Authors: Griffin T. Goodwin (Georgia State University), Dustin Kempton (Georgia State University), Reet Gupta (Georgia State University), Viacheslav M. Sadykov (Georgia State University), Petrus C. Martens (Georgia State University)

The Space Weather Analytics for Solar Flares (SWAN-SF) benchmark dataset has proven to be an invaluable resource to the flare forecasting community. Containing carefully cross-checked magnetogram data for over 4000 active regions and 10,000 flaring events, SWAN-SF has enabled researchers to efficiently train, test, and validate their predictive models with confidence. However, since its release in 2020, the dataset has seen no significant updates. As a result, the goal of this work is twofold: first, we plan to temporally expand the existing dataset to include the most recently available HMI active region patches (HARPS); and second, we aim to incorporate texture-based parameters derived from extreme ultraviolet images taken by the Solar Dynamics Observatory’s Atmospheric Imaging Assembly (SDO/AIA). The purpose of these updates is to enable researchers to investigate how flare forecasting is impacted across two solar cycles and to improve prediction accuracy near the limbs. Our methodology for producing the dataset, along with some preliminary results, will be presented here.