arXiv:2601.17126 · v2
Foundation model for particle collision data analysis
摘要
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.
作者备注：19 pages, 5 figures, 1 table, 3 Extended Data tables, and 1 Extended Data figure. Published in Nature Machine Intelligence. Supplementary Information: https://static-content.springer.com/esm/art%3A10.1038%2Fs42256-026-01322-9/MediaObjects/42256_2026_1322_MOESM1_ESM.pdf
期刊信息：Nature Machine Intelligence (2026)