{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T03:42:12Z","timestamp":1773114132047,"version":"3.50.1"},"reference-count":56,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T00:00:00Z","timestamp":1764892800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T00:00:00Z","timestamp":1764892800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100001711","name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","doi-asserted-by":"crossref","award":["TMAG-2 209263"],"award-info":[{"award-number":["TMAG-2 209263"]}],"id":[{"id":"10.13039\/501100001711","id-type":"DOI","asserted-by":"crossref"}]},{"name":"National Science Foundation","award":["No. 2411204"],"award-info":[{"award-number":["No. 2411204"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,12,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The growing luminosity frontier at the Large Hadron Collider is challenging the reconstruction and analysis of particle collision events. Increased particle multiplicities are straining latency and storage requirements at the data acquisition stage, while new complications are emerging, including higher background levels and more frequent particle vertex misassociations. This in turn necessitates the development of more holistic and scalable reconstruction methods that take advantage of recent advances in machine learning. We propose a novel heterogeneous graph neural network (HGNN) architecture featuring unique representations for diverse particle collision relationships and integrated graph pruning layers for scalability. Trained with a multi-task paradigm in an environment mimicking the LHCb experiment, this HGNN significantly improves the beauty hadron reconstruction performance. Notably, it concurrently performs particle vertex association and graph pruning within a single framework. We quantify the reconstruction and pruning performance, demonstrate enhanced inference time scaling with event complexity, and mitigate potential performance loss using a weighted message passing scheme.<\/jats:p>","DOI":"10.1088\/2632-2153\/ae22be","type":"journal-article","created":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T22:54:11Z","timestamp":1763765651000},"page":"045060","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Scalable multi-task learning for particle collision event reconstruction with heterogeneous graph neural networks"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9795-3582","authenticated-orcid":true,"given":"William","family":"Sutcliffe","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8797-1357","authenticated-orcid":false,"given":"Marta","family":"Calvi","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8444-4498","authenticated-orcid":false,"given":"Simone","family":"Capelli","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7312-3699","authenticated-orcid":true,"given":"Jonas","family":"Eschle","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2316-8829","authenticated-orcid":false,"given":"Juli\u00e1n","family":"Garc\u00eda Pardi\u00f1as","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9428-4715","authenticated-orcid":false,"given":"Abhijit","family":"Mathad","sequence":"additional","affiliation":[]},{"given":"Azusa","family":"Uzuki","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5033-0580","authenticated-orcid":true,"given":"Nicola","family":"Serra","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,12,5]]},"reference":[{"key":"mlstae22bebib1","type":"conference-proceedings","article-title":"Imagenet classification with deep convolutional neural networks","volume":"vol 25","author":"Krizhevsky","year":"2012"},{"key":"mlstae22bebib2","article-title":"Attention is all you need","author":"Vaswani","year":"2017","type":"preprint"},{"key":"mlstae22bebib3","article-title":"A comprehensive survey on graph neural networks","author":"Wu","year":"2019","type":"preprint"},{"key":"mlstae22bebib4","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v37i9.26283","type":"preprint","article-title":"Simple and efficient heterogeneous graph neural network","author":"Yang","year":"2023"},{"key":"mlstae22bebib5","article-title":"Heterogeneous graph neural network for recommendation","author":"Shi","year":"2020","type":"preprint"},{"key":"mlstae22bebib6","article-title":"Heterogeneous causal metapath graph neural network for gene-microbe-disease association prediction","author":"Zhang","year":"2024","type":"preprint"},{"key":"mlstae22bebib7","first-page":"pp 612","type":"conference-proceedings","article-title":"Cross-lingual text classification with heterogeneous graph neural network","author":"Wang","year":"2021"},{"key":"mlstae22bebib8","article-title":"Multi-task learning with deep neural networks: a survey","author":"Crawshaw","year":"2020","type":"preprint"},{"key":"mlstae22bebib9","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1038\/s41586-018-0361-2","type":"journal-article","volume":"560","author":"Radovic","year":"2018","journal-title":"Nature"},{"key":"mlstae22bebib10","doi-asserted-by":"publisher","DOI":"10.1142\/S0217751X19300199","type":"journal-article","volume":"34","author":"Bourilkov","year":"2020","journal-title":"Int. J. Mod. Phys. A"},{"key":"mlstae22bebib11","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/18\/07\/P07001","type":"journal-article","volume":"18","author":"Huang","year":"2023","journal-title":"J. Instrum."},{"key":"mlstae22bebib12","doi-asserted-by":"publisher","DOI":"10.1051\/epjconf\/202429509028","type":"journal-article","volume":"295","author":"Caillou","year":"2024","journal-title":"EPJ Web Conf."},{"key":"mlstae22bebib13","article-title":"Graph neural networks in particle physics","author":"Shlomi","year":"2020","type":"preprint"},{"key":"mlstae22bebib14","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/3\/08\/S08005","type":"journal-article","volume":"3","author":"LHCb)","year":"2008","journal-title":"J. Instrum."},{"key":"mlstae22bebib15","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/19\/05\/P05065","type":"journal-article","volume":"19","author":"(LHCb)","year":"2024","journal-title":"J. Instrum."},{"key":"mlstae22bebib16","article-title":"Physics case for an LHCb upgrade II - opportunities in flavour physics, and beyond, in the HL-LHC era lHCB-PUB-2018-009, CERN-LHCC-2018-027","author":"(LHCb)","year":"2018","type":"preprint"},{"key":"mlstae22bebib17","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/14\/04\/P04006","type":"journal-article","volume":"14","author":"Aaij","year":"2019","journal-title":"J. Instrum."},{"key":"mlstae22bebib18","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/14\/04\/P04013","type":"journal-article","volume":"14","author":"LHCb)","year":"2019","journal-title":"J. Instrum."},{"key":"mlstae22bebib19","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.108.101803","type":"journal-article","volume":"108","author":"(LHCb)","year":"2012","journal-title":"Phys. Rev. Lett."},{"key":"mlstae22bebib20","doi-asserted-by":"publisher","first-page":"743","DOI":"10.1038\/nphys3415","type":"journal-article","volume":"11","author":"LHCb)","year":"2015","journal-title":"Nat. Phys."},{"key":"mlstae22bebib21","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.134.061801","type":"journal-article","volume":"134","author":"(LHCb)","year":"2025","journal-title":"Phys. Rev. Lett."},{"key":"mlstae22bebib22","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1007\/s41781-023-00107-8","type":"journal-article","volume":"7","author":"Garc\u00eda Pardinas","year":"2023","journal-title":"Comput. Softw. Big Sci."},{"key":"mlstae22bebib23","doi-asserted-by":"publisher","DOI":"10.1088\/1748-0221\/8\/02\/P02013","type":"journal-article","volume":"8","author":"Gligorov","year":"2013","journal-title":"J. Instrum."},{"key":"mlstae22bebib24","article-title":"Development of the topological trigger for LHCb run 3","author":"Schulte","year":"2023","type":"preprint"},{"key":"mlstae22bebib25","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1007\/s41781-019-0021-8","type":"journal-article","volume":"3","author":"Keck","year":"2019","journal-title":"Comput. Softw. Big Sci."},{"key":"mlstae22bebib26","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ac8de0","type":"journal-article","volume":"3","author":"Kahn","year":"2022","journal-title":"Mach. Learn. Sci. Tech."},{"key":"mlstae22bebib27","author":"Abumusabh","year":"2025","type":"preprint"},{"key":"mlstae22bebib28","doi-asserted-by":"publisher","DOI":"10.3389\/fdata.2022.828666","type":"journal-article","volume":"5","author":"Elabd","year":"2022","journal-title":"Front. Big Data"},{"key":"mlstae22bebib29","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.110.012001","type":"journal-article","volume":"110","author":"(Belle II","year":"2024","journal-title":"Phys. Rev. D"},{"key":"mlstae22bebib30","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.101.056019","type":"journal-article","volume":"101","author":"Qu","year":"2020","journal-title":"Phys. Rev. D"},{"key":"mlstae22bebib31","article-title":"Graph neural network jet flavour tagging with the ATLAS detector","author":"ATLAS","year":"2022","type":"other"},{"key":"mlstae22bebib32","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.103.116025","type":"journal-article","volume":"103","author":"Guo","year":"2021","journal-title":"Phys. Rev. D"},{"key":"mlstae22bebib33","doi-asserted-by":"publisher","first-page":"381","DOI":"10.1140\/epjc\/s10052-021-09158-w","type":"journal-article","volume":"81","author":"Pata","year":"2021","journal-title":"Eur. Phys. J. C"},{"key":"mlstae22bebib34","doi-asserted-by":"publisher","first-page":"876","DOI":"10.1140\/epjc\/s10052-021-09675-8","type":"journal-article","volume":"81","author":"(Exa.TrkX)","year":"2021","journal-title":"Eur. Phys. J. C"},{"key":"mlstae22bebib35","article-title":"Hierarchical graph neural networks for particle track reconstruction","author":"Liu","year":"2023","type":"preprint"},{"key":"mlstae22bebib36","doi-asserted-by":"publisher","first-page":"608","DOI":"10.1140\/epjc\/s10052-019-7113-9","type":"journal-article","volume":"79","author":"Qasim","year":"2019","journal-title":"Eur. Phys. J. C"},{"key":"mlstae22bebib37","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ad66ad","type":"journal-article","volume":"5","author":"Birk","year":"2024","journal-title":"Mach. Learn. Sci. Tech."},{"key":"mlstae22bebib38","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ad64a8","type":"journal-article","volume":"5","author":"Golling","year":"2024","journal-title":"Mach. Learn. Sci. Tech."},{"key":"mlstae22bebib39","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.111.054015","type":"journal-article","volume":"111","author":"Mikuni","year":"2025","journal-title":"Phys. Rev. D"},{"key":"mlstae22bebib40","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.110.032008","type":"journal-article","volume":"110","author":"Aurisano","year":"2024","journal-title":"Phys. Rev. D"},{"key":"mlstae22bebib41","doi-asserted-by":"publisher","DOI":"10.1088\/1741-4326\/ad8014","type":"journal-article","volume":"64","author":"Zheng","year":"2024","journal-title":"Nucl. Fusion"},{"key":"mlstae22bebib42","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2017","type":"conference-proceedings"},{"key":"mlstae22bebib43","article-title":"Graph attention networks","author":"Veli\u010dkovi\u0107","year":"2018","type":"conference-proceedings"},{"key":"mlstae22bebib44","first-page":"pp 129","type":"conference-proceedings","article-title":"Neural message passing for quantum chemistry","volume":"vol 30","author":"Gilmer","year":"2017"},{"key":"mlstae22bebib45","article-title":"Relational inductive biases, deep learning, and graph networks","author":"Battaglia","year":"2018","type":"preprint"},{"key":"mlstae22bebib46","first-page":"pp 3538","type":"conference-proceedings","article-title":"Deeper insights into graph convolutional networks for semi-supervised learning","author":"Li","year":"2018"},{"key":"mlstae22bebib47","article-title":"Pytorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019","type":"preprint"},{"key":"mlstae22bebib48","article-title":"Code of paper scalable multi-task learning for particle collision event reconstruction with heterogeneous graph neural networks","author":"Sutcliffe","type":"other"},{"key":"mlstae22bebib49","author":"Kingma","year":"2015","type":"preprint"},{"key":"mlstae22bebib50","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.7799170","type":"other","article-title":"Dataset of paper \u2018GNN for deep full event interpretation and hierarchical reconstruction of heavy-hadron decays in proton-proton collisions\u2019","author":"Garc\u00eda Pardinas","year":"2023"},{"key":"mlstae22bebib51","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.15584745","type":"other","article-title":"Dataset of paper \u2018scalable multi-task learning for particle collision event reconstruction with heterogeneous graph neural networks\u2019","author":"Sutcliffe","year":"2025"},{"key":"mlstae22bebib52","doi-asserted-by":"crossref","DOI":"10.21468\/SciPostPhysCodeb.8","type":"preprint","article-title":"A comprehensive guide to the physics and usage of PYTHIA 8.3","author":"Bierlich","year":"2022"},{"key":"mlstae22bebib53","article-title":"EvtGen: a Monte Carlo generator for B-physics","author":"Ryd","year":"2005","type":"other"},{"key":"mlstae22bebib54","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2021.108026","type":"journal-article","volume":"265","author":"Billoir","year":"2021","journal-title":"Comput. Phys. Commun."},{"key":"mlstae22bebib55","article-title":"2014 LHCb tracker upgrade technical design report","type":"other"},{"key":"mlstae22bebib56","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.110.030001","type":"journal-article","volume":"110","author":"(Particle Data Group)","year":"2024","journal-title":"Phys. Rev. D"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T10:38:46Z","timestamp":1764931126000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae22be"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,5]]},"references-count":56,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,12,5]]},"published-print":{"date-parts":[[2025,12,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ae22be","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,5]]},"assertion":[{"value":"Scalable multi-task learning for particle collision event reconstruction with heterogeneous graph neural networks","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-07-10","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-11-21","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-12-05","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}