Bayesian Inferring Nucleon’s Gravitation Form Factors via Near-threshold J / ψ J/\psi Photoproduction
Yuxun Guo, Feng Yuan, Wenbin Zhao · 2025
Read Bayesian Inferring Nucleon’s Gravitation Form Factors via Near-threshold J / ψ J/\psi Photoproduction by Yuxun Guo, Feng Yuan, Wenbin Zhao (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-10532.txt.
Introduction
Shooting photons at a proton and watching a heavy particle called a J/psi come out can reveal how mass and pressure are spread inside the proton. Yuxun Guo, Feng Yuan, and Wenbin Zhao use fresh measurements of this near-threshold reaction from two experiments to pin down the proton's gravitational form factors, which encode that inner structure. They analyze the data with Bayesian methods inside a standard parton framework.
Their fits agree well with the best lattice calculations of the same quantities. In particular, the data strongly favor negative values for the quark and gluon parts of one key form factor. The authors argue that sharper experiments planned for the future could turn this reaction into a precise tool for mapping the forces that hold the proton together.
Catalog record
| CREATOR | Yuxun Guo, Feng Yuan, Wenbin Zhao |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-10532
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bayesian Inferring Nucleon’s Gravitation Form Factors via Near-threshold J / ψ J/\psi Photoproduction Yuxun Guo, Feng Yuan, Wenbin Zhao Abstract With Bayesian inference, we investigate the impact of recent near-threshold J / ψ J/\psi production measurements by the J / ψ J/\psi 007 experiment and GlueX collaboration on the extraction of proton’s gravitational form factors. We apply the generalized parton distribution framework at the next-to-leading order and demonstrate a stable expansion for the near-threshold kinematics. We find that the experimental constraints are in good agreement with the state-of-the-art lattice simulations, where negative C q ( t ) C_{q}(t) and C g ( t ) C_{g}(t) are strongly preferred. This highlights a great potential to extract them from future high-precision experiments. Introduction. Nucleon tomography encoded in the gravitational form factors (GFFs) has drawn rising interest in the past decade. They hold the keys to understanding the fundamental structure, including the mass, spin, and mechanical properties, of the nucleon [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ] . One of the most important questions is to identify the gluonic components, accessing which, nevertheless, has been challenging due to their elusive nature. It has been suggested long ago that heavy quarkonia can help to unveil the underlying dynamics of the gluon [ 8 , 9 , 10 , 11 , 12 , 13 ] . The near-threshold heavy quarkonium photoproduction, \gamma+p\to J/\psi+p\ , (1) has been extensively investigated from both experiment and theory sides to explore the gluonic GFFs in recent years [ 14 , 6 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 22 , 28 , 29 ] . Together with the first principle lattice calculations [ 30 , 31 , 32 , 31 , 33 ] , this will provide a unique opportunity to uncover the fundamental nucleon structure. One especially interesting feature of the near-threshold kinematics is that the momentum transfer is large and the skewness parameter \xi approaches 1 in the heavy-quarkonia limit. Applying the QCD factorization in the generalized parton distribution (GPD) framework and utilizing this feature, it was found that the leading contribution to the amplitude is from the lowest moments of GPDs [ 21 , 34 , 28 ] . This leading-moment dominance allows us to constrain the gluon GFFs through asymptotic expansion with large \xi [ 21 , 22 , 28 ] . In the GPD framework, the cross-sections of near-threshold heavy quarkonium productions read [ 21 ] : \begin{split}\frac{d\sigma}{dt}=\frac{\alpha_{\rm EM}e_{Q}^{2}}{4\left(W^{2}-M_{N}^{2}\right)^{2}}\frac{(16\pi\alpha_{S})^{2}}{3M_{V}^{3}}|\psi_{\rm NR}(0)|^{2}|G(t,\xi)|^{2}\ ,\end{split} (2) where t is the momentum transfer squared, W is the center of mass energy, M_{N} and M_{V} represent the masses of the nucleon and heavy quarkonium, respectively. \psi_{\rm NR}(0) stands for the non-relativistic wave function of the heavy quarkonium at r=0 in the non-relativistic QCD (NRQCD) [ 35 ] . The hadronic matrix element G(\xi,t) can be written as \displaystyle\begin{split}|G(t,\xi)|^{2}=&\left(1-\xi^{2}\right)(\mathcal{H}+\mathcal{E})^{2}-2\mathcal{E}(\mathcal{H}+\mathcal{E})\\ &+\left(1-\frac{t}{4M_{N}^{2}}\right)\mathcal{E}^{2}\ ,\end{split} (3) where \mathcal{H}=\sum_{i=q,g}\mathcal{H}_{i} and \mathcal{E}=\sum_{i=q,g}\mathcal{E}_{i} are the Compton-like form factors (CFFs) that can be factorized in terms of the quark and gluon GPDs, H_{q/g} and E_{q/g} [ 36 , 37 , 38 ] . For example, \mathcal{H}_{q/g}(\xi,t) are written as, \displaystyle\begin{split}\mathcal{H}_{q/g}(\xi,t)\equiv\int_{-1}^{1}\text{d}x\penalty\ &C_{q/g}(x,\xi,\mu_{f})H_{q/g}(x,\xi,t,\mu_{f})\ ,\end{split} (4) and similarly for \mathcal{E}_{q/g} . The Wilson coefficients C_{q/g} have been computed up to NLO [ 36 , 37 , 38 ] . At LO, only gluons contribute, and the leading-moment approximation leads to \mathcal{H}(\xi,t)\approx 2\left[A_{g}(t)+4\xi^{2}C_{g}(t)\right]/\xi^{2} and \mathcal{E}(\xi,t)\approx 2\left[B_{g}(t)-4\xi^{2}C_{g}(t)\right]/\xi^{2} with A_{g}(t) , B_{g}(t) and C_{g}(t) the gluon GFFs [ 2 ] . This provides an important constraint on the GFFs. To establish a solid case to extract the GFFs from these experiments, one has to go beyond the above LO picture. In this work, we will advance the theory developments in two aspects: apply the NLO calculations [ 36 , 37 , 38 ] in the threshold kinematics with large- \xi expansion and utilize the Bayesian inference method in the phenomenological analysis of the experimental data. Taking into account the NLO corrections will not only help to reduce the theoretical uncertainties but also lead to access to the quark GFFs because the quark GPDs will start to contribute at the NLO. Therefore, at this order, we will be able to constrain both quark and gluon GFFs, and their total contribution to the proton’s GFFs. In addition, to improve the accuracy of our analysis, we will employ the advanced Bayesian inference method. This offers a statistically rigorous and interpretable extraction, which will be extremely valuable since both quark and gluon GFFs enter at NLO and are strongly correlated in the analysis. Advanced computational methods, such as Bayesian inference, machine learning/artificial intelligence, have been widely applied in nuclear physics, see, e.g., Refs. [ 39 , 40 ] , and in particular to constrain the nucleon properties related to the GPDs [ 41 , 42 , 43 , 44 ] . In our analysis, we will also consider the constraints from the lattice simulations [ 32 ] . These lattice results will constrain, in particular, the GFFs at small momentum transfer. The combined fit to the lattice results and experimental data give us the most accurate picture of the gluon GFFs. We also notice that the underlying mechanism for near-threshold J/\psi photoproduction has been explored in different frameworks [ 45 , 46 , 47 , 48 , 49 ] . Future experiments at JLab and the electron-ion collider will help to identify the production mechanism. GPD framework at NLO in QCD. In the GPD framework, the differential cross-section of near-threshold heavy quarkonium photoproduction depends on the CFFs. As discussed in [ 28 ] , the conformal expansion of CFFs provides better convergence in the large- \xi kinematics, which we will focus on in the following analysis. The gluonic CFFs (gCFFs) can be formally written in terms of the conformal moments of GPD as, \displaystyle\begin{split}\text{Re}\mathcal{H}_{g}(\xi,t)=\sum_{n=0}^{\infty}\frac{1}{\xi^{2n+2}}\bar{C}^{(2n+1)}_{g}\bar{H}_{g}^{(2n+1)}(\xi,t)\ ,\end{split} (5) which also applies to the quark CFF (qCFF) and the \mathcal{E}_{q,g} CFFs. The basis p_{q/g}^{j}(x,\xi) and its dual c_{q/g}^{j}(x,\xi) of this conformal expansion with conformal spin j are orthonormal: \int_{-1}^{1}\text{d}x\penalty\ c_{i}^{j}(x,\xi)p_{i}^{k}(x,\xi)=(-1)^{j}\delta^{jk} for i=q,g [ 50 ] , such that the conformal moments of GPDs are, \bar{H}_{q/g}^{(2n+1)}(\xi,t)=\int_{-1}^{1}\text{d}x\penalty\ H_{q/g}(x,\xi,t)c_{q/g}^{2n+1}(x,\xi)\ , (6) and the corresponding Wilson coefficients are \bar{C}_{q/g}^{(2n+1)}=\pm\int_{-1}^{1}\text{d}x\penalty\ C_{q/g}(x,1)p_{q/g}^{2n+1}(x,1)\ . (7) As the Wilson coefficient C_{q/g}(x,\xi) depends solely on x/\xi up to an overall factor, its \xi -dependence can be pulled out and \xi is set to 1 in the above expression. A conventional - sign appears in the prefactor for the quark. For heavy vector meson, only the singlet quark GPD H_{\Sigma}(x,\xi,t)\equiv\sum_{\begin{subarray}{c}q=u,d,\cdots\end{subarray}}H_{q}(x,\xi,t)-H_{q}(-x,\xi,t) contributes, and the corresponding Wilson coefficient C_{\Sigma}(x,\xi) is normalized such that \mathcal{H}_{\Sigma}=\sum_{q}\mathcal{H}_{q} . In the following, the q will always refer to the singlet \Sigma . Analytical expressions of the Wilson coefficients in conformal spin space are generally not guaranteed beyond LO. For the leading moment with j=1 , they read \displaystyle\bar{C}_{g}^{(1)} \displaystyle=\frac{5}{4}+\alpha_{S}\left[\bar{c}_{g}^{1}-\frac{55}{16\pi}\log\left(\frac{m_{c}^{2}}{\mu_{F}^{2}}\right)\right]+\mathcal{O}(\alpha_{S}^{2})\ , (8) \displaystyle\bar{C}_{q}^{(1)} \displaystyle=0+\alpha_{S}\left[\bar{c}_{q}^{1}+\frac{10}{9\pi}\log\left(\frac{m_{c}^{2}}{\mu_{F}^{2}}\right)\right]+\mathcal{O}(\alpha_{S}^{2})\ , (9) calculated with the NLO Wilson coefficients [ 36 , 37 , 38 ] . The quantities in front of the \log\left(m_{c}^{2}/\mu_{F}^{2}\right) are nothing but the LO anomalous dimensions [ 51 ] and LO Wilson coefficients combined. On the other hand, the two genuine