University of Maryland, College Park（马里兰大学帕克分校）

Flexible Gravitational-Wave Parameter Estimation with Transformers

基于Transformer的灵活引力波参数估计

Annalena Kofler, Maximilian Dax, Stephen R. Green, Jonas Wildberger, Nihar Gupte, Jakob H. Macke, Jonathan Gair, Alessandra Buonanno, Bernhard Schölkopf

机构 * Max Planck Institute for Intelligent Systems, Max-Planck-Ring 4, 72076 T\"ubingen, Germany ； Max Planck Institute for Gravitational Physics (Albert Einstein Institute), Am M\"uhlenberg 1, 14476 Potsdam, Germany ； ELLIS Institute T\"ubingen, Maria-von-Linden-Straße 2, 72076 T\"ubingen, Germany ； T\"ubingen AI Center, Maria-von-Linden-Straße 1, 72076 T\"ubingen, Germany ； Nottingham Centre of Gravity \& School of Mathematical Sciences, University of Nottingham, University Park, Nottingham, NG7 2RD, United Kingdom ； Department of Physics, University of Maryland, College Park, MD 20742, USA ； Machine Learning in Science, University of T\"ubingen \& T\"ubingen AI Center, 72076 T\"ubingen, Germany 纠错

AI总结 提出灵活Transformer架构Dingo-T1，通过训练策略适应不同分析设置，实现引力波参数估计，支持多种配置和一致性测试，中位数采样效率从1.4%提升至4.2%。

Comments 5+15 pages, 3+8 figures

Journal ref Phys. Rev. Lett. 136, 251401, 2026