- Physical Review Journals
All JournalsPhysics Magazine

Physical Review D

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Letter
  • Open Access

Fundamental limit of jet tagging

Joep Geuskens1,*, Nishank Gite2,†, Michael Krämer1,‡, Vinicius Mikuni3,§, Alexander Mück1,∥, Benjamin Nachman4,5,¶, and Humberto Reyes-González1,**

  • *Contact author: joep.geuskens@rwth-aachen.de
  • †Contact author: nishankgite@berkeley.edu
  • ‡Contact author: mkraemer@physik.rwth-aachen.de
  • §Contact author: vmikuni@lbl.gov
  • ∥Contact author: mueck@physik.rwth-aachen.de
  • Contact author: bpnachman@lbl.gov
  • **Contact author: humberto.reyes@rwth-aachen.de
PDF

Phys. Rev. D 112, L091901 – Published 7 November, 2025

DOI: https://doi.org/10.1103/mjj1-w2b1

Abstract

Identifying the origin of high-energy hadronic jets (jet tagging) has been a critical benchmark problem for machine learning in particle physics. Jets are ubiquitous at colliders and are complex objects that serve as prototypical examples of collections of particles to be categorized. Over the last decade, machine learning-based classifiers have replaced classical observables as the state of the art in jet tagging. Increasingly complex machine learning models are leading to increasingly more effective tagger performance. Our goal is to address the question of convergence—are we getting close to the fundamental limit on jet tagging or is there still potential for computational, statistical, and physical insights for further improvements? We address this question using state-of-the-art generative models to create a realistic, synthetic dataset with a known jet tagging optimum. Various state-of-the-art taggers are deployed on this dataset, showing that there is a significant gap between their performance and the optimum. Our dataset and software are made public to provide a benchmark task for future developments in jet tagging and other areas of particle physics.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (35)

  1. A. J. Larkoski, I. Moult, and B. Nachman, Phys. Rep. 841, 1 (2020).
  2. R. Kogler et al., Rev. Mod. Phys. 91, 045003 (2019).
  3. A. Butter et al., SciPost Phys. 7, 014 (2019).
  4. A. M. Sirunyan et al. (CMS Collaboration), J. Instrum. 15, P06005 (2020).
  5. H. Qu and L. Gouskos, Phys. Rev. D 101, 056019 (2020).
  6. H. Qu, C. Li, and S. Qian, Proceedings of the 39th International Conference on Machine Learning, PMLR (2022), arXiv:2202.03772.
  7. G. Aad et al. (ATLAS Collaboration), J. Instrum. 19, P08018 (2024).
  8. J. Neyman and E. S. Pearson, Phil. Trans. R. Soc. A 231, 289 (1933).
  9. L. de Oliveira, M. Paganini, and B. Nachman, Comput. Software Big Sci. 1, 4 (2017).
  10. S. Badger et al., SciPost Phys. 14, 079 (2022).
  11. A. Adelmann et al., in 2022 Snowmass Summer Study (2022), arXiv:2203.08806.
  12. R. Verheyen, SciPost Phys. 13, 047 (2022).
  13. B. Käch, D. Krücker, I. Melzer-Pellmann, M. Scham, S. Schnake, and A. Verney-Provatas, arXiv:2211.13630.
  14. B. Käch, D. Krücker, and I. Melzer-Pellmann, arXiv:2211.13623.
  15. S. Schnake, D. Krücker, and K. Borras, arXiv:2403.15782.
  16. V. Mikuni and B. Nachman, SciPost Phys. 16, 062 (2024).
  17. T. Finke, M. Krämer, A. Mück, and J. Tönshoff, J. High Energy Phys. 06 (2023) 184.
  18. H. Qu, C. Li, and S. Qian, JetClass: A large-scale dataset for deep learning in jet physics (Zenodo, 2022), 10.5281/zenodo.6619768.
  19. M. Tanabashi et al. (Particle Data Group), Phys. Rev. D 98, 030001 (2018).
  20. J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Maltoni, O. Mattelaer, H. S. Shao, T. Stelzer, P. Torrielli, and M. Zaro, J. High Energy Phys. 07 (2014) 079.
  21. T. Sjöstrand, S. Mrenna, and P. Z. Skands, J. High Energy Phys. 05 (2006) 026.
  22. T. Sjöstrand, S. Ask, J. R. Christiansen, R. Corke, N. Desai, P. Ilten, S. Mrenna, S. Prestel, C. O. Rasmussen, and P. Z. Skands, Comput. Phys. Commun. 191, 159 (2015).
  23. J. de Favereau, C. Delaere, P. Demin, A. Giammanco, V. Lemaître, A. Mertens, and M. Selvaggi (DELPHES 3 Collaboration), J. High Energy Phys. 02 (2014) 057.
  24. A. Mertens, J. Phys. Conf. Ser. 608, 012045 (2015).
  25. M. Selvaggi, J. Phys. Conf. Ser. 523, 012033 (2014).
  26. M. Cacciari, G. P. Salam, and G. Soyez, Eur. Phys. J. C 72, 1896 (2012).
  27. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, arXiv:1706.03762.
  28. M. Zaheer, S. Kottur, S. Ravanbakhsh, B. Póczos, R. Salakhutdinov, and A. J. Smola, in Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, 2017, Long Beach, CA, USA (Curran Associates Inc., Red Hook, 2017), pp. 3391–3401.
  29. P. T. Komiske, E. M. Metodiev, and J. Thaler, J. High Energy Phys. 01 (2019) 121.
  30. S. Gong, Q. Meng, J. Zhang, H. Qu, C. Li, S. Qian, W. Du, Z.-M. Ma, and T.-Y. Liu, J. High Energy Phys. 07 (2022) 030.
  31. A. Bogatskiy, T. Hoffman, D. W. Miller, and J. T. Offermann, arXiv:2211.00454.
  32. V. Mikuni and B. Nachman, Phys. Rev. D 111, L051504 (2025).
  33. H. Reyes-González, J. Geuskens, N. Gite, M. Krämer, V. Mikuni, A. Mück, and B. Nachman, The fundamental limit of jet tagging dataset (Zenodo, 2024), 10.5281/zenodo.14023638.
  34. K. Cranmer, J. Pavez, and G. Louppe, arXiv:1506.02169.
  35. B. Nachman and J. Thaler, Phys. Rev. D 103, 116013 (2021).

Outline

Information

Phys. Rev. D 112, L091901– Published 7 November, 2025
  • Received 5 February 2025
  • Accepted 13 October 2025
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP3.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation