arXiv subjects

Gen Wang

Publications and source records attributed to Gen Wang.

At least 19 recordsLinked to original sources

arXiv2026

BMW/DMZ calculation of the hadronic vacuum polarisation for the muon magnetic moment

For twenty years, a persistent discrepancy between experimental measurements and theoretical calculations of the muon anomalous magnetic moment have provided tantalising hints of new physics. In recent years, improvements to the experimental precision have appeared to make the tension stronger and stronger. However, at the same time, our lattice calculation overturned the theoretical consensus, completely eliminating the tension. I will present the latest results from the Budapest-Marseille-Wuppertal (BMW) and DMZ collaborations, with a hybrid determination of the hadronic vacuum polarisation contribution to a precision of 0.45%

hep-lat↗

arXiv2024

aiXcoder-7B: A Lightweight and Effective Large Language Model for Code Processing

Large Language Models (LLMs) have been widely used in code completion, and researchers are focusing on scaling up LLMs to improve their accuracy. However, larger LLMs have lower inference efficiency, affecting developers' experience and productivity. In this paper, we propose a lightweight and effective LLM for code completion named aiXcoder-7B. Compared to existing LLMs, aiXcoder-7B achieves higher code completion accuracy while having smaller scales (i.e., 7 billion parameters). We attribute the superiority of aiXcoder-7B to three key factors: (1) Multi-objective training. We employ three training objectives, one of which is our proposed Structured Fill-In-the-Middle (SFIM). SFIM considers the syntax structures in code and effectively improves the performance of LLMs for code. (2) Diverse data sampling strategies. They consider inter-file relationships and enhance the capability of LLMs in understanding cross-file contexts. (3) Extensive high-quality data. We establish a rigorous data collection pipeline and consume a total of 1.2 trillion unique tokens for training aiXcoder-7B. This vast volume of data enables aiXcoder-7B to learn a broad distribution of code. We evaluate aiXcoder-7B in five popular code completion benchmarks and a new benchmark collected by this paper. The results show that aiXcoder-7B outperforms the latest six LLMs with similar sizes and even surpasses four larger LLMs (e.g., StarCoder2-15B and CodeLlama-34B), positioning aiXcoder-7B as a lightweight and effective LLM for academia and industry. Finally, we summarize three valuable insights for helping practitioners train the next generations of LLMs for code. aiXcoder-7B has been open-souced and gained significant attention. Until January 2025, aiXcoder-7B has received 2,226 GitHub Stars.

cs.CL↗

arXiv2024

Trace anomaly form factors from lattice QCD

The hadron mass can be obtained through the calculation of the trace of the energy-momentum tensor in the hadron which includes the trace anomaly and sigma terms. The anomaly due to conformal symmetry breaking is believed to be an important ingredient for hadron mass generation and confinement. In this work, we will present the calculation of the glue part of the trace anomaly form factors of the pion up to $Q^2\sim 4.3~\mathrm{GeV}^2$ and the nucleon up to $Q^2\sim 1~\mathrm{GeV}^2$. The calculations are performed on a domain wall fermion ensemble with overlap valence quarks at seven valence pion masses varying from $\sim 250$ to $\sim 540$ MeV, including the unitary point $\sim 340$ MeV. We calculate the radius of the glue trace anomaly for the pion and the nucleon from the $z$ expansion. By performing a two-dimensional Fourier transform on the glue trace anomaly form factors in the infinite momentum frame with no energy transfer, we also obtain their spatial distributions for several valence quark masses. The results are qualitatively extrapolated to the physical valence pion mass with systematic errors from the unphysical sea quark mass, discretization effects in the renormalization sum rule, and finite-volume effects to be addressed in the future. We find the pion's form factor changes sign, as does its spatial distribution, for light quark masses. This explains how the trace anomaly contribution to the pion mass approaches zero toward the chiral limit.

hep-lat↗

arXiv2023

Separation of Infrared and Bulk in Thermal QCD

A new thermal regime of QCD, featuring decoupled scale-invariant infrared glue, has been proposed to exist both in pure-glue (N$_f$=0) and ``real-world" (N$_f$=2+1 at physical quark masses) QCD. In this {\it IR phase}, elementary degrees of freedom flood the infrared, forming a distinct component independent from the bulk. This behavior necessitates non-analyticities in the theory. In pure-glue QCD, such non-analyticities have been shown to arise via Anderson-like mobility edges in Dirac spectra ($\lambda_{\rm IR} \!=\! 0$, $\pm \lambda_\text{A} \!\neq\! 0$), as manifested in the dimension function $d_{\rm IR} (\lambda)$. Here, we present the first evidence, based on lattice QCD calculation at $a$=0.105 fm, that this mechanism is also at work in real-world QCD, thus supporting the existence of the proposed IR regime in nature. An important aspect of our results is that, while at $T\!=\!234\,$MeV we find a dimensional jump between zero modes and lowest near-zero modes very close to unity ($d_{\rm IR} \!=\!3$ to $d_{\rm IR} \!\simeq\! 2$), similar to the IR phase of pure-glue QCD, at $T\!=\!187\,$MeV we observe a continuous $\lambda$-dependence. This suggests that thermal states just {\it above} the chiral crossover are non-analytically (in $T$) connected to thermal state at $T\!=\!234\,$MeV, supporting the key original proposition that the transition into the IR regime occurs at a temperature strictly above the chiral crossover.

hep-lat↗

arXiv2023

Lattice calculation of the $\pi^0$, $\eta$ and $\eta^{\prime}$ transition form factors and the hadronic light-by-light contribution to the muon $g-2$

In this paper we present a first ab-initio calculation of the $\pi^0$, $\eta$ and $\eta^{\prime}$ transition form factors performed with physical light-quark masses. We provide a complete parametrization of the form factors that includes both single and double-virtual kinematics. Our results are compared with experimental measurements of the form factors in the space-like region and with the measured two-photon decay widths. In a second step, our parametrizations of the transition form factors are used to compute the dominant pseudoscalar-pole contributions to the hadronic light-by-light scattering in the muon $g-2$. Our final result reads $a_{\mu}^{\rm hlbl, ps-pole} = (85.1 \pm 5.2) \times 10^{-11}$. Although the pion-pole is dominant, we confirm that, together, the $\eta$ and $\eta^{\prime}$ provide roughly half of its contribution.

hep-lat↗

arXiv2023

Nucleon Electric Dipole Moment from the $\theta$ Term with Lattice Chiral Fermions

We calculate the nucleon electric dipole moment (EDM) from the $\theta$ term with overlap fermions on three domain wall lattices with different sea pion masses at lattice spacing 0.11 fm. Due to the chiral symmetry conserved by the overlap fermions, we have well defined topological charge and chiral limit for the EDM. Thus, the chiral extrapolation can be carried out reliably at nonzero lattice spacings. We use three to four different partially quenched valence pion masses for each sea pion mass and find that the EDM dependence on the valence and sea pion masses behaves oppositely, which can be described by partially quenched chiral perturbation theory. With the help of the cluster decomposition error reduction (CDER) technique, we determine the neutron and proton EDM at the physical pion mass to be $d_{n}=-0.00148\left(14\right)\left(31\right)\bar\theta$ e$\cdot$fm and $d_{p}=0.0038\left(11\right)\left(8\right)\bar\theta$ e$\cdot$fm. This work is a clear demonstration of the advantages of using chiral fermions in the nucleon EDM calculation and paves the road to future precise studies of the strong $CP$ violation effects.

hep-lat↗

arXiv2022

Distance between various discretized fermion actions

We present the leading order mixed-action effect $\Delta_{\rm mix}\equiv m_{\pi,{\rm vs}}^2-\frac{m_{\pi,{\rm vv}}^2+m_{\pi,{\rm ss}}^2}{2}$ using HISQ, clover or overlap valence fermion actions on gauge ensembles using various sea fermion actions across a widely-used lattice spacing range $a\in [0.04,0.19]$~fm. The results suggest that $\Delta_{\rm mix}$ decreases as the fourth order of the lattice spacing on the gauge ensembles with dynamical chiral sea fermions, such as Domain wall or HISQ fermions. When a clover sea fermion action which has explicit chiral symmetry breaking is used in the ensemble, $\Delta_{\rm mix}$ can be much larger regardless of the valence fermion action used.

hep-lat↗

arXiv2022

Muon g-2 with overlap valence fermions

We present a lattice calculation of the leading order (LO) hadronic vacuum polarization (HVP) contribution to the muon anomalous magnetic moment for the connected light and strange quarks, $a^{\rm W}_{{\rm con}, l/s}$ in the widely used window $t_0=0.4~\mathrm{fm}$, $t_1=1.0~\mathrm{fm}$, $\Delta=0.15~\mathrm{fm}$, and also of $a^{\rm S}_{{\rm con}, l/s}$ in the short distance region. We use overlap fermions on 4 physical-point ensembles. Two 2+1 flavor RBC/UKQCD ensembles use domain wall fermions (DWF) and Iwasaki gauge actions at $a = 0.084$ and 0.114 fm, and two 2+1+1 flavor MILC ensembles use the highly improved staggered quark (HISQ) and Symanzik gauge actions at $a = 0.088$ and 0.121 fm. We have incorporated infinite volume corrections from 3 additional DWF ensembles at ${\rm L}$ = 4.8, 6.4 and 9.6 fm and physical pion mass. For $a^{\rm W}_{{\rm con}, l}$, we find that our results on the two smaller lattice spacings are consistent with those using the unitary setup, but those at the two coarser lattice spacings are slightly different. Eventually, we predict $a^{\rm W}_{{\rm con}, l}=206.7(1.5)(1.0)$ and $a^{\rm W}_{{\rm con}, s}=26.8(0.1)(0.3)$, using linear extrapolation in $a^2$, with systematic uncertainties estimated from the difference of the central values from the RBC/UKQCD and MILC ensembles.

hep-lat↗

arXiv2021

Proton momentum and angular momentum decompositions with overlap fermions

We present a calculation of the proton momentum and angular momentum decompositions using overlap fermions on a $2+1$-flavor RBC/UKQCD domain-wall lattice at 0.143 fm with a pion mass of 171 MeV which is close to the physical one. A complete determination of the momentum and angular momentum fractions carried by up, down, strange and glue inside the proton has been done with valence pion masses varying from 171 to 391 MeV. We have utilized fast Fourier transform on the stochastic-sandwich method for connected-insertion parts and the cluster-decomposition error reduction technique for disconnected-insertion parts has been used to reduce statistical errors. The full nonperturbative renormalization and mixing between the quark and glue operators are carried out. The final results are normalized with the momentum and angular momentum sum rules and reported at the physical valence pion mass at ${\overline{\rm {MS}}}\, (\mu = 2\ {\rm{GeV}})$. The renormalized momentum fractions for the quarks and glue are $\langle x \rangle^q = 0.491(20)(23)$ and $\langle x \rangle^g = 0.509(20)(23)$, respectively, and the renormalized total angular momentum fractions for quarks and glue are $2 J^q = 0.539(22)(44)$ and $2 J^g = 0.461(22)(44)$, respectively. The quark spin fraction is $\Sigma = 0.405(25)(37)$ from our previous work and the quark orbital angular momentum fraction is deduced from $2 L^q = 2 J^q - \Sigma$ to be $0.134(22)(44)$.

hep-lat↗

arXiv2021

Nucleon isovector scalar charge from overlap fermions

We calculate the nucleon isovector scalar charge in lattice QCD using overlap fermions on five ensembles of gauge configurations generated by the RBC/UKQCD collaboration using the domain-wall quark action with $2+1$ dynamical flavors. The five ensembles cover five pion masses, $m_\pi \approx$ 139, 171, 302, 337 and 371 MeV, and four lattice spacings, $a \approx $ 0.06, 0.08, 0.11 and 0.14 fm. Three to six valence quark masses are computed on each ensemble to investigate the pion mass dependence. The extrapolation to the physical pion mass, continuum and infinite volume limits is obtained by a global fit of all data to a formula originated from partially quenched chiral perturbation theory. The excited-states contamination is carefully analyzed with 3--5 sink-source separations and multi-state fits. Our final result, in the $\overline{\text{MS}}$ scheme at 2 GeV, is $g_{S}^{u-d}= 0.94 (10)_{stat}(8)_{sys}$, where the first error is the statistical error and the second is the systematic error.

hep-lat↗

arXiv2020

Lattice Calculation of Pion Form Factor with Overlap Fermions

We present a precise calculation of the pion form factor using overlap fermions on seven ensembles of 2+1-flavor domain-wall configurations with pion masses varying from 139 to 340 MeV. Taking advantage of the fast Fourier transform and other techniques to access many combinations of source and sink momenta, we find the pion mean square charge radius to be $\langle {r_\pi^2} \rangle= 0.430(5)(13)\ {\rm{fm^2}}$, which agrees well with the experimental result, and includes the systematic uncertainties from chiral extrapolation, lattice spacing and finite-volume dependence. We also find that $\langle {r_\pi^2} \rangle$ depends on both the valence and sea quark masses strongly and predict the pion form factor up to $Q^2 = 1.0 \ {\rm{GeV^2}}$ which agrees with experiments very well.

hep-ph↗

arXiv2020

Geometric quantization rules in QCPB theory

Using the quantum covariant Poisson bracket (QCPB) theory, we can accomplish much more compatible explanations of the quantum mechanics supported by the G-dynamics. We further study the generalized quantum harmonic oscillator equipped with the G-dynamics of type I, such as geometric creation and annihilation operators, and the geometric number operator as an extension of the number operator is well given for the deep discussions, the geometric Hamiltonian operator is expressed as another form. Especially, the geometric quantization rules based on the QCPB theory is then calculated.

physics.gen-ph↗

arXiv2020

The G-dynamics of the QCPB theory

In this paper, we further study the G-dynamics newly emerged in the covariant dynamics defined by the quantum covariant Poisson bracket (QCPB) theory. We propose three new operators based on the G-dynamics, we find that the non-Hermitian operators are inevitable to appear in the framework of the QCPB. Then we seek the precise formulations of the G-dynamics to calculate the practical quantum problems, the G-dynamics of three types in terms of different Hamiltonian are discovered. The geometrinetic energy operator for quantum harmonic oscillator as an application is further calculated.

physics.gen-ph↗

arXiv2020

A revision for Heisenberg uncertainty relation based on environment variable in the QCPB theory

The Heisenberg uncertainty principle and its extensions are all still inequalities form which hold the superior approximate estimations. Based on quantum covariant Poisson bracket theory, we propose quantum geomertainty relation to modify and explain the uncertainty relation to positively give a complete description of reality that enhances the outcome of each measurement with certainty. It demonstrates that entanglement term exists between the observable and the environment and nicely explains how the environment has an effect on the measurement which causes the unavoidable influences.

quant-ph↗

arXiv2020

The generalized covariant Hamilton system in complex coordinates

Imitating methods of working on the GCHS and GSPB defined on ${{\mathbb{R}}^{r}}$ in real coordinates, an attempt to follow this way in complex coordinates is considered, then we try to generalize the PB on ${{\mathbb{C}}^{n}}$ in complex coordinates to the GSPB with zero restriction of the non-degeneracy expressed in complex coordinates that is compatible with the real case in formulas. Thusly, then the GCHS in complex coordinates defined by the GSPB is self-consistent to the real situation. Meanwhile, we find some difference of the GCHS between the real and complex case in formula. Much of what distinguishes a GCHS in real coordinates from a GCHS in complex coordinates is that the TGHS in complex form has an extra expression.

math-ph↗

arXiv2020

Analogy between geodesic equation and the GCHS on Riemannian manifolds

Enlightened by the similar equation form between the GCHS \footnote{GCHS: Generalized Covariant Hamilton System\\GSPB:Generalized structural Poisson bracket} defined by the GSPB and the geodesic equation expressed by geospin variable, we find a deep connection between the geospin matrix and S-dynamics. In this contrastive way, we actually proves that the GCHS is a compatible theory suitable for the curved spacetime as primitively stated. By contrast, geospin matrix in Riemannian geometry has the same physical nature as S-dynamics in GCHS. We obtain a fact that geodesic equation can be naturally derived by the GCHS in terms of the velocity field. We strictly prove that the geometrio $\hat{S}{{\left( {{x}_{k}},{{p}_{i}},H \right)}^{T}}={{\left( {{b}_{k}},{{A}_{i}},w \right)}^{T}}$ holds by using structural operator $\hat{S}$ directly induced by structural derivative ${A}_{i}$ in terms of position ${x}_{k}$, momentum ${p}_{i}$ and Hamiltonian $H$ respectively. It evidently proves that the GCHS on the Riemannian manifold is certainly determined by the Christoffel symbols. As an application, we consider the GCHS on Riemannian geometry.

math.DS↗

arXiv2020

Geometrodynamics based on geodesic equation with Cartan structural equation on Riemannian manifolds

Motivated by the geospin matrix $W$ as a new variable in Riemannian geometry. Then, we use four real dynamical variables $\left\{ a,\alpha ,v,W \right\}$ to show the dynamical essence of Cartan structural equation, we obtain the geometrodynamics on Riemannian manifolds that can be expressed below \begin{align} & \Theta /d{{t}^{2}}=a-v\wedge W,\nonumber & d\Theta /d{{t}^{3}}=v\wedge \alpha -a\wedge W,\nonumber & \Omega /d{{t}^{2}}=\alpha -W\wedge W, \nonumber & d\Omega /d{{t}^{3}}=W\wedge \alpha -\alpha \wedge W\nonumber \end{align} that is valid more generally for any connection in a principal bundle. We can see that the first equation explains the geodesic equation very well, the second formula means the first Bianchi identity, while the last equation reveals the dynamical nature of the second Bianchi identity. It implies that Cartan structural equations on Riemannian manifolds can be rewritten in a real geometrodynamical form based on the geospin matrix.

math.DG↗

arXiv2020

Generalized geometric commutator theory and quantum geometric bracket and its uses

Inspired by the geometric bracket for the generalized covariant Hamilton system, we abstractly define a generalized geometric commutator $$\left[ a,b \right]={{\left[ a,b \right]}_{cr}}+G\left(s,a,b \right)$$ formally equipped with geomutator $G\left(s, a,b \right)=a{{\left[ s,b \right]}_{cr}}-b{{\left[ s,a \right]}_{cr}}$ defined in terms of structural function $s$ related to the structure of spacetime or manifolds itself for revising the classical representation ${{\left[ a,b \right]}_{cr}}=ab-ba$ for any elements $a$ and $b$ of any algebra. Then we use the generalized geometric commutator to define quantum covariant Poisson bracket that is related to the quantum geometric bracket defined by geomutator as a generalization of quantum Poisson bracket. The covariant dynamics includes the generalized Heisenberg equation as a natural extension of Heisenberg equation and G-dynamics based on the quantum geometric bracket, meanwhile, the geometric canonical commutation relation is induced. As an application, we reconsider the canonical commutation relation and the quantization of field to be more complete.

quant-ph↗