Our evaluation of representation learning methods is on arXiv

An Evaluation of Representation Learning Methods in Particle Physics Foundation Models is available as arXiv:2511.12829 and was presented at the ML4PS workshop at NeurIPS 2025. The paper systematically compares contrastive, masked particle modeling, and generative objectives for particle physics foundation models within a unified framework. I served as a mentor on this work, led by Michael Chen.