Skip to main content
Color-coded matrix diagram illustrating the H2 matrix representation of a 3D Boundary Element Method (BEM) matrix, with red blocks indicating fully formed matrix sections and green blocks showing low-rank compressed regions in a nested structure. A person standing outdoors in front of tall evergreen trees, with sunlight filtering through the branches. Certificate for the Best Paper Award at the 32nd IEEE Symposium on Computer Arithmetic (ARITH-2025), held in El Paso, Texas, May 4–7, 2025. The award is given to Jackson Vanover, James Demmel, Xiaoye Sherry Li, and Cindy Rubio-González for the paper titled "EXCVATE: Spoofing Exceptions and Solving Constraints to Test Exception Handling in Numerical Libraries." The certificate is signed by Program Chairs Ping Tak Peter Tang and Guillaume Melquiond. A large group of people standing on a staircase inside a building, posing for a group photo. The setting includes indoor plants and modern architectural elements. Three participants during the AI Jam event The Perlmutter supercomputer at NERSC, housed in a large facility with exposed ceiling pipes. Its cabinet panels feature a vibrant design with cosmic imagery, including orange and blue nebulae, a historical photo of physicist Saul Perlmutter with his research team, and an illustration of telescopes. The word 'Perlmutter' is prominently displayed across the panels. A photo collage of three individuals speaking at an event. The left image features a man with shoulder-length dark hair, wearing a white patterned shirt and a lanyard, gesturing with his hands while speaking. The middle image shows a woman with long black hair, wearing a striped button-up shirt and a dark cardigan, speaking into a microphone while gesturing with her hands. The right image captures a woman with curly dark hair, wearing a black jacket with decorative buttons, looking off to the side while standing in front of a screen displaying text and graphics. Group photo of ECP Leadership award recipients. Animation showing molecules represented as color-coded dots. The image illustrates how EScAIP optimizes reaction paths involving complex structures with many atoms, highlighting the speed and low memory usage critical to its efficiency. This method is under development by Samuel Blau and colleagues using data from the Open Catalyst Project. Visualization of ion collisions captured by BNL’s STAR detector. Multicolored lines trace particle tracks, created by quarks and gluons from the collision. GPTune-enhanced collisions produce more particles, offering scientists a deeper look into subatomic behavior. The image's radiant lines form a bullseye pattern against a black background, highlighting the intricate paths of the particles.
Computing Sciences logo

Connect

Our Organization