extrapolation with diversity length, widths, heights, and weights. We also propose a PackingGPT framework for packing as a sequential placement inspired by the Lego assembly process, where heterogeneous boxes of varying dimensions (bricks) are placed step-by-step into the irregular remaining cargo space (creations). Five baseline packing methods were tested on our dataset without considering the Centre-of- Mass (CoM) constraints. In sedan car simulations, 10-40% of placed boxes failed the stability check on average. When the LLP model was trained on packing sequences with CoM constraints enforced during placement, the failure rate dropped to 0.67% (SUV-500).

cs.CV↗
arXiv2026

Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers for Text-to-Image Generation

Multimodal Diffusion Transformers (MMDiTs) for text-to-image generation maintain separate text and image branches, with bidirectional information flow between text tokens and visual latents throughout denoising. In this setting, we observe a prompt forgetting phenomenon: the semantics of the prompt representation in the text branch is progressively forgotten as depth increases. We further verify this effect on three representative MMDiTs--SD3, SD3.5, and FLUX.1 by probing linguistic attributes of the representations over the layers in the text branch. Motivated by these findings, we introduce a training-free approach, prompt reinjection, which reinjects prompt representations from early layers into later layers to alleviate this forgetting. Experiments on GenEval, DPG, and T2I-CompBench++ show consistent gains in instruction-following capability, along with improvements on metrics capturing preference, aesthetics, and overall text--image generation quality.

cs.CV↗
arXiv2026

4DHumanDiff: Direct Text-to-4DGS Generation for Consistent 360-Degree Dynamic Humans

Generating high-quality 360-degree dynamic human assets from text prompts is challenging. Existing methods usually synthesize monocular or multi-view videos first and then fit a 4D representation, which is expensive and often causes incomplete geometry or view-inconsistent renderings. We present 4DHumanDiff, a diffusion framework that directly generates dynamic humans represented by 4D Gaussian Splatting (4DGS) from text prompts. By modeling the structured 4D representation space end-to-end, 4DHumanDiff avoids video pre-generation and per-scene reconstruction, making it better suited for view-consistent and temporally coherent asset generation. The model uses a 3D U-Net backbone with temporal attention for motion-aware generation. We further construct a large-scale text-to-4DGS dataset with 60,000 high-quality pairs, and introduce 2D regularization and training-free 4D interpolation to improve rendering quality and motion smoothness. Experiments show that 4DHumanDiff generates consistent 360-degree dynamic humans within one minute, achieves better temporal and multi-view consistency, and reduces inference time by more than 10x.

cs.CV↗