BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Zheng Huang

Zheng Huang appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
math.DGhep-phmath.GTmath.APArtificial IntelligenceMachine Learningnucl-thComputer VisionInformation Retrievalmath.CVComputation and Languagemath.FAmath.OAmath.RTphysics.plasm-phSound

Catalog footprint

What is connected

28works
16topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

28 published item(s)

preprint2026arXiv

A Deterministic Agentic Workflow for HS Tariff Classification: Multi-Dimensional Rule Reasoning with Interpretable Decisions

Harmonized System (HS) tariff classification is a high-stakes, expert-level task in which a free-form product description must be mapped to a specific six- or eight-digit code under the General Interpretive Rules (GIR), section notes, chapter notes, and Explanatory Notes. The difficulty lies not in knowledge volume but in *multi-dimensional rule reasoning*: a correct classification must satisfy competing priority rules along several axes simultaneously, including material, form, function, essential character, the part-versus-whole boundary, and specific listing versus residual headings. End-to-end prompting of large language models fails characteristically by resolving one axis while ignoring the priority constraints on the others. We present a *deterministic agentic workflow* in contrast to self-planning agents: the control flow is fixed, language model calls are confined to narrow stages, and reflection and verification are retained as local mechanisms. This design yields interpretability by construction--each decision is decomposed into stage-wise structured outputs with verbatim citation of the chapter or section notes that bear on it. The architecture combines offline knowledge-engineering of the Chinese HS tariff with an online six-stage pipeline. Evaluated on HSCodeComp at the six-digit level, the workflow reaches 75.0% top-1 and 91.5% top-3 at four digits, and 64.2% top-1 and 78.3% top-3 at six digits with Qwen3.6-plus; an open-weight Qwen3.6-27B-FP8 backbone in non-thinking mode achieves 84.2% four-digit and 77.4% six-digit top-1 agreement with the frontier model. A two-stage manual audit of 226 six-digit disagreements suggests that a non-trivial fraction of HSCodeComp ground-truth labels may deviate from HS general rules; full adjudication records are released in the appendix as preliminary findings for community review.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

OmniVL-Guard Pro: A Tool-Augmented Agent for Omnibus Vision-Language Forensics

Existing vision-language forgery detection and grounding methods operate under a closed-world paradigm, assuming verification can be completed by the model alone. However, self-contained MLLMs are constrained by finite parametric knowledge, static training corpora, and limited perceptual resolution, creating a practical ceiling in dynamic open-world forensics -- particularly for real-time event verification requiring external clues and forgery segmentation demanding fine-grained scrutiny of local manipulations. To address these limitations, we shift from scaling up the self-contained model toward reaching beyond it. We propose \textbf{OmniVL-Guard Pro}, a tool-augmented agent that extends unified forensics from closed-world prediction to open-world clues-driven reasoning. OmniVL-Guard Pro integrates a tool environment spanning real-time event search, local cropping and zooming, edge-anomaly screening, face detection, video frame extraction, and SAM3-based segmentation. To generate high-quality tool-reasoning trajectories, we introduce \textbf{Tree-Structured Self-Evolving Tool Trajectory Generation}, which produces diverse trajectories through seed guidance, guider-free self-evolution, and weakly-hinted hard sample synthesis, yielding the Full-Spectrum Tool Reasoning (FSTR) dataset for training. We further propose \textbf{Checker-Guided Agentic Reinforcement Learning} (CGARL), which provides process-level supervision to penalize cases where the answer is correct but the reasoning is distorted. Extensive experiments demonstrate that OmniVL-Guard Pro achieves state-of-the-art performance across various tasks, and exhibits strong zero-shot generalization. The FSTR dataset and code for OmniVL-Guard Pro will be publicly released at \url{https://github.com/shen8424/OmniVL-Guard-Pro}.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Donaldson Functional in Teichmüller Theory

In this paper we define a Donaldson type functional whose Euler-Lagrange equations are a system of differential equations which corresponds to Hitchin's self-duality equations for a suitable choice of Higgs bundle on closed Riemann surfaces. The main challenge of this functional is its lack of regularity and lack of compactness when defined in its natural domain of definition. Though a standard variational approach cannot directly be applied, we provide the appropriate analytical tools that make Donaldson functional treatable by a variational viewpoint. We prove that this functional admits a unique critical point corresponding to its global minimum. As an immediate consequence, we find that this system of self-duality equations admits a unique solution. Among the applications in geometry of this fact, we obtain a parametrization of closed constant mean curvature immersions in hyperbolic manifolds (possibly incomplete), and their moduli spaces.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Empowering Next POI Recommendation with Multi-Relational Modeling

With the wide adoption of mobile devices and web applications, location-based social networks (LBSNs) offer large-scale individual-level location-related activities and experiences. Next point-of-interest (POI) recommendation is one of the most important tasks in LBSNs, aiming to make personalized recommendations of next suitable locations to users by discovering preferences from users' historical activities. Noticeably, LBSNs have offered unparalleled access to abundant heterogeneous relational information about users and POIs (including user-user social relations, such as families or colleagues; and user-POI visiting relations). Such relational information holds great potential to facilitate the next POI recommendation. However, most existing methods either focus on merely the user-POI visits, or handle different relations based on over-simplified assumptions while neglecting relational heterogeneities. To fill these critical voids, we propose a novel framework, MEMO, which effectively utilizes the heterogeneous relations with a multi-network representation learning module, and explicitly incorporates the inter-temporal user-POI mutual influence with the coupled recurrent neural networks. Extensive experiments on real-world LBSN data validate the superiority of our framework over the state-of-the-art next POI recommendation methods.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

GFTE: Graph-based Financial Table Extraction

Tabular data is a crucial form of information expression, which can organize data in a standard structure for easy information retrieval and comparison. However, in financial industry and many other fields tables are often disclosed in unstructured digital files, e.g. Portable Document Format (PDF) and images, which are difficult to be extracted directly. In this paper, to facilitate deep learning based table extraction from unstructured digital files, we publish a standard Chinese dataset named FinTab, which contains more than 1,600 financial tables of diverse kinds and their corresponding structure representation in JSON. In addition, we propose a novel graph-based convolutional neural network model named GFTE as a baseline for future comparison. GFTE integrates image feature, position feature and textual feature together for precise edge prediction and reaches overall good results.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Varied fusion reaction probability induced by ion stopping modification in laser-driven plasma with different temperature

The dynamics of nuclear reaction in plasma is a fundamental issue in many high energy density researches, such as the astrophysical reactions and the inertial confinement fusion. The effective reaction cross-sections and ion stopping power in plasma need to be taken into account to analyze the reactivity. In this research, we have experimentally investigated the from D-D reactions from interactions between deuteron beams and deuterated polystyrene (CD) plasma, driven by two laser pulses respectively. The neutron yields, plasma density and deuteron energy loss in plasma have been measured, and the plasma temperature and deuteron stopping power have been analyzed from simulations. It is shown that, compared with a cold target, the reaction probability in plasma conditions can be enhanced or suppressed, which is ascribed to the deuteron stopping power modifications in plasma. In hotter CD plasma, the energy loss of moderate energetic deuterons reduces, which leads to higher D-D reaction probability, while the contrary happens in colder plasma. This work provides new understanding of fusion reactions in plasma environment.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Audio Recording Device Identification Based on Deep Learning

In this paper we present a research on identification of audio recording devices from background noise, thus providing a method for forensics. The audio signal is the sum of speech signal and noise signal. Usually, people pay more attention to speech signal, because it carries the information to deliver. So a great amount of researches have been dedicated to getting higher Signal-Noise-Ratio (SNR). There are many speech enhancement algorithms to improve the quality of the speech, which can be seen as reducing the noise. However, noises can be regarded as the intrinsic fingerprint traces of an audio recording device. These digital traces can be characterized and identified by new machine learning techniques. Therefore, in our research, we use the noise as the intrinsic features. As for the identification, multiple classifiers of deep learning methods are used and compared. The identification result shows that the method of getting feature vector from the noise of each device and identifying them with deep learning techniques is viable, and well-preformed.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Bi-directional LSTM Recurrent Neural Network for Chinese Word Segmentation

Recurrent neural network(RNN) has been broadly applied to natural language processing(NLP) problems. This kind of neural network is designed for modeling sequential data and has been testified to be quite efficient in sequential tagging tasks. In this paper, we propose to use bi-directional RNN with long short-term memory(LSTM) units for Chinese word segmentation, which is a crucial preprocess task for modeling Chinese sentences and articles. Classical methods focus on designing and combining hand-craft features from context, whereas bi-directional LSTM network(BLSTM) does not need any prior knowledge or pre-designing, and it is expert in keeping the contextual information in both directions. Experiment result shows that our approach gets state-of-the-art performance in word segmentation on both traditional Chinese datasets and simplified Chinese datasets.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Closed Minimal Surfaces in Cusped Hyperbolic Three-manifolds

Motivated by classical theorems on minimal surface theory in compact hyperbolic three-manifolds, we investigate the questions of existence and deformations for least area minimal surfaces in complete noncompact hyperbolic three-manifold of finite volume. We prove any closed immersed incompressible surface can be deformed to a closed immersed least area surface within its homotopy class in any cusped hyperbolic three-manifold. Our techniques highlight how special structures of these cusped hyperbolic three-manifolds prevent any least area minimal surface going too deep into the cusped region.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Counting Minimal Surfaces in Quasi-Fuchsian three-Manifolds

It is well known that every quasi-Fuchsian manifold admits at least one closed incompressible minimal surface, and at most finitely many of them. In this paper, for any prescribed integer $N>0$, we construct a quasi-Fuchsian manifold which contains at least $2^N$ such minimal surfaces. As a consequence, there exists some simple close Jordan curve on $S^2_\infty$ such that there are at least $2^N$ disk-type complete minimal surface in $\mathbb{H}^3$ sharing this Jordan curve as the asymptotic boundary.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

On Almost-Fuchsian Manifolds

Almost-Fuchsian manifold is a class of complete hyperbolic three manifolds. Such a three-manifold is a quasi-Fuchsian manifold which contains a closed incompressible minimal surface with principal curvatures everywhere in the range of (-1, 1). In such a manifold, the minimal surface is unique and embedded, hence one can parametrize these hyperbolic three-manifolds by their minimal surfaces. In this paper we obtain estimates on several geometric and analytical quantities of an almost-Fuchsian manifold M in terms of the data on the minimal surface. In particular, we obtain an upper bound for the hyperbolic volume of the convex core of M, and an upper bound on the Hausdor? dimension of the limit set associated to M. We also constructed a quasi-Fuchsian manifold which admits more than one minimal surface, and it does not admit a foliation of closed surfaces of constant mean curvature.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Holomorphic Cubic Differentials and Minimal Lagrangian Surfaces in CH2

Following earlier work of Loftin-McIntosh, we study minimal Lagrangian immersions of the universal cover of a closed surface (of genus at least 2) into CH2, with prescribed data of a conformal structure plus a holomorphic cubic differential. We show existence and non-uniqueness of such minimal Lagrangian immersions. We also establish the surface area with respect to the induced metric as a Weil-Petersson potential function for the space of holomorphic cubic differentials on the Riemann surface.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Curvatures on the Teichmüller curve

The Teichmüller curve is the fiber space over Teichmüller space of closed Riemann surfaces, where the fiber over a point in Teichmüller space is the underlying surface. We derive formulas for sectional curvatures on the Teichmüller curve. In particular, our method can be applied to investigate the geometry of the Weil-Petersson geodesic as a three-manifold, and the degeneration of the curvatures near the infinity of the augmented Teichmüller space along a Weil-Petersson geodesic, as well as the minimality of hyperbolic surfaces in this three-manifold.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Mean Curvature Flows in Almost Fuchsian Manifolds

An almost Fuchsian manifold is a quasi-Fuchsian hyperbolic three-manifold that contains a closed incompressible minimal surface with principal curvatures everywhere in the range of (-1,1). In such a hyperbolic three-manifold, the minimal surface is unique and embedded, hence one can parametrize these three-manifolds by their minimal surfaces. We prove that any closed surface which is a graph over any fixed surface of small principal curvatures can be deformed into the minimal surface via the mean curvature flow. We also obtain an upper bound for the hyperbolic volume of the convex core of M, as well as estimates of the Hausdorff dimension of the limit set for $M$.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Minimal immersions of closed surfaces in hyperbolic three-manifolds

We study minimal immersions of closed surfaces (of genus $g \ge 2$) in hyperbolic 3-manifolds, with prescribed data $(σ, tα)$, where $σ$ is a conformal structure on a topological surface $S$, and $αdz^2$ is a holomorphic quadratic differential on the surface $(S,σ)$. We show that, for each $t \in (0,τ_0)$ for some $τ_0 > 0$, depending only on $(σ, α)$, there are at least two minimal immersions of closed surface of prescribed second fundamental form $Re(tα)$ in the conformal structure $σ$. Moreover, for $t$ sufficiently large, there exists no such minimal immersion. Asymptotically, as $t \to 0$, the principal curvatures of one minimal immersion tend to zero, while the intrinsic curvatures of the other blow up in magnitude.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Quasi-Fuchsian 3-Manifolds and Metrics on Teichmüller Space

An almost Fuchsian 3-manifold is a quasi-Fuchsian manifold which contains an incompressible closed minimal surface with principal curvatures in the range of $(-1,1)$. Such a 3-manifold $M$ admits a foliation of parallel surfaces, whose locus in Teichmüller space is represented as a path $γ$, we show that $γ$ joins the conformal structures of the two components of the conformal boundary of $M$. Moreover, we obtain an upper bound for the Teichmüller distance between any two points on $γ$, in particular, the Teichmüller distance between the two components of the conformal boundary of $M$, in terms of the principal curvatures of the minimal surface in $M$. We also establish a new potential for the Weil-Petersson metric on Teichmüller space.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

Surfaces of Prescribed Mean Curvature in Quasi-Fuchsian Manifolds

Let $M$ be a quasi-Fuchsian three-manifold that contains a closed incompressible surface with principal curvatures within the range of the unit interval, for a prescribed function $H$ (with mild conditions) on $M$, we construct a closed incompressible surface with mean curvature $H$ . A direct application is the existence of embedded surfaces of prescribed constant mean curvatures with constants in $(-2,2)$.

0 citations0 reviewsNo code linkedOpen access
preprint1996arXiv

Dilepton and Photon Productions from a Coherent Pion Oscillation

Since the electromagnetic current for a pion system coincides with the third component of the isovector current, the isospin angular oscillation of a coherent field can be a significant source for the electromagnetic emissions. We study the characteristic dilepton and photon emissions from the classical pion field oscillation in the QCD vacuum. The general analytical solution obtained in the nonlinear sigma model is used to calculate the electromagnetic current density, which exhibits a light-front singularity and decreases rapidly as inverse square of the proper time due to a longitudinal expansion. The momentum and invariant mass spectra of the direct photon and dilepton are found to be a sensitive probe of the space-time evolution of the chiral condensate field.

0 citations0 reviewsNo code linkedOpen access
preprint1995arXiv

Isospin Coherence and Final-State Scattering of Disoriented Chiral Condensate

We examine the validity of the notion of the coherent state for pions and the quantum scattering effect in the final state of pion emission. When the number of particles is large, the effect caused by the small but finite mass difference between the neutral and charged pions can add up substantially in the quantum evolution of an initially coherent state. As a result, the states with quite different numbers of neutral or charged pions are essentially incoherent. The importance of the quantum scattering in the final-state isospin charge distribution of a disoriented chiral condensate (DCC) is investigated. We find that the scattering effect significantly reduces the spectacular Centauro and anti-Centauro events. The deformation of a charge distribution $dP/df$ predicted by the classical field theory is significant only for a DCC with the size of 10 fm or more.

0 citations0 reviewsNo code linkedOpen access
preprint1995arXiv

Reinterpretation of Thermal Dilepton Emission Rate by Spectral Functions

We reinterpret the dilepton emission rate from a hadronic gas expected to be produced in heavy ion collisions in terms of the spectral functions available from the $e^+e^-$ annihilation and the $τ$ lepton decays experiments. We take into account all possible hadronic state especially the multi-pion contributions to the dilepton emission and the parity mixing phenomenon due to the soft final-state corrections. A new compilation of the experimental data for the spectral functions is presented.

0 citations0 reviewsNo code linkedOpen access
preprint1992arXiv

Comments on the Vacuum Orientations in QCD

We study the QCD vacuum orientation angles in correlation with the strong CP phases. A vacuum alignment equation of the dynamical chiral symmetry breaking is derived based on the anomalous Ward identity. It is emphasized that a chiral rotation of the quark field causes a change of the vacuum orientation and a change in the definition of the light pseudoscalar generators. As an illustration of the idea, $\h\rightarrow 2\p$ decays are carefully studied in different chiral frames. Contrary to the claim in Ref.[7], the $θ$-term does not directly contribute to the CP-violating processes.

0 citations0 reviewsNo code linkedOpen access
preprint1992arXiv

On the Anomalous Discrete Symmetry

We examine an interesting scenario to solve the domain wall problem recently suggested by Preskill, Trivedi, Wilczek and Wise. The effective potential is calculated in the presence of the QCD axial anomaly. It is shown that some discrete symmetries such as CP and Z_2 can be anomalous due to a so-called $K$-term induced by instantons. We point out that Z_2 domain-wall problem in the two-doublet standard model can be resolved by two types of solutions: the CP-conserving one and the CP-breaking one. In the first case, there exist two Z_2-related local minima whose energy splitting is provided by the instanton effect. In the second case, there is only one unique vacuum so that the domain walls do not form at all. The consequences of this new source of CP violation are discussed and shown to be well within the experimental limits in weak interactions.

0 citations0 reviewsNo code linkedOpen access
preprint1992arXiv

The Measure of Strong CP Violation

We investigate a controversial issue on the measure of CP violation in strong in teractions. In the presence of nontrivial topological gauge configurations, the $θ$-term in QCD has a profound effect: it breaks the CP symmetry. The CP-violating amplitude is shown to be determined by the vacuum tunneling process, where the semiclassical method makes most sense. We discuss a long-standing dispute on whether the instanton dynamics satisfies or not the anomalous Ward identity (AWI). The strong CP violation measure, when complying with the vacuum alignment, is proportional to the topological susceptibility. We obtain an effective CP-violating lagrangian different from that provided by Baluni. To solve the IR divergence problem of the instanton computation, We present a ``classically gauged'' Georgi-Manohar model and derive an effective potential which uniquely determines an explicit $U(1)_A$ symmetry breaking sector. The CP violation effects are analyzed in this model. It is shown that the strong CP problem and the $U(1)$ problem are closely related. Some possible solutions to both problems are also discussed with new insights.

0 citations0 reviewsNo code linkedOpen access