The objective of the Scale to See workshop is to bring together researchers in high-performance computing (HPC), artificial intelligence (AI), scientific imaging, and spatiotemporal data science to advance scalable AI methods, algorithms, and computing systems that enable scientific analysis at unprecedented scales. The workshop aims to identify key challenges and emerging opportunities at the intersection of these fields, foster interdisciplinary collaboration, and explore how scaling AI can unlock new scientific capabilities and discoveries.

                                                                                                                  Registration               8:00 - 9:15

   Scale to See From Extreme-Scale Computing and Scalable AI to New Scientific Discovery (Invited Talk from Xiao Wang)                                                   8:30- 9:00

   Exploring the Exascale Universe  (Invited Talk from JD Emberson)                                                                                                                                                 9:00 - 9:30

   Designing Scalable Vision Models for Scientific Understanding (Invited Talk from Mohamed Wahib)                                                                                   9:30 - 10:00

                                                                                                                   Morning Break        10:00 - 10:30

   Faster Scientific Multi-dimensional data analytics with NVIDIA cuPhoton (Invited Talk from Quynh Nguyen)                                                                   10:30 - 11:00

   AC-STRR: Training-Aware Visual Token Reduction for Scalable High-Resolution Medical Vision-Language Models (workshop paper)                        11:00 - 11: 10

   Scalable In-Domain Self-Supervised Foundation Model for Dense Representation Transfer in High-Resolution Plant Imaging (workshop paper)     11:10 - 11: 20

   Fast, Differentiable Modeling for Millimeter-Sky Inference (workshop paper)                                                                                                                          11:20 - 11: 30

   Characterizing Low-bit Deployment of 3D Medical Image Analysis on Portable AI Supercomputer  (workshop paper)                                                   11:30 - 11: 40

   Out-of-View Depth and Motion Prediction in Robotic Endoscopy  (workshop paper)                                                                                                             11:40 - 11: 50

   Toward a Foundation Plug-and-Play Prior for Computed Tomography Reconstruction via a Multimodal Diffusion Model (workshop paper)             11:50 - 12: 00

   SuperSAM: Crafting a SAM Supernetwork via Structured Pruning and Unstructured Parameter Prioritization (workshop paper)                                 12:00 - 12: 10