Leveraging Multimodality for Real-Time Classification of Transients and Variables found by the Zwicky Transient Facility
利用多模态对Zwicky瞬变设施发现的瞬变和变星进行实时分类
机构 * Department of Physics and Astronomy, Northwestern University(西北大学物理与天文学系) ; Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA)(天体物理跨学科探索与研究中心) ; NSF-Simons AI Institute for the Sky (SkAI)(NSF-西蒙斯天空人工智能研究所) ; School of Physics and Astronomy, University of Minnesota(明尼苏达大学物理与天文学学院) ; Division of Physics, Mathematics, and Astronomy, California Institute of Technology(加州理工学院物理、数学与天文学部)
专题命中 多模态评测 :multimodal(abstract)
AI总结 针对时域巡天中实时分类的挑战,本文提出ORACLE-2模型,结合光变曲线、元数据和图像进行多模态层次分类,在ZTF数据上宏F1达0.73,较仅用光变曲线提升40%。
Comments 29 Pages, 15 Figures, 8 Tables. Comments welcome