Source author record

Lei Wu

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

ResearcherUnclaimed source record

Catalog footprint

What is connected

117works

37topics

4close collaborators

Actions

Connect this record

Open graph Browse works

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.

preprint2026arXiv

Accelerated simulation of multiscale gas-radiation coupling flows via a general synthetic iterative scheme

Gas-radiation coupling critically influences hypersonic reentry flows, where extreme temperatures induce pronounced non-equilibrium gas and radiative heat transport. Accurate and efficient simulation of radiative gas dynamics is therefore indispensable for reliable design of thermal protection systems for atmospheric entry vehicles. In this study, a Boltzmann-type kinetic model for radiative gas flows is solved across a broad spectrum of flow and radiation transport regimes using the general synthetic iterative scheme (GSIS). The approach integrates an unstructured finite-volume discrete velocity method with a set of macroscopic synthetic equations. Within this framework, the kinetic model provides high-order closures for the constitutive relations in the synthetic equations. Simultaneously, the macroscopic synthetic equations drive the evolution of the mesoscopic kinetic system, significantly accelerating steady-state convergence in near-continuum regimes, as substantiated by linear Fourier stability analysis. Crucially, the algorithm is proven to be asymptotic-preserving, correctly recovering the continuum and optically thick limits, represented by the radiative Navier-Stokes-Fourier equations governing distinct translational, rotational, vibrational, and radiative temperatures, on coarse meshes independent of the mean free path. Numerical simulations of challenging benchmarks, including three-dimensional hypersonic flow over an Apollo reentry capsule, demonstrate that GSIS achieves orders-of-magnitude speedup over conventional iterative schemes in multiscale simulations of radiative gas flows while accurately capturing non-equilibrium effects and radiative heat transfer in hypersonic environments.

preprint2026arXiv

Dual-Diffusional Generative Fashion Recommendation

Personalized generative recommender systems have emerged as a promising solution for fashion recommendation. However, existing methods primarily rely on implicit visual embeddings from historical interactions, which often contain preference-irrelevant information and result in insufficient user behavior modeling. Moreover, these models typically generate only item images, providing limited interpretability. To address these limitations, we propose DualFashion, a Dual-Diffusional Generative Fashion Recommendation Architecture that jointly models image and text modalities for personalized and explainable recommendation. DualFashion adopts a dual-diffusion Transformer with image and text branches, where structured attribute-level captions and visual outfit information are jointly used as conditioning signals to model user behavior. The proposed architecture produces both fashion item images and textual descriptions, ensuring visual compatibility while providing explicit semantic interpretability. Furthermore, we introduce a text-augmented fine-tuning strategy that enhances generation diversity and enables effective cross-modal knowledge transfer without incurring heavy computational costs. Extensive experiments on iFashion and Polyvore-U across Personalized Fill-in-the-Blank and Generative Outfit Recommendation tasks demonstrate that DualFashion achieves strong performance in behavior modeling, interpretability, and efficiency compared to state-of-the-art methods. Our code and model checkpoints are available at https://github.com/LinkMingzhe/DualFashion.

preprint2026arXiv

Geomagnetic constraints on Millicharged Dark Matter

Millicharged particles are well-motivated dark matter candidates arising in many extensions of the Standard Model. We show that, despite their tiny coupling $e_m$ to photons, millicharged dark matter (mDM) in the Earth's geomagnetic field can generate a quasi-static, monochromatic magnetic signal with angular frequency twice the mDM mass. Using null results from the SuperMAG and SNIPE Hunt collaborations, we constrain the effective charge of bosonic mDM in the mass range $10^{-18}$--$10^{-14}\,\text{eV}$. The resulting upper bounds exceed stellar cooling constraints by over thirteen orders of magnitude, demonstrating the power of this method.

preprint2023arXiv

Embedding Inequalities for Barron-type Spaces

An important problem in machine learning theory is to understand the approximation and generalization properties of two-layer neural networks in high dimensions. To this end, researchers have introduced the Barron space $\mathcal{B}_s(Ω)$ and the spectral Barron space $\mathcal{F}_s(Ω)$, where the index $s\in [0,\infty)$ indicates the smoothness of functions within these spaces and $Ω\subset\mathbb{R}^d$ denotes the input domain. However, the precise relationship between the two types of Barron spaces remains unclear. In this paper, we establish a continuous embedding between them as implied by the following inequality: for any $δ\in (0,1), s\in \mathbb{N}^{+}$ and $f: Ω\mapsto\mathbb{R}$, it holds that \[ δ\|f\|_{\mathcal{F}_{s-δ}(Ω)}\lesssim_s \|f\|_{\mathcal{B}_s(Ω)}\lesssim_s \|f\|_{\mathcal{F}_{s+1}(Ω)}. \] Importantly, the constants do not depend on the input dimension $d$, suggesting that the embedding is effective in high dimensions. Moreover, we also show that the lower and upper bound are both tight.

preprint2023arXiv

Field-Induced Lifshitz Transition in the Magnetic Weyl Semimetal Candidate PrAlSi

Lifshitz transition (LT) refers to an abrupt change in the electronic structure and Fermi surface, and is associated to a variety of emergent quantum phenomena. Amongst the LTs observed in known materials, the field-induced LT has been rare and its origin remains elusive. To understand the origin of field-induced LT, it is important to extend the material basis beyond the usual setting of heavy fermion metals. Here, we report on a field-induced LT in PrAlSi, a magnetic Weyl semimetal candidate with localized 4$f$ electrons, through a study of magnetotransport up to 55 T. The quantum oscillation analysis reveals that across a threshold field $B^*\approx$14.5 T the oscillation frequency ($F_1$ = 43 T) is replaced by two new frequencies ($F_2$ = 62 T and $F_3$ = 103 T). Strikingly, the LT occurs well below quantum limit, with obvious temperature-dependent oscillation frequency and field-dependent cyclotron mass. Our work not only enriches the rare examples of field-induced LTs, but also paves the way for further investigation on the interplay among topology, magnetism and electronic correlation.

preprint2022arXiv

A kinetic model for rarefied flows of molecular gas with vibrational modes

A kinetic model is proposed for rarefied flows of molecular gas with rotational and temperature-dependent vibrational degrees of freedom. The model reduces to the Boltzmann equation for monatomic gas when the energy exchange between the translational and internal modes is absent, thus the influence of intermolecular potential can be captured. Moreover, not only the transport coefficients but also their fundamental relaxation processes are recovered. The accuracy of our kinetic model is validated by the direct simulation Monte Carlo method in several rarefied gas flows, including the shock wave, Fourier flow, Couette flow, and the creep flow driven by Maxwell's demon. Then the kinetic model is adopted to investigate thermally-induced flows. By adjusting the viscosity index in the Boltzmann collision operator, we find that the intermolecular potential significantly influences the velocity and Knudsen force. Interestingly, in the transition flow regime, the Knudsen force exerting on a heated beam could reverse the direction when the viscosity index changes from 0.5 (hard-sphere gas) to 1 (Maxwell gas). This discovery is useful in the design of micro-electromechanical systems for microstructure actuation and gas sensing.

preprint2022arXiv

A Logarithm Depth Quantum Converter: From One-hot Encoding to Binary Encoding

Within the quantum computing, there are two ways to encode a normalized vector $\{ α_i \}$. They are one-hot encoding and binary coding. The one-hot encoding state is denoted as $\left | ψ_O^{(N)} \right \rangle=\sum_{i=0}^{N-1} α_i \left |0 \right \rangle^{\otimes N-i-1} \left |1 \right \rangle \left |0 \right \rangle ^{\otimes i}$ and the binary encoding state is denoted as $\left | ψ_B^{(N)} \right \rangle=\sum_{i=0}^{N-1} α_i \left |b_i \right \rangle$, where $b_i$ is interpreted in binary of $i$ as the tensor product sequence of qubit states. In this paper, we present a method converting between the one-hot encoding state and the binary encoding state by taking the Edick state as the transition state, where the Edick state is defined as $\left | ψ_E^{(N)} \right \rangle=\sum_{i=0}^{N-1} α_i \left |0 \right \rangle^{\otimes N-i-1} \left |1 \right \rangle ^{\otimes i}$. Compared with the early work, our circuit achieves the exponential speedup with $O(\log^2 N)$ depth and $O(N)$ size.

preprint2022arXiv

A spectral-based analysis of the separation between two-layer neural networks and linear methods

We propose a spectral-based approach to analyze how two-layer neural networks separate from linear methods in terms of approximating high-dimensional functions. We show that quantifying this separation can be reduced to estimating the Kolmogorov width of two-layer neural networks, and the latter can be further characterized by using the spectrum of an associated kernel. Different from previous work, our approach allows obtaining upper bounds, lower bounds, and identifying explicit hard functions in a united manner. We provide a systematic study of how the choice of activation functions affects the separation, in particular the dependence on the input dimension. Specifically, for nonsmooth activation functions, we extend known results to more activation functions with sharper bounds. As concrete examples, we prove that any single neuron can instantiate the separation between neural networks and random feature models. For smooth activation functions, one surprising finding is that the separation is negligible unless the norms of inner-layer weights are polynomially large with respect to the input dimension. By contrast, the separation for nonsmooth activation functions is independent of the norms of inner-layer weights.

preprint2022arXiv

A Survey on EOSIO Systems Security: Vulnerability, Attack, and Mitigation

EOSIO, as one of the most representative blockchain 3.0 platforms, involves lots of new features, e.g., delegated proof of stake consensus algorithm and updatable smart contracts, enabling a much higher transaction per second and the prosperous decentralized applications (DApps) ecosystem. According to the statistics, it has reached nearly 18 billion USD, taking the third place of the whole cryptocurrency market, following Bitcoin and Ethereum. Loopholes, however, are hiding in the shadows. EOSBet, a famous gambling DApp, was attacked twice within a month and lost more than 1 million USD. No existing work has surveyed the EOSIO from a security researcher perspective. To fill this gap, in this paper, we collected all occurred attack events against EOSIO, and systematically studied their root causes, i.e., vulnerabilities lurked in all relying components for EOSIO, as well as the corresponding attacks and mitigations. We also summarized some best practices for DApp developers, EOSIO official team, and security researchers for future directions.

preprint2022arXiv

Axion Dark Radiation: Hubble Tension and Hyper-kamiokande Neutrino Experiment

In this work, we investigate the dark sector of a supersymmetric axion model, consisting of the late-decaying gravitino/axino dark matter and axion dark radiation. In the early universe, the decay of the scalar superpartner of the axion (saxion) will produce a large amount of entropy. The additional entropy can not only dilute the relic density of the gravitino/axino dark matter to avoid overclosing the universe but also relax the constraint on the reheating temperature $T_{R}$ after inflation. Meanwhile, the axion dark radiation from the saxion decay will increase the effective number of neutrino species $N_{\rm eff}$, which can help to reduce the cosmological Hubble tension. In the late universe, the decay of long-lived gravitino/axino dark matter produces the axions with MeV-scale kinetic energy. We study the potential of searching for such energetic axions through the inverse Primakoff process $a+A \to γ+ A$ in the neutrino experiments, such as Hyper-Kamiokande.

preprint2022arXiv

Beyond the Quadratic Approximation: the Multiscale Structure of Neural Network Loss Landscapes

A quadratic approximation of neural network loss landscapes has been extensively used to study the optimization process of these networks. Though, it usually holds in a very small neighborhood of the minimum, it cannot explain many phenomena observed during the optimization process. In this work, we study the structure of neural network loss functions and its implication on optimization in a region beyond the reach of a good quadratic approximation. Numerically, we observe that neural network loss functions possesses a multiscale structure, manifested in two ways: (1) in a neighborhood of minima, the loss mixes a continuum of scales and grows subquadratically, and (2) in a larger region, the loss shows several separate scales clearly. Using the subquadratic growth, we are able to explain the Edge of Stability phenomenon [5] observed for the gradient descent (GD) method. Using the separate scales, we explain the working mechanism of learning rate decay by simple examples. Finally, we study the origin of the multiscale structure and propose that the non-convexity of the models and the non-uniformity of training data is one of the causes. By constructing a two-layer neural network problem we show that training data with different magnitudes give rise to different scales of the loss function, producing subquadratic growth and multiple separate scales.

preprint2022arXiv

Deep Learning Jet Image as a Probe of Light Higgsino Dark Matter at the LHC

Higgsino in supersymmetric standard models can play the role of dark matter particle. In conjunction with the naturalness criterion, the higgsino mass parameter is expected to be around the electroweak scale. In this work, we explore the potential of probing the nearly degenerate light higgsinos with machine learning at the LHC. By analyzing jet images and other jet substructure information, we use the Convolutional Neural Network(CNN) to enhance the signal significance. We find that our deep learning jet image method can improve the previous result based on the conventional cut-flow by about a factor of two at the High-Luminosity LHC.

preprint2022arXiv

Electroweak Precision Fit and New Physics in light of $W$ Boson Mass

The $W$ boson mass is one of the most important electroweak precision observables for testing the Standard Model or its extensions. The very recent measured $W$ boson mass at CDF shows about $7σ$ deviations from the SM prediction, which may challenge the internal consistency of the SM. By performing the global electroweak fit with the new $W$-boson, we present the new values of the oblique parameters: $S = 0.06 \pm 0.10$, $T= 0.11 \pm 0.12$, $U=0.13 \pm 0.09$, or $S=0.14 \pm 0.08$, $T= 0.26 \pm 0.06$ with $U =0$ and the corresponding correlation matrices, which strongly indicates the need for the non-degenerate multiplets beyond the SM. As a proof-of-concept, we show that the new results can be accommodated in the two-Higgs doublet model, where the charged Higgs boson has to be either heavier or lighter than both two heavy neutral Higgs bosons. Therefore, searching for these non-SM Higgs bosons will provide a complementary way to test the new physics for the $W$ boson mass anomaly.

preprint2022arXiv

Exploiting the Potential of Datasets: A Data-Centric Approach for Model Robustness

Robustness of deep neural networks (DNNs) to malicious perturbations is a hot topic in trustworthy AI. Existing techniques obtain robust models given fixed datasets, either by modifying model structures, or by optimizing the process of inference or training. While significant improvements have been made, the possibility of constructing a high-quality dataset for model robustness remain unexplored. Follow the campaign of data-centric AI launched by Andrew Ng, we propose a novel algorithm for dataset enhancement that works well for many existing DNN models to improve robustness. Transferable adversarial examples and 14 kinds of common corruptions are included in our optimized dataset. In the data-centric robust learning competition hosted by Alibaba Group and Tsinghua University, our algorithm came third out of more than 3000 competitors in the first stage while we ranked fourth in the second stage. Our code is available at \url{https://github.com/hncszyq/tianchi_challenge}.

preprint2022arXiv

Global fits of SUSY at future Higgs factories

In this work, we study the impact of electroweak and Higgs precision measurements at future electron-positron colliders on several typical supersymmetric models, including the Constrained Minimal Supersymmetric Standard Model (CMSSM), Non-Universal Higgs Mass generalisations (NUHM1, NUHM2), and the 7-dimensional Minimal Supersymmetric Standard Model (MSSM7). Using publicly-available data from the \textsf{GAMBIT} community, we post-process previous SUSY global fits with additional likelihoods to explore the discovery potential of Higgs factories, such as the Circular Electron Positron Collider (CEPC), the Future Circular Collider (FCC) and the International Linear Collider (ILC). We show that the currently allowed parameter space of these models will be further tested by future precision measurements. In particular, dark matter annihilation mechanisms may be distinguished by precise measurements of Higgs observables.

preprint2022arXiv

Hilbert Expansion for the Relativistic Landau Equation

In this paper, we study the local-in-time validity of the Hilbert expansion for the relativistic Landau equation. We justify that solutions of the relativistic Landau equation converge to small classical solutions of the limiting relativistic Euler equations as the Knudsen number shrinks to zero in a weighted Sobolev space. The key difficulty comes from the temporal and spatial derivatives of the local Maxwellian, which produce momentum growth terms and are uncontrollable by the standard $L^2$-based energy and dissipation. We introduce novel time-dependent weight functions to generate additional dissipation terms to suppress the large momentum. The argument relies on a hierarchy of energy-dissipation structures with or without weights. As far as the authors are aware of, this is the first result of the Hilbert expansion for the Landau-type equation.

preprint2022arXiv

HoVer-Trans: Anatomy-aware HoVer-Transformer for ROI-free Breast Cancer Diagnosis in Ultrasound Images

Ultrasonography is an important routine examination for breast cancer diagnosis, due to its non-invasive, radiation-free and low-cost properties. However, the diagnostic accuracy of breast cancer is still limited due to its inherent limitations. It would be a tremendous success if we can precisely diagnose breast cancer by breast ultrasound images (BUS). Many learning-based computer-aided diagnostic methods have been proposed to achieve breast cancer diagnosis/lesion classification. However, most of them require a pre-define ROI and then classify the lesion inside the ROI. Conventional classification backbones, such as VGG16 and ResNet50, can achieve promising classification results with no ROI requirement. But these models lack interpretability, thus restricting their use in clinical practice. In this study, we propose a novel ROI-free model for breast cancer diagnosis in ultrasound images with interpretable feature representations. We leverage the anatomical prior knowledge that malignant and benign tumors have different spatial relationships between different tissue layers, and propose a HoVer-Transformer to formulate this prior knowledge. The proposed HoVer-Trans block extracts the inter- and intra-layer spatial information horizontally and vertically. We conduct and release an open dataset GDPH&SYSUCC for breast cancer diagnosis in BUS. The proposed model is evaluated in three datasets by comparing with four CNN-based models and two vision transformer models via five-fold cross validation. It achieves state-of-the-art classification performance with the best model interpretability. In the meanwhile, our proposed model outperforms two senior sonographers on the breast cancer diagnosis when only one BUS image is given.

preprint2022arXiv

iLibScope: Reliable Third-Party Library Detection for iOS Mobile Apps

Vetting security impacts introduced by third-party libraries in iOS apps requires a reliable library detection technique. Especially when a new vulnerability (or a privacy-invasive behavior) was discovered in a third-party library, there is a practical need to precisely identify the existence of libraries and their versions for iOS apps. However, few studies have been proposed to tackle this problem, and they all suffer from the code duplication problem in different libraries. In this paper, we focus on third-party library detection in iOS apps. Given an app, we aim to identify the integrated libraries and pinpoint their versions (or the version range).To this end, we first conduct an in-depth study on iOS third-party libraries to demystify the code duplication challenge. By doing so, we have two key observations: 1) even though two libraries can share classes, the shared classes cannot be integrated into an app simultaneously without causing a class name conflict; and 2) code duplication between multiple versions of two libraries can vary. Based on these findings, we propose a novel profile-based similarity comparison approach to perform the detection. Specifically, we build a library database consists of original library binaries with distinct versions. After extracting profiles for each library version and the target app, we conduct a similarity comparison to find the best matches. We implemented this approach in iLibScope. We built a benchmark consists of 5,807 apps with 10,495 library integrations and applied our tool to it. Our evaluation shows that iLibScope achieves a recall exceeds 99% and a precision exceeds 97% for library detection. We also applied iLibScope to detect the presence of well-known vulnerable third-party libraries in real-world iOS mobile apps to show the promising usage of our tool. It successfully identified 405 vulnerable library usage from 4,249 apps.

preprint2022arXiv

Inert Higgs Dark Matter for CDF-II W-boson Mass and Detection Prospects

The $W$-boson mass, which was recently measured at FermiLab with an unprecedented precision, suggests the presence of new multiplets beyond the Standard Model (SM). One of the minimal extensions of the SM is to introduce an additional scalar doublet, in which the non-SM scalars can enhance $W$-boson mass via the loop corrections. On the other hand, with a proper discrete symmetry, the lightest new scalar in the doublet can be stable and play the role of dark matter particle. We show that the inert two Higgs doublet model can naturally handle the new $W$-boson mass without violating other constraints, and the preferred dark matter mass is between $54$ and $74$ GeV. We identify three feasible parameter regions for the thermal relic density: the $SA$ co-annihilation, the Higgs resonance, and the $SS \to WW^*$ annihilation. We find that the first region can be fully tested by the HL-LHC, the second region will be tightly constrained by direct detection experiments, and the third region could yield detectable GeV gamma-ray and antiproton signals in the Galaxy that may have been observed by Fermi-LAT and AMS-02.

preprint2022arXiv

Learning a Single Neuron for Non-monotonic Activation Functions

We study the problem of learning a single neuron $\mathbf{x}\mapsto σ(\mathbf{w}^T\mathbf{x})$ with gradient descent (GD). All the existing positive results are limited to the case where $σ$ is monotonic. However, it is recently observed that non-monotonic activation functions outperform the traditional monotonic ones in many applications. To fill this gap, we establish learnability without assuming monotonicity. Specifically, when the input distribution is the standard Gaussian, we show that mild conditions on $σ$ (e.g., $σ$ has a dominating linear part) are sufficient to guarantee the learnability in polynomial time and polynomial samples. Moreover, with a stronger assumption on the activation function, the condition of input distribution can be relaxed to a non-degeneracy of the marginal distribution. We remark that our conditions on $σ$ are satisfied by practical non-monotonic activation functions, such as SiLU/Swish and GELU. We also discuss how our positive results are related to existing negative results on training two-layer neural networks.

preprint2022arXiv

Logarithmic cotangent bundles, Chern-Mather classes, and the Huh-Sturmfels Involution conjecture

Using compactifications in the logarithmic cotangent bundle, we obtain a formula for the Chern classes of the pushforward of Lagrangian cycles under an open embedding with normal crossing complement. This generalizes earlier results of Aluffi and Wu-Zhou. The first application of our formula is a geometric description of Chern-Mather classes of an arbitrary very affine variety, generalizing earlier results of Huh which held under the smooth and schon assumptions. As the second application, we confirm an involution formula relating sectional maximum likelihood (ML) degrees and ML bidegrees, which was conjectured by Huh and Sturmfels in 2013.

preprint2022arXiv

Non-monotonic heat dissipation phenomenon in close-packed quasi-2D and 3D hotspot system

Transient heat dissipation in close-packed quasi-2D nanoline and 3D nanocuboid hotspot systems is studied based on phonon Boltzmann transport equation. It is found that, counter-intuitively, the heat dissipation efficiency is not a monotonic function of the distance between adjacent nanoscale heat sources: the heat dissipation efficiency reaches the highest value when this distance is comparable to the phonon mean free path. This is due to the competition of two thermal transport processes: quasiballistic transport when phonons escape from the nanoscale heat source and the scattering among phonons originating from adjacent nanoscale heat source.

preprint2022arXiv

On self-supervised multi-modal representation learning: An application to Alzheimer's disease

Introspection of deep supervised predictive models trained on functional and structural brain imaging may uncover novel markers of Alzheimer's disease (AD). However, supervised training is prone to learning from spurious features (shortcut learning) impairing its value in the discovery process. Deep unsupervised and, recently, contrastive self-supervised approaches, not biased to classification, are better candidates for the task. Their multimodal options specifically offer additional regularization via modality interactions. In this paper, we introduce a way to exhaustively consider multimodal architectures for contrastive self-supervised fusion of fMRI and MRI of AD patients and controls. We show that this multimodal fusion results in representations that improve the results of the downstream classification for both modalities. We investigate the fused self-supervised features projected into the brain space and introduce a numerically stable way to do so.

preprint2022arXiv

On the emergence of heat waves in the transient thermal grating geometry

The propagation of heat in the transient thermal grating geometry is studied based on phonon Boltzmann transport equation (BTE) in different phonon transport regimes. Our analytical and numerical results show that the phonon dispersion relation and temperature play a significant role in the emergence of heat wave. For the frequency-independent BTE, the heat wave appears as long as the phonon resistive scattering is not sufficient, while for the frequency-dependent BTE, the heat wave could disappear in the ballistic regime, depending on the grating period and temperature. We predict that the heat wave could appear in the suspended graphene and silicon in extremely low temperature but disappear at room temperature.

preprint2022arXiv

Penny Wise and Pound Foolish: Quantifying the Risk of Unlimited Approval of ERC20 Tokens on Ethereum

The prosperity of decentralized finance motivates many investors to profit via trading their crypto assets on decentralized applications (DApps for short) of the Ethereum ecosystem. Apart from Ether (the native cryptocurrency of Ethereum), many ERC20 (a widely used token standard on Ethereum) tokens obtain vast market value in the ecosystem. Specifically, the approval mechanism is used to delegate the privilege of spending users' tokens to DApps. By doing so, the DApps can transfer these tokens to arbitrary receivers on behalf of the users. To increase the usability, unlimited approval is commonly adopted by DApps to reduce the required interaction between them and their users. However, as shown in existing security incidents, this mechanism can be abused to steal users' tokens. In this paper, we present the first systematic study to quantify the risk of unlimited approval of ERC20 tokens on Ethereum. Specifically, by evaluating existing transactions up to 31st July 2021, we find that unlimited approval is prevalent (60%, 15.2M/25.4M) in the ecosystem, and 22% of users have a high risk of their approved tokens for stealing. After that, we investigate the security issues that are involved in interacting with the UIs of 22 representative DApps and 9 famous wallets to prepare the approval transactions. The result reveals the worrisome fact that all DApps request unlimited approval from the front-end users and only 10% (3/31) of UIs provide explanatory information for the approval mechanism. Meanwhile, only 16% (5/31) of UIs allow users to modify their approval amounts. Finally, we take a further step to characterize the user behavior into five modes and formalize the good practice, i.e., on-demand approval and timely spending, towards securely spending approved tokens. However, the evaluation result suggests that only 0.2% of users follow the good practice to mitigate the risk.

preprint2022arXiv

Probing ultra-light dark photon from inverse Compton-like scattering

Dark photon not only provides a portal linking dark sector particles and ordinary matter but also is a well-motivated dark matter candidate. We propose to detect the dark photon dark matter through the inverse Compton-like scattering process $p+γ^\prime \to p+γ$. Thanks to the ultra-high energy primary cosmic rays, we find that such a method is able to probe the dark photon mass from $10^{-2}$ eV down to $10^{-19}$ eV with the expected sensitivity of eROSITA $X$-ray telescope, which can extend the current lower limit of dark photon mass from Jupiter's magnetic fields experiment by about three orders of magnitude.

preprint2022arXiv

Rarefaction effects in head-on collision of two identical droplets

The head-on collision of two identical droplets is investigated based on the BGK-Boltzmann equation. Gauss-Hermite quadratures with different degree of precision are used to solve the kinetic equation, so that the continuum (solution truncated at the Navier-Stokes order) and non-continuum (rarefied gas dynamics) solutions can be compared. When the kinetic equation is solved with adequate accuracy, prominent variations of the vertical velocity (the collision is in the horizontal direction), the viscous stress components, and droplet morphology are observed during the formation of liquid bridge, which demonstrates the importance of the rarefaction effects and the failure of the Navier-Stokes equation. The rarefaction effects change the topology of streamlines near the droplet surface, suppress the high-magnitude vorticity concentration inside the interdroplet region, and promote the vorticity diffusion around outer droplet surface. Two physical mechanisms responsible for the local energy conversion between the free and kinetic energies are identified, namely, the total pressure-dilatation coupling effect and the interaction between the density gradient and strain rate tensor. An energy conversion analysis is performed to show that the rarefaction effects can enhance the conversion from free energy to kinetic energy and facilitate the discharge of interdroplet gas film along the vertical direction, thereby boosting droplet coalescence.

preprint2022arXiv

Self-supervised multimodal neuroimaging yields predictive representations for a spectrum of Alzheimer's phenotypes

Recent neuroimaging studies that focus on predicting brain disorders via modern machine learning approaches commonly include a single modality and rely on supervised over-parameterized models.However, a single modality provides only a limited view of the highly complex brain. Critically, supervised models in clinical settings lack accurate diagnostic labels for training. Coarse labels do not capture the long-tailed spectrum of brain disorder phenotypes, which leads to a loss of generalizability of the model that makes them less useful in diagnostic settings. This work presents a novel multi-scale coordinated framework for learning multiple representations from multimodal neuroimaging data. We propose a general taxonomy of informative inductive biases to capture unique and joint information in multimodal self-supervised fusion. The taxonomy forms a family of decoder-free models with reduced computational complexity and a propensity to capture multi-scale relationships between local and global representations of the multimodal inputs. We conduct a comprehensive evaluation of the taxonomy using functional and structural magnetic resonance imaging (MRI) data across a spectrum of Alzheimer's disease phenotypes and show that self-supervised models reveal disorder-relevant brain regions and multimodal links without access to the labels during pre-training. The proposed multimodal self-supervised learning yields representations with improved classification performance for both modalities. The concomitant rich and flexible unsupervised deep learning framework captures complex multimodal relationships and provides predictive performance that meets or exceeds that of a more narrow supervised classification analysis. We present elaborate quantitative evidence of how this framework can significantly advance our search for missing links in complex brain disorders.

preprint2022arXiv

Simple and statistically sound recommendations for analysing physical theories

Physical theories that depend on many parameters or are tested against data from many different experiments pose unique challenges to statistical inference. Many models in particle physics, astrophysics and cosmology fall into one or both of these categories. These issues are often sidestepped with statistically unsound ad hoc methods, involving intersection of parameter intervals estimated by multiple experiments, and random or grid sampling of model parameters. Whilst these methods are easy to apply, they exhibit pathologies even in low-dimensional parameter spaces, and quickly become problematic to use and interpret in higher dimensions. In this article we give clear guidance for going beyond these procedures, suggesting where possible simple methods for performing statistically sound inference, and recommendations of readily-available software tools and standards that can assist in doing so. Our aim is to provide any physicists lacking comprehensive statistical training with recommendations for reaching correct scientific conclusions, with only a modest increase in analysis burden. Our examples can be reproduced with the code publicly available at https://doi.org/10.5281/zenodo.4322283.

preprint2022arXiv

Tasting the cake: evaluating self-supervised generalization on out-of-distribution multimodal MRI data

Self-supervised learning has enabled significant improvements on natural image benchmarks. However, there is less work in the medical imaging domain in this area. The optimal models have not yet been determined among the various options. Moreover, little work has evaluated the current applicability limits of novel self-supervised methods. In this paper, we evaluate a range of current contrastive self-supervised methods on out-of-distribution generalization in order to evaluate their applicability to medical imaging. We show that self-supervised models are not as robust as expected based on their results in natural imaging benchmarks and can be outperformed by supervised learning with dropout. We also show that this behavior can be countered with extensive augmentation. Our results highlight the need for out-of-distribution generalization standards and benchmarks to adopt the self-supervised methods in the medical imaging community.

preprint2021arXiv

A fast-converging scheme for the Phonon Boltzmann equation with dual relaxation times

Callaway's dual relaxation times model, which takes into account the normal and resistive scatterings of phonon, is used to describe the heat conduction in materials like graphene. For steady-state problems, the Callaway model is usually solved by the conventional iterative scheme (CIS), which is efficient in the ballistic regime, but inefficient in the diffusive/hydrodynamic regime. In this paper, a general synthetic iterative scheme (GSIS) is proposed to expedite the convergence to steady-state solutions. First, macroscopic synthetic equations are designed to guide the evolution of equilibrium distribution functions for normal and resistive scatterings, so that fast convergence can be achieved even in the diffusive/hydrodynamic regime. Second, the Fourier stability analysis is conducted to find the convergence rate for both CIS and GSIS, which rigorously proves the efficiency of GSIS over CIS. Finally, several numerical simulations are carried out to demonstrate the accuracy and efficiency of GSIS, where up to three orders of magnitude of convergence acceleration is achieved.

preprint2021arXiv

Direct Detection of Spin-Dependent Sub-GeV Dark Matter via Migdal Effect

Motivated by the current strong constraints on the spin-independent dark matter (DM)-nucleus scattering, we investigate the spin-dependent (SD) interactions of the light Majorana DM with the nucleus mediated by an axial-vector boson. Due to the small nucleus recoil energy, the ionization signals have now been used to probe the light dark matter particles in direct detection experiments. With the existing ionization data, we derive the exclusion limits on the SD DM-nucleus scattering through Migdal effect in the MeV-GeV DM mass range. It is found that the lower limit of the DM mass can reach about several MeVs. Due to the momentum transfer correction induced by the light mediator, the bounds on the SD DM-nucleus scattering cross sections can be weakened in comparison with the heavy mediator.

preprint2021arXiv

On the Quantum Boltzmann Equation near Maxwellian and Vacuum

We consider the non-relativistic quantum Boltzmann equation for fermions and bosons. Using the nonlinear energy method and mild formulation, we justify the global well-posedness when the density function is near the global Maxwellian and vacuum. This work is a generalization and adaptation of the classical Boltzmann theory. Our main contribution is a detailed analysis of the nonlinear operator $Q$ in the quantum context. This is the first piece of a long-term project on the quantum kinetic equations.

preprint2021arXiv

The Generalization Error of the Minimum-norm Solutions for Over-parameterized Neural Networks

We study the generalization properties of minimum-norm solutions for three over-parametrized machine learning models including the random feature model, the two-layer neural network model and the residual network model. We proved that for all three models, the generalization error for the minimum-norm solution is comparable to the Monte Carlo rate, up to some logarithmic terms, as long as the models are sufficiently over-parametrized.

preprint2021arXiv

Towards Understanding and Demystifying Bitcoin Mixing Services

One reason for the popularity of Bitcoin is due to its anonymity. Although several heuristics have been used to break the anonymity, new approaches are proposed to enhance its anonymity at the same time. One of them is the mixing service. Unfortunately, mixing services have been abused to facilitate criminal activities, e.g., money laundering. As such, there is an urgent need to systematically understand Bitcoin mixing services. In this paper, we take the first step to understand state-of-the-art Bitcoin mixing services. Specifically, we propose a generic abstraction model for mixing services and observe that there are two mixing mechanisms in the wild, i.e. {swapping} and {obfuscating}. Based on this model, we conduct a transaction-based analysis and successfully reveal the mixing mechanisms of four representative services. Besides, we propose a method to identify mixing transactions that leverage the obfuscating mechanism. The proposed approach is able to identify over $92$\% of the mixing transactions. Based on identified transactions, we then estimate the profit of mixing services and provide a case study of tracing the money flow of stolen Bitcoins.

preprint2021arXiv

VM Matters: A Comparison of WASM VMs and EVMs in the Performance of Blockchain Smart Contracts

WebAssemly is an emerging runtime for Web applications and has been supported in almost all browsers. Recently, WebAssembly is further regarded to be a the next-generation environment for blockchain applications, and has been adopted by Ethereum, namely eWASM, to replace the state-of-the-art EVM. However, whether and how well current eWASM outperforms EVM on blockchain clients is still unknown. This paper conducts the first measurement study, to measure the performance on WASM VM and EVM for executing smart contracts on blockchain. To our surprise, the current WASM VM does not perform in expected performance. The overhead introduced by WASM is really non-trivial. Our results highlight the challenges when deploying WASM in practice, and provide insightful implications for improvement space.

preprint2020arXiv

A Comparative Analysis of the Optimization and Generalization Property of Two-layer Neural Network and Random Feature Models Under Gradient Descent Dynamics

A fairly comprehensive analysis is presented for the gradient descent dynamics for training two-layer neural network models in the situation when the parameters in both layers are updated. General initialization schemes as well as general regimes for the network width and training data size are considered. In the over-parametrized regime, it is shown that gradient descent dynamics can achieve zero training loss exponentially fast regardless of the quality of the labels. In addition, it is proved that throughout the training process the functions represented by the neural network model are uniformly close to that of a kernel method. For general values of the network width and training data size, sharp estimates of the generalization error is established for target functions in the appropriate reproducing kernel Hilbert space.

preprint2020arXiv

A Priori Estimates of the Population Risk for Two-layer Neural Networks

New estimates for the population risk are established for two-layer neural networks. These estimates are nearly optimal in the sense that the error rates scale in the same way as the Monte Carlo error rates. They are equally effective in the over-parametrized regime when the network size is much larger than the size of the dataset. These new estimates are a priori in nature in the sense that the bounds depend only on some norms of the underlying functions to be fitted, not the parameters in the model, in contrast with most existing results which are a posteriori in nature. Using these a priori estimates, we provide a perspective for understanding why two-layer neural networks perform better than the related kernel methods.

preprint2020arXiv

Atmospheric Dark Matter and Xenon1T Excess

Very recently, the Xenon1T collaboration has reported an intriguing electron recoil excess, which may imply for light dark matter. In order to interpret this anomaly, we propose the atmospheric dark matter (ADM) from the inelastic collision of cosmic rays (CRs) with the atmosphere. Due to the boost effect of high energy CRs, we show that the light ADM can be fast-moving and successfully fit the observed electron recoil spectrum through the ADM-electron scattering process. Meanwhile, our ADM predicts the scattering cross section $σ_e \sim {\cal O}(10^{-38}- 10^{-39}$) cm$^{2}$, and thus can evade other direct detection constraints. The search for light meson rare decays, such as $η\to π+ \slashed E_T$, would provide a complementary probe of our ADM in the future.

preprint2020arXiv

Automated Deobfuscation of Android Native Binary Code

With the popularity of Android apps, different techniques have been proposed to enhance app protection. As an effective approach to prevent reverse engineering, obfuscation can be used to serve both benign and malicious purposes. In recent years, more and more sensitive logic or data have been implemented as obfuscated native code because of the limitations of Java bytecode. As a result, native code obfuscation becomes a great obstacle for security analysis to understand the complicated logic. In this paper, we propose DiANa, an automated system to facilitate the deobfuscation of native binary code in Android apps. Specifically, given a binary obfuscated by Obfuscator-LLVM (the most popular native code obfuscator), DiANa is capable of recovering the original Control Flow Graph. To the best of our knowledge, DiANa is the first system that aims to tackle the problem of Android native binary deobfuscation. We have applied DiANa in different scenarios, and the experimental results demonstrate the effectiveness of DiANa based on generic similarity comparison metrics.

preprint2020arXiv

Bernstein-Sato ideals and hyperplane arrangements

We study the relation between zero loci of Bernstein-Sato ideals and roots of b-functions and obtain a criterion to guarantee that roots of b-functions of a reducible polynomial are determined by the zero locus of the associated Bernstein-Sato ideal. Applying the criterion together with a result of Maisonobe we prove that the set of roots of the b-function of a free hyperplane arrangement is determined by its intersection lattice. We also study the zero loci of Bernstein-Sato ideals and the associated relative characteristic cycles for arbitrary central hyperplane arrangements. We prove the multivariable n/d conjecture of Budur for complete factorizations of arbitrary hyperplane arrangements, which in turn proves the strong monodromy conjecture for the associated multivariable topological zeta functions.

preprint2020arXiv

Calibrating the dynamic Huff model for business analysis using location big data

The Huff model has been widely used in location-based business analysis for delineating a trading area containing potential customers to a store. Calibrating the Huff model and its extensions requires empirical location visit data. Many studies rely on labor-intensive surveys. With the increasing availability of mobile devices, users in location-based platforms share rich multimedia information about their locations in a fine spatiotemporal resolution, which offers opportunities for business intelligence. In this research, we present a time-aware dynamic Huff model (T-Huff) for location-based market share analysis and calibrate this model using large-scale store visit patterns based on mobile phone location data across ten most populated U.S. cities. By comparing the hourly visit patterns of two types of stores, we demonstrate that the calibrated T-Huff model is more accurate than the original Huff model in predicting the market share of different types of business (e.g., supermarkets vs. department stores) over time. We also identify the regional variability where people in large metropolitan areas with a well-developed transit system show less sensitivity to long-distance visits. In addition, several socioeconomic and demographic factors (e.g., median household income) that potentially affect people's visit decisions are examined and summarized.

preprint2020arXiv

Characterizing Cryptocurrency Exchange Scams

As the indispensable trading platforms of the ecosystem, hundreds of cryptocurrency exchanges are emerging to facilitate the trading of digital assets. While, it also attracts the attentions of attackers. A number of scam attacks were reported targeting cryptocurrency exchanges, leading to a huge mount of financial loss. However, no previous work in our research community has systematically studied this problem. In this paper, we make the first effort to identify and characterize the cryptocurrency exchange scams. We first identify over 1,500 scam domains and over 300 fake apps, by collecting existing reports and using typosquatting generation techniques. Then we investigate the relationship between them, and identify 94 scam domain families and 30 fake app families. We further characterize the impacts of such scams, and reveal that these scams have incurred financial loss of 520k US dollars at least. We further observe that the fake apps have been sneaked to major app markets (including Google Play) to infect unsuspicious users. Our findings demonstrate the urgency to identify and prevent cryptocurrency exchange scams. To facilitate future research, we have publicly released all the identified scam domains and fake apps to the community.

preprint2020arXiv

Characterizing EOSIO Blockchain

EOSIO has become one of the most popular blockchain platforms since its mainnet launch in June 2018. In contrast to the traditional PoW-based systems (e.g., Bitcoin and Ethereum), which are limited by low throughput, EOSIO is the first high throughput Delegated Proof of Stake system that has been widely adopted by many applications. Although EOSIO has millions of accounts and billions of transactions, little is known about its ecosystem, especially related to security and fraud. In this paper, we perform a large-scale measurement study of the EOSIO blockchain and its associated DApps. We gather a large-scale dataset of EOSIO and characterize activities including money transfers, account creation and contract invocation. Using our insights, we then develop techniques to automatically detect bots and fraudulent activity. We discover thousands of bot accounts (over 30\% of the accounts in the platform) and a number of real-world attacks (301 attack accounts). By the time of our study, 80 attack accounts we identified have been confirmed by DApp teams, causing 828,824 EOS tokens losses (roughly 2.6 million US\$) in total.

preprint2020arXiv

Complexity Measures for Neural Networks with General Activation Functions Using Path-based Norms

A simple approach is proposed to obtain complexity controls for neural networks with general activation functions. The approach is motivated by approximating the general activation functions with one-dimensional ReLU networks, which reduces the problem to the complexity controls of ReLU networks. Specifically, we consider two-layer networks and deep residual networks, for which path-based norms are derived to control complexities. We also provide preliminary analyses of the function spaces induced by these norms and a priori estimates of the corresponding regularized estimators.

preprint2020arXiv

DEPOSafe: Demystifying the Fake Deposit Vulnerability in Ethereum Smart Contracts

Cryptocurrency has seen an explosive growth in recent years, thanks to the evolvement of blockchain technology and its economic ecosystem. Besides Bitcoin, thousands of cryptocurrencies have been distributed on blockchains, while hundreds of cryptocurrency exchanges are emerging to facilitate the trading of digital assets. At the same time, it also attracts the attentions of attackers. Fake deposit, as one of the most representative attacks (vulnerabilities) related to exchanges and tokens, has been frequently observed in the blockchain ecosystem, causing large financial losses. However, besides a few security reports, our community lacks of the understanding of this vulnerability, for example its scale and the impacts. In this paper, we take the first step to demystify the fake deposit vulnerability. Based on the essential patterns we have summarized, we implement DEPOSafe, an automated tool to detect and verify (exploit) the fake deposit vulnerability in ERC-20 smart contracts. DEPOSafe incorporates several key techniques including symbolic execution based static analysis and behavior modeling based dynamic verification. By applying DEPOSafe to 176,000 ERC-20 smart contracts, we have identified over 7,000 vulnerable contracts that may suffer from two types of attacks. Our findings demonstrate the urgency to identify and prevent the fake deposit vulnerability.

preprint2020arXiv

Fast convergence and asymptotic preserving of the General Synthetic Iterative Scheme

Recently the general synthetic iteration scheme (GSIS) is proposed to find the steady-state solution of the Boltzmann equation~\cite{SuArXiv2019}, where various numerical simulations have shown that (i) the steady-state solution can be found within dozens of iterations at any Knudsen number $K$, and (ii) the solution is accurate even when the spatial cell size in the bulk region is much larger than the molecular mean free path, i.e. Navier-Stokes solutions are recovered at coarse grids. The first property indicates that the error decay rate between two consecutive iterations decreases to zero with $K$, while the second one implies that the GSIS is asymptotically preserving the Navier-Stokes limit. This paper is dedicated to the rigorous proof of both properties.

preprint2020arXiv

General synthetic iteration scheme for non-linear gas kinetic simulation of multi-scale rarefied gas flows

The general synthetic iteration scheme (GSIS) is extended to find the steady-state solution of nonlinear gas kinetic equation, removing the long-standing problems of slow convergence and requirement of ultra-fine grids in near-continuum flows. The key ingredients of GSIS are that the gas kinetic equation and macroscopic synthetic equations are tightly coupled, and the constitutive relations in macroscopic synthetic equations explicitly contain Newton's law of shear stress and Fourier's law of heat conduction. The higher-order constitutive relations describing rarefaction effects are calculated from the velocity distribution function, however, their constructions are simpler than our previous work (Su et al. Journal of Computational Physics 407 (2020) 109245) for linearized gas kinetic equations. On the other hand, solutions of macroscopic synthetic equations are used to inform the evolution of gas kinetic equation at the next iteration step. A rigorous linear Fourier stability analysis in periodic system shows that the error decay rate of GSIS can be smaller than 0.5, which means that the deviation to steady-state solution can be reduced by 3 orders of magnitude in 10 iterations. Other important advantages of the GSIS are (i) it does not rely on the specific form of Boltzmann collision operator and (ii) it can be solved by sophisticated techniques in computational fluid dynamics, making it amenable to large scale engineering applications. In this paper, the efficiency and accuracy of GSIS is demonstrated by a number of canonical test cases in rarefied gas dynamics.

preprint2020arXiv

Isotriviality of smooth families of varieties of general type

In this paper, we proved that a log smooth family of log general type klt pairs with a special (in the sense of Campana) quasi-projective base is isotrivial. As a consequence, we proved the generalized Kebekus-Kovács conjecture \cite[Conjecture 1.1]{WW19}, for smooth families of general type varieties as well as log smooth families of log canonical pairs of log general type, assuming the existence of relative good minimal models.

preprint2020arXiv

LFV and g-2 in non-universal SUSY models with light higgsinos

We consider a supersymmetric type-I seesaw framework with non-universal scalar masses at the GUT scale to explain the long-standing discrepancy of the anomalous magnetic moment of the muon. We find that it is difficult to accommodate the muon g-2 while keeping charged-lepton flavor violating processes under control for the conventional SO(10)-based relation between the up sector and neutrino sector. However, such tension can be relaxed by adding a Georgi-Jarlskog factor for the Yukawa matrices, which requires a non-trivial GUT-based model. In this model, we find that both observables are compatible for small mixings, CKM-like, in the neutrino Dirac Yukawa matrix.

preprint2020arXiv

Log D-modules and Index theorems

We study log D-modules on smooth log pairs and construct a comparison theorem of log de Rham complexes. The proof uses Sabbah's generalized b-functions. As applications, we deduce a log index theorem and a Riemann-Roch type formula for perverse sheaves on smooth quasi-projective varieties. The log index theorem naturally generalizes the Dubson-Kashiwara index theorem on smooth projective varieties.

preprint2020arXiv

Nearby and vanishing cycles for perverse sheaves and D-modules

We survey nearby and vanishing cycles for both perverse sheaves and D-modules under analytic setting. Following ideas of A. Beilinson, M. Kashiwara and M. Saito, we explain in detail the proof of the comparison theorem between them in the sense of Riemann-Hilbert correspondence.

preprint2020arXiv

On the comparison of nearby cycles via $b$-functions

In this article, we give a simple proof of the comparison of nearby and vanishing cycles in the sense of Riemann-Hilbert correspondence following the idea of Beilinson and Bernstein, without using the Kashiwara-Malgrange $V$-filtrations.

preprint2020arXiv

On the coverage of neutralino dark matter in coannihilations at the upgraded LHC

In the supersymmetric models, the coannihilation of the neutralino DM with a lighter supersymmetric particle provides a feasible way to accommodate the observed cosmological DM relic density. Such a mechanism predicts a compressed spectrum of the neutralino DM and its coannihilating partner, which results in the soft final states and makes the searches for sparticles challenging at colliders. On the other hand, the abundance of the freeze-out neutralino DM usually increases as the DM mass becomes heavier. This implies an upper bound on the mass of the neutralino DM. Given these observations, we explore the HE-LHC coverage of the neutralino DM for the coannihilations. By analyzing the events of the multijet with the missing transverse energy ($E^{miss}_T$), the monojet, the soft lepton pair plus $E^{miss}_T$, and the monojet plus a hadronic tau, we find that the neutralino DM mass can be excluded up to 2.6, 1.7 and 0.8 TeV in the gluino, stop and wino coannihilations at the $2σ$ level, respectively. However, there is still no sensitivity of the neutralino DM in stau coannihilation at the HE-LHC, due to the small cross section of the direct stau pair production and the low tagging efficiency of soft tau from the stau decay.

preprint2020arXiv

Reinterpretation of LHC Results for New Physics: Status and Recommendations after Run 2

We report on the status of efforts to improve the reinterpretation of searches and measurements at the LHC in terms of models for new physics, in the context of the LHC Reinterpretation Forum. We detail current experimental offerings in direct searches for new particles, measurements, technical implementations and Open Data, and provide a set of recommendations for further improving the presentation of LHC results in order to better enable reinterpretation in the future. We also provide a brief description of existing software reinterpretation frameworks and recent global analyses of new physics that make use of the current data.

preprint2020arXiv

Security Analysis of EOSIO Smart Contracts

The EOSIO blockchain, one of the representative Delegated Proof-of-Stake (DPoS) blockchain platforms, has grown rapidly recently. Meanwhile, a number of vulnerabilities and high-profile attacks against top EOSIO DApps and their smart contracts have also been discovered and observed in the wild, resulting in serious financial damages. Most of EOSIO's smart contracts are not open-sourced and they are typically compiled to WebAssembly (Wasm) bytecode, thus making it challenging to analyze and detect the presence of possible vulnerabilities. In this paper, we propose EOSAFE, the first static analysis framework that can be used to automatically detect vulnerabilities in EOSIO smart contracts at the bytecode level. Our framework includes a practical symbolic execution engine for Wasm, a customized library emulator for EOSIO smart contracts, and four heuristics-driven detectors to identify the presence of four most popular vulnerabilities in EOSIO smart contracts. Experiment results suggest that EOSAFE achieves promising results in detecting vulnerabilities, with an F1-measure of 98%. We have applied EOSAFE to all active 53,666 smart contracts in the ecosystem (as of November 15, 2019). Our results show that over 25% of the smart contracts are vulnerable. We further analyze possible exploitation attempts against these vulnerable smart contracts and identify 48 in-the-wild attacks (25 of them have been confirmed by DApp developers), resulting in financial loss of at least 1.7 million USD.

preprint2020arXiv

The Quenching-Activation Behavior of the Gradient Descent Dynamics for Two-layer Neural Network Models

A numerical and phenomenological study of the gradient descent (GD) algorithm for training two-layer neural network models is carried out for different parameter regimes when the target function can be accurately approximated by a relatively small number of neurons. It is found that for Xavier-like initialization, there are two distinctive phases in the dynamic behavior of GD in the under-parametrized regime: An early phase in which the GD dynamics follows closely that of the corresponding random feature model and the neurons are effectively quenched, followed by a late phase in which the neurons are divided into two groups: a group of a few "activated" neurons that dominate the dynamics and a group of background (or "quenched") neurons that support the continued activation and deactivation process. This neural network-like behavior is continued into the mildly over-parametrized regime, where it undergoes a transition to a random feature-like behavior. The quenching-activation process seems to provide a clear mechanism for "implicit regularization". This is qualitatively different from the dynamics associated with the "mean-field" scaling where all neurons participate equally and there does not appear to be qualitative changes when the network parameters are changed.

preprint2020arXiv

The Slow Deterioration of the Generalization Error of the Random Feature Model

The random feature model exhibits a kind of resonance behavior when the number of parameters is close to the training sample size. This behavior is characterized by the appearance of large generalization gap, and is due to the occurrence of very small eigenvalues for the associated Gram matrix. In this paper, we examine the dynamic behavior of the gradient descent algorithm in this regime. We show, both theoretically and experimentally, that there is a dynamic self-correction mechanism at work: The larger the eventual generalization gap, the slower it develops, both because of the small eigenvalues. This gives us ample time to stop the training process and obtain solutions with good generalization property.

preprint2020arXiv

Top quark as a probe of heavy Majorana neutrino at the LHC and future collider

Right-handed (RH) Majorana neutrinos play a crucial role in understanding the origin of neutrino mass, the nature of dark matter and the mechanism of matter-antimatter asymmetry. In this work, we investigate the observability of heavy Majorana neutrino through the top quark neutrinoless double beta decay process $t \to b \ell^+ \ell^+ j j$ at hadron colliders. By performing detector level simulation, we demonstrate that our method can give stronger limits on the light-heavy neutrino mixing parameters $|V_{eN, μN}|$ in the mass range of 15 GeV $< m_N <$ 80 GeV than other existing collider bounds.

preprint2019arXiv

Can we find steady-state solutions to multiscale rarefied gas flows within dozens of iterations?

One of the central problems in the study of rarefied gas dynamics is to find the steady-state solution of the Boltzmann equation quickly. When the Knudsen number is large, i.e. the system is highly rarefied, the conventional iteration scheme can lead to convergence within a few iterations. However, when the Knudsen number is small, i.e. the flow falls in the near-continuum regime, hundreds of thousands iterations are needed, and yet the "converged" solutions are prone to be contaminated by accumulated error and large numerical dissipation. Recently, based on the gas kinetic models, the implicit unified gas kinetic scheme (UGKS) and its variants have significantly reduced the iterations in the near-continuum flow regime, but still much higher than that of the highly rarefied gas flows. In this paper, we put forward a general synthetic iteration scheme (GSIS) to find the steady-state solutions of general rarefied gas flows within dozens of iterations at any Knudsen number. As the GSIS does not rely on the specific kinetic model/collision operator, it can be naturally extended to quickly find converged solutions for mixture flows and even flows involving chemical reactions. These two superior advantages are also expected to accelerate the slow convergence in simulation of near-continuum flows via the direct simulation Monte Carlo method and its low-variance version.

preprint2019arXiv

Testing electroweak SUSY for muon $g-2$ and dark matter at the LHC and beyond

Given that the LHC experiment has produced strong constraints on the colored supersymmetric particles (sparticles), testing the electroweak supersymmetry (EWSUSY) will be the next crucial task at the LHC. On the other hand, the light electroweakinos and sleptons in the EWSUSY can also contribute to the dark matter (DM) and low energy lepton observables. The precision measurements of them will provide the indirect evidence of SUSY. In this work, we confront the EWSUSY with the muon $g-2$ anomaly, the DM relic density, the direct detection limits and the latest LHC Run-2 data. We find that the sneutrino DM or the neutralino DM with sizable higgsino component has been excluded by the direct detections. Then two viable scenarios are pinned down: one has the light compressed bino and sleptons but heavy higgsinos, and the other has the light compressed bino, winos and sleptons. In the former case, the LSP and slepton masses have to be smaller than about 350 GeV. While in the latter case, the LSP and slepton masses have to be smaller than about 700 GeV and 800 GeV, respectively. From investigating the observability of these sparticles in both scenarios at future colliders, it turns out that the HE-LHC with a luminosity of 15 ab$^{-1}$ can exclude the whole BHL and most part of BWL scenarios at $2σ$ level. The precision measurement of the Higgs couplings at the lepton colliders could play a complementary role of probing the BWL scenario.

preprint2019arXiv

Unveiling CP property of top-Higgs coupling with graph neural networks at the LHC

The top-Higgs coupling plays an important role in particle physics and cosmology. The precision measurements of this coupling can provide an insight to new physics beyond the Standard Model. In this paper, we propose to use Message Passing Neural Network (MPNN) to reveal the CP nature of top-Higgs interaction through semi-leptonic channel $pp \to t(\to b\ell^-ν_\ell)\bar{t}(\to \bar{b}jj)h(\to b\bar{b})$. Using the test statistics constructed from the event classification probabilities given by the MPNN, we find that the pure CP-even and CP-odd components can be well distinguished at the LHC, with at most 300 fb$^{-1}$ experimental data.

preprint2018arXiv

Boundary Layer of Transport Equation with In-Flow Boundary

Consider the steady neutron transport equation in 2D convex domains with in-flow boundary condition. In this paper, we establish the diffusive limit while the boundary layers are present. Our contribution relies on a delicate decomposition of boundary data to separate the regular and singular boundary layers, novel weighted $W^{1,\infty}$ estimates for the Milne problem with geometric correction in convex domains, as well as an $L^{2m}-L^{\infty}$ framework which yields stronger remainder estimates.

preprint2018arXiv

High-Order Implicit Hybridizable Discontinuous Galerkin Method for the Boltzmann Equation

The high-order hybridizable discontinuous Galerkin (HDG) method combining with an implicit iterative scheme is used to find the steady-state solution of the Boltzmann equation with full collision integral on two-dimensional triangular meshes. The velocity distribution function and its trace are approximated in the piecewise polynomial space of degree up to 4. The fast spectral method (FSM) is incorporated into the DG discretization to evaluate the collision operator. Specific polynomial approximation is proposed for the collision term to reduce the computational cost. The proposed scheme is proved to be accurate and efficient.

preprint2016arXiv

750 GeV Composite Axion as the LHC Diphoton Resonance

We propose that the 750 GeV resonance, presumably observed in the early LHC Run 2 data, could be a heavy composite axion that results from condensation of a hypothetical quark in a high-colour representation of conventional QCD. The model, motivated by a recently proposed solution to the strong CP problem, is very economical and is essentially defined by the properties of the additional quark - its colour charge, hypercharge and mass. The axion mass and its coupling to two photons (via axial anomaly) can be computed in terms of these parameters. The axion is predominantly produced via photon fusion ($γγ\to {\cal A}$) which is followed by $ Z $ vector boson fusion and associated production at the LHC. We find that the total diphoton cross section of the axion can be fitted with the observed excess. Combining the requirement on the cross-section, such that it reproduces the diphoton excess events, with the bounds on the total width ($Γ_{tot} \leqslant 45$ GeV), we obtain the effective coupling in the range $1.6\times 10^{-4}$ GeV$^{-1}\gtrsim C_{\cal A} \gtrsim 6.5\times 10^{-5}$ GeV$^{-1}$. Within this window of allowed couplings the model favours a narrow width resonance and $ y_{Q}^2 \sim \mathcal{O}(10)$. In addition, we observe that the associated production $q\bar{q} \to {\cal A}γ\to γγγ$ can potentially produce a sizeable number of three photon events at future LHC and $ e^{+} e^{-} $ colliders. However, the rare decay $Z\to\mathcal{A}^*γ\to γγγ$ is found to be too small to be probed at the LHC.

preprint2016arXiv

750 GeV diphoton resonance in a top and bottom seesaw model

The top/bottom seesaw model, which extends the top seesaw in order to accomodate a 125 GeV Higgs boson, predicts vector-like top/bottom partners and these partners can be bounded to form several neutral and charged singlet composite scalars by some new strong dynamics. In this letter, we use such a singlet scalar to interpret the 750 GeV diphoton reseanance. This singlet scalar is dominantly produced through the gluon fusion process induced by the partners and its diphoton decay is induced by both the partners and the charged singlet scalars. We show that this scenario can readily account for the observed 750 GeV diphoton signal under the current LHC constraints. Further, this scenario predicts some other phenomenology, such as a strong correlation between the decays to $γγ$, $Z γ$ and $ZZ$, a three-photon singal from the associate production of a singlet scalar and a photon, as well as some signals from the partner cascade decays. These signals may jointly allow for a test of this framework in future LHC and ILC experiments.

preprint2016arXiv

750 GeV Diphoton Resonance, 125 GeV Higgs and Muon g-2 Anomaly in Deflected Anomaly Mediation SUSY Breaking Scenario

We propose to interpret the 750 GeV diphoton excess in deflected anomaly mediation supersymmetry breaking scenarios, which can naturally predict couplings between a singlet field and vector-like messengers. The CP-even scalar component (S) of the singlet field can serve as the 750 GeV resonance. The messenger scale, which is of order the gravitino scale, can be as light as F_ϕ\sim {\cal O}(10) TeV when the messenger species N_F and the deflection parameter d are moderately large. Such messengers can induce the large loop decay process S \to γγ. Our results show that such a scenario can successfully accommodate the 125 GeV Higgs boson, the 750 GeV diphoton excess and the muon g-2 without conflicting with the LHC constraints. We also comment on the possible explanations in the gauge mediation supersymmetry breaking scenario.

preprint2016arXiv

A comparative study of discrete velocity methods for rarefied gas flows

In the study of rarefied gas dynamics, the discrete velocity method (DVM) has been widely employed to solve the gas kinetic equations. Although various versions of DVM have been developed, their performance, in terms of accuracy and computational efficiency, is yet to be compreheively studied in the whole flow regime. Here, the traditional third-order time-implicit Godunov DVM (GDVM) and the recently developed discrete unified gas-kinetic scheme (DUGKS) are analysed in finding steady-state solutions of the force-driven Poiseuille and lid-driven cavity flows. With the molecular collision and free streaming being treated simultaneously, the DUGKS preserves the second-order accuracy in the spatial and temporal discretizations in all flow regimes. Towards the hydrodynamic flow regime, the DUGKS is not only faster than the GDVM when using the same spatial mesh, but also requires less spatial resolution than that of the GDVM to achieve the same numerical accuracy. From the slip to free molecular flow regimes, however, the DUGKS is slower than the GDVM, due to the complicated flux evaluation and the time step is less than the maximum effective time step of the GDVM. Therefore, the DUGKS is preferable for problems involving different flow regimes, particularly when the hydrodynamic flow regime is dominant. For highly rarefied gas flows, if the steady-state solution is concerned, the implicit DVM, which can boost the convergence significantly, is a better choice.

preprint2016arXiv

Asymptotic Analysis of Transport Equation in Annulus

We consider the diffusive limit of a steady neutron transport equation with one-speed velocity in a two-dimensional annulus. A classical theorem states that the solution can be approximated in $L^{\infty}$ by the leading order interior solution plus the corresponding Knudsen layers in the diffusive limit. In this paper, we construct a counterexample of this result via a different boundary layer expansion with geometric correction.

preprint2016arXiv

Closing up a light stop window in natural SUSY at LHC

Top squark (stop) plays a key role in the radiative stability of the Higgs boson mass in supersymmetry (SUSY). The LHC searches for stop have made a great progress and tightly constrained the stop mass during Run-1. In this work, we use the LHC Run-1 data to determine the lower mass limit of the right-handed stop in a natural SUSY scenario, where the higgsinos $\tildeχ^0_{1,2}$ and $\tildeχ^\pm_{1}$ are light and nearly degenerate. We find that the stop mass has been excluded up to 430 GeV for $m_{\tildeχ^0_1} \lesssim 250$ GeV and to 540 GeV for $m_{\tildeχ^0_1} \simeq 100$ GeV by the Run-1 SUSY searches for $2b+E^{miss}_T$ and $1\ell+jets+E^{miss}_T$, respectively. In a small strip of parameter space with $m_{\tildeχ^0_1} \gtrsim 190$ GeV, the stop mass can still be as light as 210 GeV and compatible with the Higgs mass measurement and the monojet bound. The 14 TeV LHC with a luminosity of 20 fb$^{-1}$ can further cover such a light stop window by monojet and $2b+E^{miss}_T$ searches and push the lower bound of the stop mass to 710 GeV. We also explore the potential to use the Higgs golden ratio, $D_{γγ}=σ(pp \to h \to γγ)/σ(pp \to h \to ZZ^* \to4\ell^\pm)$, as a complementary probe for the light and compressed stop. If this golden ratio can be measured at percent level at the high luminosity LHC (HL-LHC) or future $e^+e^-$ colliders, the light stop can be excluded for most of the currently allowed parameter region.

preprint2016arXiv

Diffusive Limit with Geometric Correction of Unsteady Neutron Transport Equation

We consider the diffusive limit of an unsteady neutron transport equation in a two-dimensional plate with one-speed velocity. We show the solution can be approximated by the sum of interior solution, initial layer, and boundary layer with geometric correction. Also, we construct a counterexample to the classical theory in \cite{Bensoussan.Lions.Papanicolaou1979} which states the behavior of solution near boundary can be described by the Knudsen layer derived from the Milne problem.

preprint2016arXiv

Electroweak Baryogenesis with Anomalous Higgs Couplings

We investigate feasibility of efficient baryogenesis at the electroweak scale within the effective field theory framework based on a non-linear realisation of the electroweak gauge symmetry. In this framework the LHC Higgs boson is described by a singlet scalar field, which, therefore, admits new interactions. Assuming that Higgs couplings with the eletroweak gauge bosons are as in the Standard Model, we demonstrate that the Higgs cubic coupling and the CP-violating Higgs-top quark anomalous couplings alone may drive the a strongly first-order phase transition. The distinguished feature of this transition is that the anomalous Higgs vacuum expectation value is generally non-zero in both phases. We identify a range of anomalous couplings, consistent with current experimental data, where sphaleron rates are sufficiently fast in the 'symmetric' phase and are suppressed in the 'broken' phase and demonstrate that the desired baryon asymmetry can indeed be generated in this framework. This range of the Higgs anomalous couplings can be further constrained from the LHC Run 2 data and be probed at high luminosity LHC and beyond.

preprint2016arXiv

Explaining 750 GeV diphoton excess from top/bottom partner cascade decay in two-Higgs-doublet model extension

In this paper, we interpret the 750 GeV diphoton excess in the Zee-Babu extension of the two-Higgs-doublet model by introducing a top partner ($T$)/bottom partner ($B$). In the alignment limit, the 750 GeV resonance is identified as the heavy CP-even Higgs boson ($H$), which can be sizably produced via the QCD process $pp \to T\bar{T}$ or $pp \to B\bar{B}$ followed by the decay $T\to Ht$ or $B \to Hb$. The diphoton decay rate of $H$ is greatly enhanced by the charged singlet scalars predicted in the Zee-Babu extension and the total width of $H$ can be as large as 7 GeV. Under the current LHC constraints, we scan the parameter space and find that such an extension can account for the observed diphoton excess.

preprint2016arXiv

Finding the magnetic center of sextupole using vibrating wire technique

In order to obtain stable and high-quality synchrotron radiation photon, the magnets in the storage ring of High Energy Photon Source(HEPS) need to have a stable support and precise positioning. The vibrating wire technique can be used to pre-align the quadrupoles and sextupoles on one girder with high precision to meet the extremely low emittance requirement of HEPS. This thesis introduces the measurement of magnetic center of sextupole using vibrating wire. According to the measurement results, the magnetic field distribution is consistent with theoretical expectation. And vibrating wire has achieved the purpose of measuring the magnetic center and has reached a certain precision.

preprint2016arXiv

Lattice Boltzmann modelling of intrinsic permeability

Lattice Boltzmann method (LBM) has been applied to predict flow properties of porous media including intrinsic permeability, where it is implicitly assumed that the LBM is equivalent to the incompressible (or near incompressible) Navier-Stokes equation. However, in LBM simulations, high-order moments, which are completely neglected in the Navier-Stokes equation, are still available through particle distribution functions. To ensure that the LBM simulation is correctly working at the Navier-Stokes hydrodynamic level, the high-order moments have to be negligible. This requires that the Knudsen number (Kn) is small so that rarefaction effect can be ignored. In this technical note, we elaborate this issue in LBM modelling of porous media flows, which is particularly important for gas flows in ultra-tight media.

preprint2016arXiv

Regularity of Milne Problem with Geometric Correction in 3D

Consider the Milne problem with geometric correction in a 3D convex domain. Via bootstrapping arguments, we establish $W^{1,\infty}$ regularity for its solutions. Combined with a uniform $L^6$ estimate, such regularity leads to the validity of diffusive expansion for the neutron transport equation with diffusive boundary conditions.

preprint2016arXiv

Single top partner production in the Higgs to diphoton channel in the Littlest Higgs Model with $T$-parity

The top partner as a hallmark of the Littlest Higgs model with $T$-parity (LHT model) has been extensively searched for during the Large Hadron Collider (LHC) Run-1. With the increasing mass limits on the top partner, the single production of the top partner will be dominant over the pair production. Under the constraints from the Higgs data, the electroweak precision observables and $R_b$, we find that the mass of $T$-even top partner ($T_+$) has to be heavier than 730 GeV. Then, we investigate the observability of the single $T$-even top partner production through the process $pp \to T_+ j$ with the sequent decay $T_+ \to th$ in the di-photon channel in the LHT model at the LHC. We find that the mass of $T_+$ can be excluded up to 800 GeV at $2σ$ level at 14 TeV LHC with the integrated luminosity ${\cal L}=3$ ab$^{-1}$.

preprint2016arXiv

Single top squark production as a probe of natural supersymmetry at the LHC

Light top squarks (stops) and light higgsinos are the key features of natural SUSY, where the higgsinos $\tildeχ^{\pm}_{1}$ and $\tildeχ^0_{1,2}$ are nearly degenerate and act as the missing transverse energy ($E^{miss}_T$) at the LHC. Besides the strong production, the stop can be produced via the electroweak interaction. The determination of the electroweak properties of the stop is an essential task for the LHC and future colliders. So in this paper, we investigate the single stop ($\tilde{t}_1$) production $pp \to \tilde{t}_1+ E^{miss}_T$ in the natural SUSY at the LHC, which gives the monotop signature $t+ E^{miss}_T$ from $\tilde{t}_1\to t \tildeχ^0_{1,2}$ or the monobottom signature $b+ E^{miss}_T$ from $\tilde{t}_1 \to b \tildeχ^+_{1}$. We perform Monte Carlo simulations for these signatures and obtain the results: (1) The signal $b+ E^{miss}_T$ has a better sensitivity than $t+ E^{miss}_T$ for probing natural SUSY; (2) The parameter region with a higgsino mass 100 GeV$\lesssim μ\lesssim$ 225 GeV and stop mass $m_{\tilde{t}_1} \lesssim$ 620 GeV, can be probed through the single stop production with $S/\sqrt{B} > 3$ and $4\% \lesssim S/B \lesssim19\%$ at 14 TeV HL-LHC with an integrated luminosity of 3000 fb$^{-1}$.

preprint2016arXiv

Weak positivity for Hodge modules

We prove the weak positivity of the kernels of Kodaira-Spencer- type maps for pure Hodge module extensions of generically defined variations of Hodge structure.

preprint2015arXiv

ATLAS Z-peaked excess in MSSM with a light sbottom or stop

We attempt to explain the recent ATLAS 3$σ$ excess of dilepton events with an invariant mass near $Z$ peak through the gluino-mediated sbottom production in a simplified scenario inspired by the Minimal Supersymmetric Standard Model (MSSM). The additional $Z$ bosons can be produced through the cascade decay chain $\tilde{g} \to b \tilde{b}^{\dagger}_{1} \to b \bar{b} \tildeχ^{0}_{2,3} \to b \bar{b} \tildeχ^{0}_1 Z$, in which $\tilde b_1$ is the right-handed sbottom, $\tildeχ_1^0$ is the bino-like lightest supersymmetric particle (LSP) and $\tildeχ_{2,3}^0$ are two nearly degenerate higgsino-like next-to LSPs (NLSPs). Taking into account the constraint from the LHC search for the gluino-mediated sbottom production in the final states with the missing transverse energy and at least three $b$-jets, we find that the ATLAS on-$Z$ excess can only be marginally explained at $2σ$ level. We also note that within the scenario where the gluino predominantly decays to the right-handed stop instead of the sbottom, the excess can hardly be explained, since the $Z$-boson is not produced in the stop dominant decay channel, $\tilde{t}_1 \to b \tildeχ^+_1$.

preprint2015arXiv

Enhancing $thj$ Production from Top-Higgs FCNC Couplings

In this paper, we study the single top and Higgs associated production $pp \to thj$ in the presence of top-Higgs FCNC couplings($κ_{tqh}, q=u,c$) at the LHC. Under the current constraints, we find that the cross section of $pp \to thj$ can be sizably enhanced in comparison with the SM predictions at 8 and 14 TeV LHC. We also find that the full cross section of $pp \to thj$ with $κ_{tch}$ is larger than the resonant cross section of $pp \to t\bar{t} \to thj$ by a factor 1.16 at 8 TeV LHC and 1.12 at 14 TeV LHC, respectively. We further explore the observability of top-Higgs FCNC couplings through $pp \to t(\to b\ell^{+} ν_{\ell}) h( \to γγ) j$ and find that the branching ratios $Br(t\to qh)$, $Br(t \to uh)$ and $Br(t \to ch)$ can be respectively probed to $0.12\%,~0.23\%$ and $~0.26\%$ at $3σ$ sensitivity at 14 TeV LHC with ${\cal L} =3000$ fb$^{-1}$.

preprint2015arXiv

hhh Coupling in SUSY models after LHC run I

We examine the Higgs triple coupling in MSSM and NMSSM under current constraints, which include the LHC measurements, Higgs data, B physics, electroweak precision observables, relic density and so on. The ratio $λ^{\rm MSSM}_{hhh}/λ^{\rm SM}_{hhh}$ is above 0.97, due to the highly constrained parameter space. While the ratio $λ^{\rm NMSSM}_{hhh}/λ^{\rm SM}_{hhh}$ can reach 0.1 under current constraints. The precise measurement in future collider will give a tighter constraint to the Higgs triple coupling in MSSM and NMSSM.

preprint2015arXiv

Higgs self-coupling in the MSSM and NMSSM after the LHC Run 1

Measuring the Higgs self-coupling is one of the crucial physics goals at the LHC Run-2 and other future colliders. In this work, we attempt to figure out the size of SUSY effects on the trilinear self-coupling of the 125 GeV Higgs boson in the MSSM and NMSSM after the LHC Run-1. Taking account of current experimental constraints, such as the Higgs data, flavor constraints, electroweak precision observables and dark matter detections, we obtain the observations: (1) In the MSSM, the ratio of $λ^{MSSM}_{3h}/λ^{SM}_{3h}$ has been tightly constrained by the LHC data, which can be only slightly smaller than 1 and minimally reach 97\%; (2) In the NMSSM with $λ<0.7$, a sizable reduction of $λ^{NMSSM}_{3h_2}/λ^{SM}_{3h_2}$ can occur and minimally reach 10\% when the lightest CP-even Higgs boson mass $m_{h_1}$ is close to the SM-like Higgs boson $m_{h_2}$ due to the large mixing angle between the singlet and doublet Higgs bosons; (3) In the NMSSM with $λ>0.7$, a large enhancement or reduction $-1.1<λ^{NMSSM}_{3h_1}/λ^{SM}_{3h_1}<2$ can occur, which is accompanied by a sizable change of $h_1τ^+τ^-$ coupling. The future colliders, such as the HL-LHC and ILC, will have the capacity to test these large deviations in the NMSSM.

preprint2015arXiv

Hydrodynamic Limit with Geometric Correction of Stationary Boltzmann Equation

We consider the hydrodynamic limit of a stationary Boltzmann equation in a unit plate with in-flow boundary. We prove the solution can be approximated in $L^{\infty}$ by the sum of interior solution which satisfies steady incompressible Navier-Stokes-Fourier system, and boundary layer with geometric correction. Also, we construct a counterexample to the classical theory which states the behavior of solution near boundary can be described by the Knudsen layer derived from the Milne problem.

preprint2015arXiv

Smoothed Dissipative Particle Dynamics model for mesoscopic multiphase flows in the presence of thermal fluctuations

Thermal fluctuations cause perturbations of fluid-fluid interfaces and highly nonlinear hydrodynamics in multiphase flows. In this work, we develop a novel multiphase smoothed dissipative particle dynamics model. This model accounts for both bulk hydrodynamics and interfacial fluctuations. Interfacial surface tension is modeled by imposing a pairwise force between SDPD particles. We show that the relationship between the model parameters and surface tension, previously derived under the assumption of zero thermal fluctuation, is accurate for fluid systems at low temperature but overestimates the surface tension for intermediate and large thermal fluctuations. To analyze the effect of thermal fluctuations on surface tension, we construct a coarse-grained Euler lattice model based on the mean field theory and derive a semi-analytical formula to directly relate the surface tension to model parameters for a wide range of temperatures and model resolutions. We demonstrate that the present method correctly models the dynamic processes, such as bubble coalescence and capillary spectra across the interface.

preprint2015arXiv

Vanishing and injectivity theorems for Hodge modules

We prove a surjectivity theorem for the Deligne canonical extension of a polarizable variation of Hodge structure with quasi-unipotent monodromy at infinity along the lines of Esnault-Viehweg. We deduce from it several injectivity theorems and vanishing theorems for pure Hodge modules. We also give an inductive proof of Kawamata-Viehweg vanishing for the lowest graded piece of the Hodge filtration of a pure Hodge module using mixed Hodge modules of nearby cycles.

preprint2014arXiv

A light SUSY dark matter after CDMS-II, LUX and LHC Higgs data

In SUSY, a light dark matter is usually accompanied by light scalars to achieve the correct relic density, which opens new decay channels of the SM like Higgs boson. Under current experimental constraints including the latest LHC Higgs data and the dark matter relic density, we examine the status of a light neutralino dark matter in the framework of NMSSM and confront it with the direct detection results of CoGeNT, CDMS-II and LUX. We have the following observations: (i) A dark matter as light as 8 GeV is still allowed and its scattering cross section off the nucleon can be large enough to explain the CoGeNT/CDMS-II favored region; (ii) The LUX data can exclude a sizable part of the allowed parameter space, but still leaves a light dark matter viable; (iii) The SM-like Higgs boson can decay into the light dark matter pair with an invisible branching ratio reaching 30% under the current LHC Higgs data, which may be tested at the 14 TeV LHC experiment.

preprint2014arXiv

A new approach for detecting compressed bino/wino at the LHC

In some supersymmetric models like split supersymmetry or models with non-universal gaugino mass, bino (LSP) and winos (NLSP) may have rather small mass splitting in order to provide the correct dark matter relic density through bino/wino co-annihilation. Such a scenario with the compressed bino/wino is difficult to explore at the LHC. In this work we propose to probe this scenario from $pp \to j \tildeχ^0_2 \tildeχ^\pm_1$ followed by $\tildeχ^0_2 \to γ\tildeχ^0_1$ and $\tildeχ^\pm_1 \to W^{*}\tildeχ^0_1\to \ell^\pm ν\tildeχ^0_1$ (this method is also applicable to the compressed bino/higgsino scenario). Through a detailed Monte Carlo simulation for both the signal and the backgrounds, we find that for a mass splitting $ΔM \sim 5-15$ GeV between bino (LSP) and wino (NLSP), the 14 TeV LHC with luminosity of 500$fb^{-1}$ can probe the wino up to 150 GeV (the sensitivity can reach $5σ$ for $ΔM = 5$ GeV and $2σ$ for $ΔM = 15$ GeV). We also investigate the dark matter detection sensitivities for this scenario and find that the planned XENON-1T(2017) cannot fully cover the parameter space with wino below 150 GeV allowed by relic density and the LUX limits.

preprint2014arXiv

Anomalous Top-Higgs Couplings and Top Polarisation in Single Top and Higgs Associated Production at the LHC

In this paper, we put constraints on anomalous $\mathcal{CP}$-violating top-Higgs couplings using the currently available Higgs data and explore the prospect of measuring these couplings at 240 GeV TLEP. We find that the $\mathcal{CP}$-violating phase $ξ$ is currently limited in the range $|ξ|< 0.6π$ at 95\% C.L. and may be further constrained to $|ξ| <0.07π$ at TLEP. Under this consideration, we further investigate the observability of the scalar ($ξ=0$), pseudoscalar ($ξ=0.5π$) and mixed ($ξ=0.25π$) top-Higgs interactions through the channel $pp \to t(\to \ell^+ν_\ell b)h(\to b\overline{b})j$. We find that it is most promising to observe pure pseudoscalar interactions with $y_t=y_t^{SM}$, although this will be challenging due to a low signal to background ratio. We also find that the anomalous top-Higgs couplings can lead to sizeable differences in lepton forward-backward asymmetries and can be distinguished by measuring the lepton angular distributions from polarised top quarks at 14 TeV LHC.

preprint2014arXiv

Constraining Top partner and Naturalness at the LHC and TLEP

We investigate indirect constraints on the top partner within the minimal fermionic top partner model. By performing a global fit of the latest Higgs data, $B_s \to μ^+μ^-$ measurements and the electroweak precision observables we find that the top partner with the mass up to 830 GeV is excluded at $2σ$ level. Our bound on the top partner mass is much stronger than the bounds obtained from the direct searches at the LHC. Under the current constraints the fine-tuning measure is less than 9% and the branching ratio of $T \to tZ$ is bounded between 14% and 25%. We also find that precise measurements of Higgs couplings at 240 GeV TLEP will constrain the top partner mass in multi-TeV region.

preprint2014arXiv

Full one-loop electroweak corrections to $e^+e^- \to ZHγ$ at a Higgs factory

Motivated by the future precision test of the Higgs boson at an $e^+e^-$ Higgs factory, we calculate the production $e^+e^- \to ZHγ$ in the Standard Model with complete next-to-leading order electroweak corrections. We find that for $\sqrt{s}=240$ (350) GeV the cross section of this production is sizably reduced by the electroweak corrections, which is $1.03$ (5.32) fb at leading order and 0.72 (4.79) fb at next-to-leading order. The transverse momentum distribution of the photon in the final states is also presented.

preprint2014arXiv

Geometric Correction for Diffusive Expansion of Steady Neutron Transport Equation

We revisit the diffusive limit of a steady neutron transport equation in a $2$-D unit disk with one-speed velocity. We show the classical result in [4] with Milne expansion is incorrect in $L^{\infty}$ and we give the right answer in studying the $ε$-Milne expansion with geometric correction.

preprint2014arXiv

Higgs pair production with SUSY QCD correction: revisited under current experimental constraints

We consider the current experimental constraints on the parameter space of the MSSM and NMSSM. Then in the allowed parameter space we examine the Higgs pair production at the 14 TeV LHC via $b\bar{b}\to hh$ ($h$ is the 125 GeV SM-like Higg boson) with one-loop SUSY QCD correction and compare it with the production via $gg\to hh$. We obtain the following observations: (i) For the MSSM the production rate of $b\bar{b} \to hh$ can reach 50 fb and thus can be competitive with $gg \to hh$, while for the NMSSM $b\bar{b} \to hh$ has a much smaller rate than $gg \to hh$ due to the suppression of the $hb\bar{b}$ coupling; (ii) The SUSY-QCD correction to $b\bar{b} \to hh$ is sizable, which can reach $45\%$ for the MSSM and $15\%$ for the NMSSM within the $1σ$ region of the Higgs data; (iii) In the heavy SUSY limit (all soft mass parameters become heavy), the SUSY effects decouple rather slowly from the Higgs pair production (especially the $gg\to hh$ process), which, for $M_{\rm SUSY}=5$ TeV and $m_A<1$ TeV, can enhance the production rate by a factor of 1.5 and 1.3 for the MSSM and NMSSM, respectively. So, the Higgs pair production may be helpful for unraveling the effects of heavy SUSY.

preprint2014arXiv

Local Wellposedness of Viscous Surface Wave without Surface Tension

We consider an incompressible viscous flow without surface tension in a finite- depth domain of three dimension, with free top boundary. This system is governed by a Naiver-Stokes equation in a moving domain and a transport equation for the top boundary. Traditionally, we consider this problem in Lagrangian coordinate and perturbed linear form. In [1], I. Tice and Y. Guo introduced a new framework using geometric structure in Eulerian coordinate to study both local and global wellposedness of this system. Following this path, we extend their result in local wellposedness from small data case to arbitrary data case. Other than the geometric energy estimate and time-dependent Galerkin method introduced in [1], we utilize a few new techniques: (1) using parameterized Poisson integral to construct a nontrivial transform between fixed domain and moving domain; (2) using bootstrapping argument to prove a comparison result for steady Navier-Stokes equation for arbitrary data of free surface.

preprint2014arXiv

Previous R&D of vibrating wire alignment technique for HEPS

The alignment tolerance of multipoles on a girder is better than 30um in the storage ring of High Energy Photon Source HEPS which will be the next project at IHEP Institute of High Energy Physics. This is difficult to meet the precision only using the traditional optical survey method. In order to achieve this goal, vibrating wire alignment technique with high precision and sensitivity is considered to be used in this project. This paper presents some previous research works about theory, scheme design and achievements.

preprint2014arXiv

Probing Light Higgsinos in Natural SUSY from Monojet Signals at the LHC

We investigate a strategy to search for light, nearly degenerate higgsinos within the natural MSSM at the LHC. We demonstrate that the higgsino mass range $μ$ in $100-150$ GeV, which is preferred by the naturalness, can be probed at $2σ$ significance through the monojet search at 14 TeV HL-LHC with 3000 fb$^{-1}$ luminosity. The proposed method can also probe certain region in the parameter space for the lightest neutralino with a high higgsino purity, that cannot be reached by planned direct detection experiments at XENON-1T(2017).

preprint2014arXiv

Revisiting Associated Production of 125 GeV Higgs Boson with a Photon at a Higgs Factory

Considering the constraints from the flavor physics, precision electroweak measurements, Higgs data and dark matter detections, we scan over the parameter space of the MSSM and calculate the cross section of $e^+e^- \to h γ$ in the allowed parameter space. Since the loop-induced gauge couplings $hγγ$ and $hZγ$ can simultaneously contribute to the process $e^+e^- \to h γ$, we find the cross section can be sizably enhanced by a light stau, maximally 1.47(1.38) times larger than the SM prediction at $\sqrt{s}=240(350)$ GeV. So with the high luminosity, the measurement of $e^+e^- \to h γ$ may be used to test the anomalous gauge couplings $hγγ$ and $hZγ$ in the MSSM at a Higgs Factory.

preprint2014arXiv

SUSY effects in Higgs productions at high energy $e^+e^-$ colliders

Considering the constraints from collider experiments and dark matter detections, we investigate the SUSY effects in the Higgs productions $e^+e^- \to Zh$ at an $e^+e^-$ collider with a center-of-mass energy above 240 GeV and $γγ\to h \to b\bar{b}$ at a photon collider with a center-of-mass energy above 125 GeV. In the parameter space allowed by current experiments, we find that the SUSY corrections to $e^+e^- \to Zh$ can reach a few percent and the production rate of $γγ\to h \to b\bar{b}$ can be enhanced by a factor of 1.2 over the SM prediction. We also calculate the exotic Higgs productions $e^+e^-\to Zh_1$ and $e^+e^-\rightarrow A_1h$ in the next-to-minimal supersymmetric model (NMSSM) ($h$ is the SM-like Higgs, $h_1$ and $A_1$ are respectively the CP-even and CP-odd singlet-dominant Higgs bosons which can be much lighter than $h$). We find that at a 250 GeV $e^+e^-$ collider the production rates of $e^+e^-\rightarrow Zh_1$ and $e^+e^-\to A_1h$ can reach 60 fb and 0.1 fb, respectively.

preprint2014arXiv

SUSY induced top quark FCNC decay $t \to c h$ after Run I of LHC

In light of the Higgs discovery and the nonobservation of sparticles at the LHC, we revisit the SUSY induced top quark flavor changing decay into the Higgs boson. We perform a scan over the relevant SUSY parameter space by considering the constraints from the Higgs mass measurement, the LHC search for SUSY, the vacuum stability, the precision electro-weak observables as well as $B \to X_s γ$. We have the following observations: (1) In the MSSM, the branching ratio of $ t \to c h$ can only reach $3.0\times10^{-6}$, which is about one order smaller than previous results obtained before the advent of the LHC. Among the considered constraints, the Higgs mass and the LHC search for sparticles are found to play an important role in limiting the prediction. (2) In the singlet extension of the MSSM, since the squark sector is less constrained by the Higgs mass, the branching ratio of $t \to c h$ can reach the order of $10^{-5}$ in the allowed parameter space. (3) The chiral-conserving mixings $δ_{LL}$ and $δ_{RR}$ may have remanent effects on $t \to c h$ in heavy SUSY limit. In the MSSM with squarks and gluino above 3 TeV and meanwhile the CP-odd Higgs boson mass around 1 TeV, the branching ratio of $t\to c h$ can still reach the order of $10^{-8}$ under the constraints.

preprint2013arXiv

Complete one-loop effects of SUSY QCD in $b\bar{b}h$ production at the LHC under current experimental constraints

Inspired by the recent LHC Higgs data and null search results of supersymmetry (SUSY), we scan the parameter space of the Minimal Supersymmetric Standard Model (MSSM) with relatively heavy sparticles (1-3 TeV). Then in the parameter space allowed by current collider experiments and dark matter detections, we calculate the complete one-loop SUSY QCD corrections to $pp\to b\bar{b}h$ at the LHC with $\sqrt{s}=14$ TeV and obtain the following observations: (i) For the large values of $\tanβ$ and low values of $m_{A}$, the SUSY QCD effects can be quite large, which, however, have been excluded by the latest results of LHC search for $H/A \to τ^{+}τ^{-}$; (ii) For modest values of $\tanβ$ and $m_{A}$ which so far survived all experimental constraints, the SUSY QCD corrections can maximally reach about -9%.

preprint2013arXiv

Current experimental bounds on stop mass in natural SUSY

Motivated by the recent progress of direct search for the productions of stop pair and sbottom pair at the LHC, we examine the constraints of the search results on the stop ($\tilde{t}_1$) mass in natural SUSY. We first scan the parameter space of natural SUSY in the framework of MSSM, considering the constraints from the Higgs mass, B-physics and electroweak precision measurements. Then in the allowed parameter space we perform a Monte Carlo simulation for stop pair production followed by $\tilde{t}_{1} \to t \tildeχ_{1}^{0}$ or $\tilde{t}_{1} \to b \tildeχ_{1}^{+}$ and sbottom pair production followed by $\tilde{b}_{1} \to b \tildeχ_{1}^{0}$ or $\tilde{b}_{1} \to t \tildeχ_{1}^{-}$. Using the combined results of ATLAS with 20.1 fb$^{-1}$ from the search of $\ell+{\rm jets}+\slashed E_{T}$, hadronic $t\bar{t}+\slashed E_{T}$ and $2b+\slashed E_{T}$, we find that a stop lighter than 600 GeV can be excluded at 95% CL in this scenario.

preprint2013arXiv

MIGRAINE: MRI Graph Reliability Analysis and Inference for Connectomics

Currently, connectomes (e.g., functional or structural brain graphs) can be estimated in humans at $\approx 1~mm^3$ scale using a combination of diffusion weighted magnetic resonance imaging, functional magnetic resonance imaging and structural magnetic resonance imaging scans. This manuscript summarizes a novel, scalable implementation of open-source algorithms to rapidly estimate magnetic resonance connectomes, using both anatomical regions of interest (ROIs) and voxel-size vertices. To assess the reliability of our pipeline, we develop a novel nonparametric non-Euclidean reliability metric. Here we provide an overview of the methods used, demonstrate our implementation, and discuss available user extensions. We conclude with results showing the efficacy and reliability of the pipeline over previous state-of-the-art.

preprint2013arXiv

The Z+photon and diphoton decays of the Higgs boson as a joint probe of low energy SUSY models at LHC

In light of recent remarkable progress in Higgs search at the LHC, we study the rare decay process $h \to Zγ$ and show its correlation with the decay $h \to γγ$ in low energy SUSY models such as CMSSM, MSSM, NMSSM and nMSSM. Under various experimental constraints, we scan the parameter space of each model, and present in the allowed parameter space the SUSY predictions on the $Zγ$ and $γγ$ signal rates in the Higgs production at the LHC and future e+e- linear colliders. We have following observations: (i) Compared with the SM prediction, the $Zγ$ and $γγ$ signal rates in the CMSSM are both slightly suppressed; (ii) In the MSSM, both the $Zγ$ and $γγ$ rates can be either enhanced or suppressed, and in optimal case, the enhancement factors can reach 1.2 and 2 respectively; (iii) In the NMSSM, the $Zγ$ and $γγ$ signal rates normalized by their SM predictions are strongly correlated, and vary from 0.2 to 2; (iv) In the nMSSM, the $Zγ$ and $γγ$ rates are greatly reduced. Since the correlation behavior between the Zγsignal and the γγsignal is so model-dependent, it may be used to distinguish the models in future experiments.

preprint2013arXiv

Two-Higgs-doublet model with a color-triplet scalar: a joint explanation for top quark forward-backward asymmetry and Higgs decay to diphoton

The excess of top quark forward-backward asymmetry ($A^t_{FB}$) reported by the Tevatron and the enhancement of the Higgs decay to diphoton observed by the LHC may point to a same origin of new physics. In this note we examined such anomalies in the two-Higgs-doublet model with a color-triplet scalar. We found that under current experimental constraints this model can simultaneously explain both anomalies at $1σ$ level. Also, we examined the Higgs decay $h\to Zγ$ and displayed its correlation with $h\to γγ$. We found that unlike other models, this model predicts a special correlation between $h\to Zγ$ and $h\to γγ$, i.e., the $Zγ$ rate is highly suppressed while the $γγ$ rate is enhanced. This behavior may help to distinguish this model in the future high luminosity run of the LHC.

preprint2012arXiv

Pattern Detection with Rare Item-set Mining

The discovery of new and interesting patterns in large datasets, known as data mining, draws more and more interest as the quantities of available data are exploding. Data mining techniques may be applied to different domains and fields such as computer science, health sector, insurances, homeland security, banking and finance, etc. In this paper we are interested by the discovery of a specific category of patterns, known as rare and non-present patterns. We present a novel approach towards the discovery of non-present patterns using rare item-set mining.

preprint2012arXiv

Probing natural SUSY from stop pair production at the LHC

We consider the natural supersymmetry scenario in the framework of the R-parity conserving minimal supersymmetric standard model (called natural MSSM) and examine the observability of stop pair production at the LHC. We first scan the parameters of this scenario under various experimental constraints, including the SM-like Higgs boson mass, the indirect limits from precision electroweak data and B-decays. Then in the allowed parameter space we study the stop pair production at the LHC followed by the stop decay into a top quark plus a lightest neutralino or into a bottom quark plus a chargino. From detailed Monte Carlo simulations of the signals and backgrounds, we find the two decay modes are complementary to each other in probing the stop pair production, and the LHC with $\sqrt{s}= 14$ TeV and 100 $fb^{-1}$ luminosity is capable of discovering the stop predicted in natural MSSM up to 450 GeV. If no excess events were observed at the LHC, the 95% C.L. exclusion limits of the stop masses can reach around 537 GeV.

preprint2012arXiv

Probing topcolor-assisted technicolor from top charge asymmetry and triple-top production at the LHC

In a topcolor-assisted technicolor model (TC2) with large FCNC top quark couplings, we study its correlated contributions to the top quark forward-backward asymmetry ($A_{FB}$) at the Tevatron, the top charge asymmetry ($A_{C}$) and the triple-top production at the LHC. Under current constraints on the top quark from the LHC and Tevatron(such as the total and differential production rates), we scan the parameter space of such a TC2 model. We find that in the allowed parameter space the TC2 model can explain the Tevatron measured $A_{FB}$ at $2σ$ level, but meanwhile significantly enhance $A_{C}$ at the LHC. Such enhanced $A_{C}$, albeit currently allowed by the LHC measurement at $2σ$ level, will serve as a test of TC2 with the improvement of measurement precision at the LHC. Then with all the constraints (including the requirement to explain $A_{FB}$ at $2σ$ level and satisfying the current LHC measurement of $A_{C}$ at $2σ$ level), we find that the TC2 model can induce sizable triple-top production at the 14 TeV LHC (the production rate can maximally reach 16 pb). Due to the low SM backgrounds, the triple-top production can also be a good probe for TC2 model, complementary to $A_{C}$.

preprint2012arXiv

Testing new physics models by top charge asymmetry and polarization at the LHC

As a top quark factory, the LHC can test the new physics models used to explain the top quark forward-backward asymmetry A^t_FB measured at the Tevatron. In this work we perform a comparative study for two such models: the W'-model and the color triplet diquark (ϕ) model. Requiring these models to explain A^t_FB and also satisfy the top pair production rate measured at the Tevatron, we examine their contributions to the LHC observables such as the polarization and charge asymmetry in top quark and W' (or ϕ) productions. We find that these observables can be enhanced to the observable level and the current LHC measurement on the top charge asymmetry can already tightly constrain the W'-model. We also find that each observable shows different characteristics in different models, which can be utilized to discriminate the models.

preprint2012arXiv

Top quark forward-backward asymmetry and charge asymmetry in left-right twin Higgs model

In order to explain the Tevatron anomaly of the top quark forward-backward asymmetry $A_{FB}^t$ in the left-right twin Higgs model, we choose to give up the lightest neutral particle of $\hat{h}$ field as a stable dark matter candidate. Then a new Yukawa interaction for $\hat{h}$ is allowed, which can be free from the constraint of same-sign top pair production and contribute sizably to $A_{FB}^t$. Considering the constraints from the production rates of the top pair ($t\bar t$), the top decay rates and $t\bar{t}$ invariant mass distribution, we find that this model with such new Yukawa interaction can explain $A_{FB}^t$ measured at the Tevatron while satisfying the charge asymmetry $A_{C}^t$ measured at the LHC.Moreover, this model predicts a strongly correlation between $A_{C}^t$ at the LHC and $A_{FB}^t$ at the Tevatron, i.e., $A_{C}^t$ increases as $A_{FB}^t$ increases.

preprint2012arXiv

Wellposedness and Decaying Property of Viscous Surface Wave

In this paper, we consider an incompressible viscous flow without surface tension in a finite-depth domain of three dimensions, with free top boundary and fixed bottom boundary. This system is governed by a Naiver-Stokes equation in above moving domain and a transport equation for the top boundary. Traditionally, we consider this problem in Lagrangian coordinates with perturbed linear form. In the series of papers [1], [2] and [3], I. Tice and Y. Guo introduced a new framework using geometric structure in Eulerian coordinates to study both local and global wellposedness of this system. Following this path, we extend their result in local wellposedness from small data case to arbitrary data case. Also, we give a simpler proof for global wellposedness in infinite domain. Other than the geometric energy estimates, time-dependent Galerkin method, and interpolation estimate with Riesz potential and minimum count, which are introduced in these papers, we utilize three new techniques: (1) using ε-Poisson integral to construct a diffeomorphism between fixed domain and moving domain; (2) using bootstrapping argument to prove a comparison result for steady Navier-Stokes equation for arbitrary data of free surface; (3) redefining the energy and dissipation to replace the original complicated bootstrapping argument to show interpolation estimate.

preprint2011arXiv

New physics effects on top quark spin correlation and polarization at the LHC: a comparative study in different models

Extensions of the Standard Model often predict new chiral interactions for top quark, which will contribute to top quark spin correlation and polarization in $t\bar{t}$ production at the LHC. In this work, under the constraints from the current Tevatron measurements, a comparative study of the spin correlation and polarization is performed in three new physics models: the minimal supersymmetric model without R-parity (RPV-MSSM), the third-generation enhanced left-right model and the axigluon model. We find that the polarization asymmetry may be enhanced to the accessible level in all these models while the correction to the spin correlation may be detectable in the axigluon model and the RPV-MSSM with $λ"$ couplings.

preprint2011arXiv

Top quark forward-backward asymmetry, FCNC decays and like-sign pair production as a joint probe of new physics

The anomaly of the top quark forward-backward asymmetry $A^t_{FB}$ observed at the Tevatron can be explained by the t-channel exchange of a neutral gauge boson (Z') which has sizable flavor changing coupling for top and up quarks. This gauge boson can also induce the top quark flavor-changing neutral-current (FCNC) decays and the like-sign top pair production at the LHC. In this work we focus on two models which predict such a Z', namely the left-right model and the U(1)_X model, to investigate the correlated effects on $A^t_{FB}$, the FCNC decays $t -> u V$ ($V=g,Z,γ$) and the like-sign top pair production at the LHC. We also pay special attention to the most recently measured $A^t_{FB}$ in the large top pair invariant mass region. We find that under the current experimental constraints both models can alleviate the deviation of $A^t_{FB}$ and, meanwhile, enhance the like-sign top pair production to the detectable level of the LHC. We also find that the two models give different predictions for the observables and their correlations, and thus they may even be distinguished by jointly studying these top quark observables.

preprint2010arXiv

Exact exact solutions of the Gross-Pitaevskii equation for stable vortex modes

We construct exact solutions of the Gross-Pitaevskii equation for solitary vortices, and approximate ones for fundamental solitons, in 2D models of Bose-Einstein condensates with a spatially modulated nonlinearity of either sign and a harmonic trapping potential. The number of vortex-soliton (VS) modes is determined by the discrete energy spectrum of a related linear Schrödinger equation. The VS families in the system with the attractive and repulsive nonlinearity are mutually complementary. \emph{% Stable} VSs with vorticity $S\geq 2$ and those corresponding to higher-order radial states are reported for the first time, in the case of the attraction and repulsion, respectively.

preprint2010arXiv

Lepton flavor-changing processes in R-parity violating MSSM: Z->l_i l_j and γγ->l_i l_j under new bounds from l_i->l_j γ

We examine the lepton flavor-changing processes in R-parity violating MSSM. First, we update the constraints on the relevant R-violating couplings by using the latest data on the rare decays l_i->l_j γ. We find that the updated constraints are much stronger than the old ones from rare Z-decays at LEP. Then we calculate the processes Z->l_i l_j and γγ->l_i l_j. We find that with the updated constraints the R-violating couplings can still enhance the rates of these processes to the sensitivity of GigaZ and photon-photon collision options of the ILC.

preprint2010arXiv

Matter-wave solitons and finite-amplitude Bloch waves in optical lattices with a spatially modulated nonlinearity

We investigate solitons and nonlinear Bloch waves in Bose-Einstein condensates trapped in optical lattices. By introducing specially designed localized profiles of the spatial modulation of the attractive nonlinearity, we construct an infinite number of exact soliton solutions in terms of the Mathieu and elliptic functions, with the chemical potential belonging to the semi-infinite bandgap of the optical-lattice-induced spectrum. Starting from the exact solutions, we employ the relaxation method to construct generic families of soliton solutions in a numerical form. The stability of the solitons is investigated through the computation of the eigenvalues for small perturbations, and also by direct simulations. Finally, we demonstrate a virtually exact (in the numerical sense) composition relation between nonlinear Bloch waves and solitons.

preprint2010arXiv

Top quark forward-backward asymmetry at the Tevatron: a comparative study in different new physics models

The top quark forward-backward asymmetry A_{FB}^t measured at the Tevatron is above the Standard Model prediction by more than 2-sigma deviation, which might be a harbinger for new physics. In this work we examine the contribution to A_{FB}^t in two different new physics models: one is the minimal supersymmetric model without R-parity (RPV-MSSM) which contributes to A_{FB}^t via sparticle-mediated t-channel process d d_bar-> t t_bar; the other is the third-generation enhanced left-right model (LR model) which contributes to A_{FB}^t via Z'-mediated t-channel or s-channel processes. We find that in the parameter space allowed by the tt_bar production rate and the tt_bar invariant mass distribution at the Tevatron, the LR model can enhance A_{FB}^t to within the 2-sigma region of the Tevatron data for the major part of the parameter space, and in optimal case A_{FB}^t can reach 12% which is slightly below the 1-sigma lower bound. For the RPV-MSSM, only in a narrow part of the parameter space can the λ'' couplings enhance A_{FB}^t to within the 2-sigma region while the λ' couplings just produce negative contributions to worsen the fit.

preprint2010arXiv

Vortices and ring dark solitons in nonlinear amplifying waveguides

We consider the generation and propagation of (2+1)-dimensional beams in a nonlinear waveguide with the linear gain. Simple self-similar evolution of the beams is achieved at the asymptotic stage, if the input beams represent the fundamental mode. On the contrary, if they carry vorticity or amplitude nodes (or phase slips), vortex tori and ring dark solitons (RDSs) are generated, featuring another type of the self-similar evolution, with an exponentially shrinking vortex core or notch of the RDS. Numerical and analytical considerations reveal that these self-similar structures are robust entities in amplifying waveguides, being \emph{stable} against azimuthal perturbations.

Lei Wu

What is connected

Connect this record

See the researcher in context

Building this map preview

117 published item(s)

Accelerated simulation of multiscale gas-radiation coupling flows via a general synthetic iterative scheme

Dual-Diffusional Generative Fashion Recommendation

Geomagnetic constraints on Millicharged Dark Matter

Embedding Inequalities for Barron-type Spaces

Field-Induced Lifshitz Transition in the Magnetic Weyl Semimetal Candidate PrAlSi

A kinetic model for rarefied flows of molecular gas with vibrational modes

A Logarithm Depth Quantum Converter: From One-hot Encoding to Binary Encoding

A spectral-based analysis of the separation between two-layer neural networks and linear methods

A Survey on EOSIO Systems Security: Vulnerability, Attack, and Mitigation

Axion Dark Radiation: Hubble Tension and Hyper-kamiokande Neutrino Experiment

Beyond the Quadratic Approximation: the Multiscale Structure of Neural Network Loss Landscapes

Deep Learning Jet Image as a Probe of Light Higgsino Dark Matter at the LHC

Electroweak Precision Fit and New Physics in light of $W$ Boson Mass

Exploiting the Potential of Datasets: A Data-Centric Approach for Model Robustness

Global fits of SUSY at future Higgs factories

Hilbert Expansion for the Relativistic Landau Equation

HoVer-Trans: Anatomy-aware HoVer-Transformer for ROI-free Breast Cancer Diagnosis in Ultrasound Images

iLibScope: Reliable Third-Party Library Detection for iOS Mobile Apps

Inert Higgs Dark Matter for CDF-II W-boson Mass and Detection Prospects

Learning a Single Neuron for Non-monotonic Activation Functions

Logarithmic cotangent bundles, Chern-Mather classes, and the Huh-Sturmfels Involution conjecture

Non-monotonic heat dissipation phenomenon in close-packed quasi-2D and 3D hotspot system

On self-supervised multi-modal representation learning: An application to Alzheimer's disease

On the emergence of heat waves in the transient thermal grating geometry

Penny Wise and Pound Foolish: Quantifying the Risk of Unlimited Approval of ERC20 Tokens on Ethereum

Probing ultra-light dark photon from inverse Compton-like scattering

Rarefaction effects in head-on collision of two identical droplets

Self-supervised multimodal neuroimaging yields predictive representations for a spectrum of Alzheimer's phenotypes

Simple and statistically sound recommendations for analysing physical theories

Tasting the cake: evaluating self-supervised generalization on out-of-distribution multimodal MRI data

A fast-converging scheme for the Phonon Boltzmann equation with dual relaxation times

Direct Detection of Spin-Dependent Sub-GeV Dark Matter via Migdal Effect

On the Quantum Boltzmann Equation near Maxwellian and Vacuum

The Generalization Error of the Minimum-norm Solutions for Over-parameterized Neural Networks

Towards Understanding and Demystifying Bitcoin Mixing Services

VM Matters: A Comparison of WASM VMs and EVMs in the Performance of Blockchain Smart Contracts

A Comparative Analysis of the Optimization and Generalization Property of Two-layer Neural Network and Random Feature Models Under Gradient Descent Dynamics

A Priori Estimates of the Population Risk for Two-layer Neural Networks

Atmospheric Dark Matter and Xenon1T Excess

Automated Deobfuscation of Android Native Binary Code

Bernstein-Sato ideals and hyperplane arrangements

Calibrating the dynamic Huff model for business analysis using location big data

Characterizing Cryptocurrency Exchange Scams

Characterizing EOSIO Blockchain

Complexity Measures for Neural Networks with General Activation Functions Using Path-based Norms

DEPOSafe: Demystifying the Fake Deposit Vulnerability in Ethereum Smart Contracts

Fast convergence and asymptotic preserving of the General Synthetic Iterative Scheme

General synthetic iteration scheme for non-linear gas kinetic simulation of multi-scale rarefied gas flows

Isotriviality of smooth families of varieties of general type

LFV and g-2 in non-universal SUSY models with light higgsinos

Log D-modules and Index theorems

Nearby and vanishing cycles for perverse sheaves and D-modules

On the comparison of nearby cycles via $b$-functions

On the coverage of neutralino dark matter in coannihilations at the upgraded LHC

Reinterpretation of LHC Results for New Physics: Status and Recommendations after Run 2

Security Analysis of EOSIO Smart Contracts

The Quenching-Activation Behavior of the Gradient Descent Dynamics for Two-layer Neural Network Models

The Slow Deterioration of the Generalization Error of the Random Feature Model

Top quark as a probe of heavy Majorana neutrino at the LHC and future collider

Can we find steady-state solutions to multiscale rarefied gas flows within dozens of iterations?

Testing electroweak SUSY for muon $g-2$ and dark matter at the LHC and beyond

Unveiling CP property of top-Higgs coupling with graph neural networks at the LHC

Boundary Layer of Transport Equation with In-Flow Boundary

High-Order Implicit Hybridizable Discontinuous Galerkin Method for the Boltzmann Equation

750 GeV Composite Axion as the LHC Diphoton Resonance

750 GeV diphoton resonance in a top and bottom seesaw model

750 GeV Diphoton Resonance, 125 GeV Higgs and Muon g-2 Anomaly in Deflected Anomaly Mediation SUSY Breaking Scenario

A comparative study of discrete velocity methods for rarefied gas flows

Asymptotic Analysis of Transport Equation in Annulus

Closing up a light stop window in natural SUSY at LHC

Diffusive Limit with Geometric Correction of Unsteady Neutron Transport Equation

Electroweak Baryogenesis with Anomalous Higgs Couplings

Explaining 750 GeV diphoton excess from top/bottom partner cascade decay in two-Higgs-doublet model extension

Finding the magnetic center of sextupole using vibrating wire technique