pipette
ESEspañol

Foundation model for particle collision data analysis

Ting-Hsiang Hsu, B. Zhou, K. Lie, B. Yabsley, Shu Li, W.-S. Hou, Benjamin Philip Nachman, S.-C. Hsu, V. M. Mikuni, M. C. Chu, Z. Zhang

Peer-reviewed journalBold claims, read critically

In the authors' words

Abstract Data-intensive collider experiments increasingly require models that transfer across high-dimensional, partially observed measurements, yet most analyses still optimize separate models for each task. Here, to address this bottleneck, we introduce EVENET, an event-level foundation model pretrained on 500 million simulated collision events using self-supervised learning and physics-informed supervision. The model constructs a shared particle-cloud representation from the variable-length set of physics objects in each event. Across searches for heavy resonances and exotic Higgs decays, EVENET outperforms both established and state-of-the-art baselines, particularly in low-statistics regimes. We validate transfer to experimental data by rediscovering the ϒ meson in CMS Open Data. The pretrained model also supports precise extraction of quantum-correlation observables that remain stable under the tested systematic variations. Together, these results establish a reusable backbone for diverse collider analyses, improving downstream adaptation after a substantial one-time pretraining investment.

Main resultThe abstract does not state a limitation.

Appeared: Saturday, October 31. Nature Machine Intelligence. Peer-reviewed journal.

DOI: 10.1038/s42256-026-01322-9