The Discharge Me! shared task invites participants to streamline the generation of discharge summary sections in the EHR, with the goal of alleviating clinician burden and enhancing patient care quality. Leveraging a dataset derived from MIMIC-IV, participants are tasked with generating the "Brief Hospital Course" and "Discharge Instructions" sections using over 100,000 admissions from the Emergency Department (ED). Submission guidelines and data access agreements are detailed on the task and competition website (https://stanford-aimi.github.io/discharge-me https://gcc02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstanford-aimi.github.io%2Fdischarge-me&data=05%7C02%7Cddemner%40mail.nih.gov%7C62ce665f2fbb4e0edaa708dc330ecb00%7C14b77578977342d58507251ca2dc2b06%7C0%7C0%7C638441385832644699%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=RDI5XS5sY0%2BifU2CWSRCosqAo5jse75%2BTzgg279SZBU%3D&reserved=0), with system submissions due by May 10th, 2024. Accepted papers will be presented at the 23rd Workshop on Biomedical Natural Language Processing at ACL 2024. Join us in revolutionizing clinical documentation and improving healthcare workflows! For further details and registration, please visit the Codabench competition page linked on the task website.