State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack. Existing fault-tolerance mechanisms either impose non-trivial overhead during failure-free e…
Machine learning (ML) methods have been widely explored for predicting material properties. However, due to the rapid development of ML techniques and the diversity of available models, performance comparisons between traditional and graph-based machine learning models remain lim…