Community challenges

Major community-organized challenges in HEP–ML, with links to their summary papers. These are valuable not only as benchmarks but also as snapshots of the state of the art at a given moment.

Jet physics

• The Machine Learning Landscape of Top Taggers, Butter et al. (SciPost Phys. 7 (2019) 014, arXiv:1902.09914) — comparison of top-tagging architectures on a common dataset, widely used as a reference benchmark.

Anomaly detection

• The LHC Olympics 2020, Kasieczka et al. (Rept. Prog. Phys. 84 (2021) 124201, arXiv:2101.08320) — community challenge for anomaly detection at the LHC.
• The Dark Machines Anomaly Score Challenge, Aarrestad et al. (SciPost Phys. 12 (2022) 043, arXiv:2105.14027) — model-independent event classification challenge.

Fast simulation

• CaloChallenge 2022, Amram et al. (Rept. Prog. Phys. 88 (2025) 116201, arXiv:2410.21611) — community challenge for fast calorimeter simulation, comparing a wide range of generative architectures.

Note

To propose a challenge for inclusion, please open a pull request — see Contribute.