PanelAccelerating Distributed DLRM Training With Optimized TT Decomposition and Micro-Batching1:30–2:00 p.m. ET PaperBandwidth-Optimal, Fully-Offloaded Collectives11:30 a.m.–12:00 p.m. ET Paper Exploring GPU-to-GPU Communication: Insights Into Supercomputer Interconnects4:30–5:00 p.m. ET...