{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T12:10:02Z","timestamp":1779451802587,"version":"3.53.1"},"reference-count":47,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100002701","name":"Ministry of Education","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100002701","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS-2025-25399600"],"award-info":[{"award-number":["RS-2025-25399600"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS-2024-00410940"],"award-info":[{"award-number":["RS-2024-00410940"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Knowledge-Based Systems"],"published-print":{"date-parts":[[2026,5]]},"DOI":"10.1016\/j.knosys.2026.115821","type":"journal-article","created":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T07:53:34Z","timestamp":1774338814000},"page":"115821","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Energy-ensemble concept bottleneck models for enhancing interpretability and accuracy in concept-based learning"],"prefix":"10.1016","volume":"341","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-4123-072X","authenticated-orcid":false,"given":"Dasom","family":"Ahn","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7452-3897","authenticated-orcid":false,"given":"Sangwon","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7284-0768","authenticated-orcid":false,"given":"Byoung Chul","family":"Ko","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.knosys.2026.115821_bib0001","series-title":"Proceedings of the International Conference on Machine Learning (ICML)","first-page":"5338","article-title":"Concept bottleneck models","author":"Koh","year":"2020"},{"key":"10.1016\/j.knosys.2026.115821_bib0002","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"21400","article-title":"Concept embedding models: beyond the accuracy-explainability trade-off","author":"Espinosa Zarlenga","year":"2022"},{"key":"10.1016\/j.knosys.2026.115821_bib0003","series-title":"Proceedings of the International Conference on Learning Representations (ICLR)","first-page":"1","article-title":"Post-hoc concept bottleneck models","author":"Yuksekgonul","year":"2023"},{"key":"10.1016\/j.knosys.2026.115821_bib0004","series-title":"Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)","first-page":"5948","article-title":"Interactive concept bottleneck models","author":"Chauhan","year":"2023"},{"key":"10.1016\/j.knosys.2026.115821_bib0005","series-title":"Proceedings of the International Conference on Machine Learning (ICML)","first-page":"16521","article-title":"Probabilistic concept bottleneck models","author":"Kim","year":"2023"},{"key":"10.1016\/j.knosys.2026.115821_bib0006","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"10286","article-title":"A framework for learning ante-hoc explainable models via concepts","author":"Sarkar","year":"2022"},{"key":"10.1016\/j.knosys.2026.115821_bib0007","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"23386","article-title":"Addressing leakage in concept bottleneck models","author":"Havasi","year":"2022"},{"key":"10.1016\/j.knosys.2026.115821_bib0008","series-title":"Proceedings of the Asian Conference on Computer Vision (ACCV)","first-page":"3432","article-title":"EQ-CBM: a probabilistic concept bottleneck with energy-based models and quantized vectors","author":"Kim","year":"2024"},{"key":"10.1016\/j.knosys.2026.115821_bib0009","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","article-title":"Coarse-to-fine concept bottleneck models","author":"Panousis","year":"2024"},{"key":"10.1016\/j.knosys.2026.115821_bib0010","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","article-title":"Beyond concept bottleneck models: how to make black boxes intervenable?","author":"Laguna","year":"2024"},{"issue":"3","key":"10.1016\/j.knosys.2026.115821_bib0011","doi-asserted-by":"crossref","DOI":"10.1088\/2632-2153\/ad6ad2","article-title":"Concept graph embedding models for enhanced accuracy and interpretability","volume":"5","author":"Kim","year":"2024","journal-title":"Mach. Learn. Sci. Technol."},{"key":"10.1016\/j.knosys.2026.115821_bib0012","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"1","article-title":"Auxiliary losses for learning generalizable concept-based models","author":"Sheth","year":"2023"},{"key":"10.1016\/j.knosys.2026.115821_bib0013","series-title":"Proceedings of the International Conference on Learning Representations (ICLR)","first-page":"1","article-title":"Energy-based concept bottleneck models","author":"Xu","year":"2024"},{"issue":"0","key":"10.1016\/j.knosys.2026.115821_bib0014","first-page":"1","article-title":"A tutorial on energy-based learning","volume":"1","author":"LeCun","year":"2006","journal-title":"Predicting Struct. Data"},{"key":"10.1016\/j.knosys.2026.115821_bib0015","series-title":"Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)","first-page":"5272","article-title":"On the anatomy of mcmc-based maximum likelihood learning of energy-based models","author":"Nijkamp","year":"2020"},{"key":"10.1016\/j.knosys.2026.115821_bib0016","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"361","article-title":"Learning non-convergent non-persistent short-run mcmc toward energy-based model","author":"Nijkamp","year":"2019"},{"key":"10.1016\/j.knosys.2026.115821_bib0017","series-title":"Proceedings of the European Conference on Computer Vision (ECCV)","first-page":"361","article-title":"Learning multi-layer latent variable model via variational optimization of short run mcmc for approximate inference","author":"Nijkamp","year":"2020"},{"issue":"1","key":"10.1016\/j.knosys.2026.115821_bib0018","first-page":"1","article-title":"Alternating back-propagation for generator network","volume":"31","author":"Han","year":"2017","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"issue":"1","key":"10.1016\/j.knosys.2026.115821_bib0019","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1006\/jmva.1994.1033","article-title":"Two-sample test statistics for measuring discrepancies between two multivariate probability density functions using kernel-based density estimates","volume":"50","author":"Anderson","year":"1994","journal-title":"J. Multivar. Anal."},{"issue":"1","key":"10.1016\/j.knosys.2026.115821_bib0020","first-page":"723","article-title":"A kernel two-sample test","volume":"13","author":"Gretton","year":"2012","journal-title":"J. Mach. Learn. Res."},{"key":"10.1016\/j.knosys.2026.115821_bib0021","series-title":"Technical Report","article-title":"Caltech-UCSD Birds 200","author":"Welinder","year":"2010"},{"issue":"9","key":"10.1016\/j.knosys.2026.115821_bib0022","doi-asserted-by":"crossref","first-page":"2251","DOI":"10.1109\/TPAMI.2018.2857768","article-title":"Zero-shot learning-a comprehensive evaluation of the good, the bad and the ugly","volume":"41","author":"Xian","year":"2018","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.knosys.2026.115821_bib0023","series-title":"Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)","first-page":"590","article-title":"CheXpert: a large chest radiograph dataset with uncertainty labels and expert comparison","author":"Irvin","year":"2019"},{"key":"10.1016\/j.knosys.2026.115821_bib0024","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.1016\/j.knosys.2026.115821_bib0025","series-title":"Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)","first-page":"1","article-title":"Inception-v4, inception-resnet and the impact of residual connections on learning","author":"Szegedy","year":"2017"},{"key":"10.1016\/j.knosys.2026.115821_bib0026","series-title":"Proceedings of the International Conference on Learning Representations (ICLR)","first-page":"1","article-title":"Label-free concept bottleneck models","author":"Oikarinen","year":"2022"},{"key":"10.1016\/j.knosys.2026.115821_bib0027","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"19187","article-title":"Language in a bottle: language model guided concept bottlenecks for interpretable image classification","author":"Yang","year":"2023"},{"key":"10.1016\/j.knosys.2026.115821_bib0028","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"11030","article-title":"Incremental residual concept bottleneck models","author":"Shang","year":"2024"},{"issue":"2","key":"10.1016\/j.knosys.2026.115821_bib0029","doi-asserted-by":"crossref","first-page":"1250","DOI":"10.1109\/TPAMI.2024.3489597","article-title":"The decoupling concept bottleneck model","volume":"47","author":"Zhang","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"10.1016\/j.knosys.2026.115821_bib0030","first-page":"3401","article-title":"V2C-CBM: building concept bottlenecks with vision-to-concept tokenizer","volume":"39","author":"He","year":"2025","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"10.1016\/j.knosys.2026.115821_bib0031","series-title":"Proceedings of the Computer Vision and Pattern Recognition Conference","first-page":"20179","article-title":"Hybrid concept bottleneck models","author":"Liu","year":"2025"},{"issue":"1","key":"10.1016\/j.knosys.2026.115821_bib0032","first-page":"147","article-title":"A learning algorithm for Boltzmann machines","volume":"9","author":"Ackley","year":"1985","journal-title":"Cogn. Sci."},{"issue":"7","key":"10.1016\/j.knosys.2026.115821_bib0033","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1162\/neco.2006.18.7.1527","article-title":"A fast learning algorithm for deep belief nets","volume":"18","author":"Hinton","year":"2006","journal-title":"Neural Comput."},{"key":"10.1016\/j.knosys.2026.115821_bib0034","series-title":"Artificial Intelligence and Statistics","first-page":"448","article-title":"Deep boltzmann machines","author":"Salakhutdinov","year":"2009"},{"issue":"11","key":"10.1016\/j.knosys.2026.115821_bib0035","first-page":"2","article-title":"MCMC using Hamiltonian dynamics","volume":"2","author":"Neal","year":"2011","journal-title":"Handb. Markov Chain Monte Carlo"},{"key":"10.1016\/j.knosys.2026.115821_bib0036","series-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV)","first-page":"847","article-title":"Grade: gibbs reaction and diffusion equations","author":"Zhu","year":"1998"},{"key":"10.1016\/j.knosys.2026.115821_bib0037","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"9155","article-title":"Learning generative convnets via multi-grid modeling and sampling","author":"Gao","year":"2018"},{"key":"10.1016\/j.knosys.2026.115821_bib0038","series-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV)","first-page":"22952","article-title":"EGC: image generation and classification via a diffusion energy-based model","author":"Guo","year":"2023"},{"key":"10.1016\/j.knosys.2026.115821_bib0039","series-title":"Proceedings of the International Conference on Learning Representations (ICLR)","first-page":"1","article-title":"Energy-based generative adversarial network","author":"Zhao","year":"2017"},{"key":"10.1016\/j.knosys.2026.115821_bib0040","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"15608","article-title":"Unsupervised learning of compositional energy concepts","author":"Du","year":"2021"},{"key":"10.1016\/j.knosys.2026.115821_bib0041","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"21994","article-title":"Learning latent space energy-based prior model","author":"Pang","year":"2020"},{"key":"10.1016\/j.knosys.2026.115821_bib0042","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"1","article-title":"Implicit generation and modeling with energy based models","author":"Du","year":"2019"},{"key":"10.1016\/j.knosys.2026.115821_bib0043","series-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","first-page":"7978","article-title":"Joint training of variational auto-encoder and latent energy-based model","author":"Han","year":"2020"},{"key":"10.1016\/j.knosys.2026.115821_bib0044","series-title":"Proceedings of the International Conference on Learning Representations (ICLR)","first-page":"1","article-title":"Your classifier is secretly an energy based model and you should treat it like one","author":"Grathwohl","year":"2020"},{"key":"10.1016\/j.knosys.2026.115821_bib0045","series-title":"Advances in Neural Information Processing Systems (NeurIPS)","first-page":"4358","article-title":"Energy-based contrastive learning of visual representations","author":"Kim","year":"2022"},{"key":"10.1016\/j.knosys.2026.115821_bib0046","series-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV)","first-page":"6494","article-title":"JEM++: improved techniques for training JEM","author":"Yang","year":"2021"},{"issue":"11","key":"10.1016\/j.knosys.2026.115821_bib0047","first-page":"1","article-title":"Visualizing data using t-SNE","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"J. Mach. Learn. Res."}],"container-title":["Knowledge-Based Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0950705126005472?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0950705126005472?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,5,22]],"date-time":"2026-05-22T11:41:46Z","timestamp":1779450106000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0950705126005472"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":47,"alternative-id":["S0950705126005472"],"URL":"https:\/\/doi.org\/10.1016\/j.knosys.2026.115821","relation":{},"ISSN":["0950-7051"],"issn-type":[{"value":"0950-7051","type":"print"}],"subject":[],"published":{"date-parts":[[2026,5]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Energy-ensemble concept bottleneck models for enhancing interpretability and accuracy in concept-based learning","name":"articletitle","label":"Article Title"},{"value":"Knowledge-Based Systems","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.knosys.2026.115821","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"115821"}}