AI 中文总结
本文综述了广义部分子分布（GPDs）的现象学、提取方法及其在强子成像中的应用，重点讨论了从实验数据提取康普顿形状因子和重建GPDs的逆问题，并展望了未来精确成像时代的新机遇。
AI 中文摘要
广义部分子分布（GPDs）为研究强子中夸克和胶子的关联动量与空间结构，以及获取角动量和QCD能量-动量张量等基本性质提供了一个框架。在这篇综述中，我们讨论了GPD现象学的现状，强调了将深度虚排他测量与底层部分子结构联系起来所面临的挑战。我们将这一问题组织为两个相继的逆问题：从测量观测量中提取康普顿形状因子（CFFs），以及从定义CFFs的卷积积分中重建GPDs。我们回顾了GPDs的理论描述和现象学参数化、当前CFF和GPD提取的策略，以及格点QCD、贝叶斯推断、不确定性量化和人工智能（包括神经网络和可解释机器学习方法）的作用。我们讨论了当前测定的局限性以及杰斐逊实验室计划、互补排他过程和未来电子-离子对撞机所提供的机会。最后，我们考虑了日益精确和多维的信息，连同新的统计和人工智能方法，将如何在精确强子成像的新兴时代改变GPD现象学。
英文摘要
Generalized Parton Distributions (GPDs) provide a framework for investigating the correlated momentum and spatial structure of quarks and gluons in hadrons and for accessing fundamental properties such as angular momentum and the QCD energy-momentum tensor. In this review, we discuss the present status of GPD phenomenology, emphasizing the challenges involved in connecting deeply virtual exclusive measurements to the underlying partonic structure. We organize this problem in terms of two successive inverse problems: the extraction of Compton Form Factors (CFFs) from measured observables and the reconstruction of GPDs from the convolution integrals defining the CFFs. We review the theoretical description and phenomenological parametrizations of GPDs, current strategies for CFF and GPD extraction, and the role of lattice QCD, Bayesian inference, uncertainty quantification, and artificial intelligence, including neural networks and interpretable machine-learning approaches. We discuss the limitations of present determinations and the opportunities offered by the Jefferson Lab program, complementary exclusive processes, and the future Electron-Ion Collider. Finally, we consider how increasingly precise and multidimensional information, together with new statistical and AI methodologies, will transform GPD phenomenology in the emerging era of precision hadron imaging.
Comments31 pages, 10 figures
