Foundation models play an increasingly important role in physics. With @ozamram.bsky.social @joschkabirk.bsky.social @kasieczka.bsky.social et al. we have trained a FM on 180M particle physics events from the LHC and shown the performance in generating synthetic data: arxiv.org/abs/2412.10504.

Aspen Open Jets: Unlocking LHC Data for Foundation Models in Particle Physics

Foundation models are deep learning models pre-trained on large amounts of data which are capable of generalizing to multiple datasets and/or downstream tasks. This work demonstrates how data collecte...

arxiv.org

New paper on #arxiv, inspired by discussions this summer at the Aspen Center for Physics 🤔 Foundation models need huge amounts of training data. 🤔 There now exist large volumes of #OpenData by e.g. @cmsexperiment.bsky.social 💡 Why not use that data for training?