{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T17:49:57Z","timestamp":1773942597560,"version":"3.50.1"},"reference-count":47,"publisher":"Wiley","issue":"12","license":[{"start":{"date-parts":[[2024,9,30]],"date-time":"2024-09-30T00:00:00Z","timestamp":1727654400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62106116"],"award-info":[{"award-number":["62106116"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["advanced.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Advanced Intelligent Systems"],"published-print":{"date-parts":[[2024,12]]},"abstract":"<jats:p>Explainable artificial intelligence (XAI) aims to develop transparent explanatory approaches for \u201cblack\u2010box\u201d deep learning models. However, it remains difficult for existing methods to achieve the trade\u2010off of the three key criteria in interpretability, namely, reliability, understandability, and usability, which hinder their practical applications. In this article, we propose a self\u2010supervised automatic semantic interpretable explainable artificial intelligence (AS\u2010XAI) framework, which utilizes transparent orthogonal embedding semantic extraction spaces and row\u2010centered principal component analysis (PCA) for global semantic interpretation of model decisions in the absence of human interference, without additional computational costs. In addition, the invariance of filter feature high\u2010rank decomposition is used to evaluate model sensitivity to different semantic concepts. Extensive experiments demonstrate that robust and orthogonal semantic spaces can be automatically extracted by AS\u2010XAI, providing more effective global interpretability for convolutional neural networks (CNNs) and generating human\u2010comprehensible explanations. The proposed approach offers broad fine\u2010grained extensible practical applications, including shared semantic interpretation under out\u2010of\u2010distribution (OOD) categories, auxiliary explanations for species that are challenging to distinguish, and classification explanations from various perspectives. In a systematic evaluation by users with varying levels of AI knowledge, AS\u2010XAI demonstrated superior \u201cglass box\u201d characteristics.<\/jats:p>","DOI":"10.1002\/aisy.202400359","type":"journal-article","created":{"date-parts":[[2024,9,30]],"date-time":"2024-09-30T13:13:37Z","timestamp":1727702017000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["AS\u2010XAI: Self\u2010Supervised Automatic Semantic Interpretation for CNN"],"prefix":"10.1002","volume":"6","author":[{"given":"Changqi","family":"Sun","sequence":"first","affiliation":[{"name":"School of Environmental Science and Engineering Southern University of Science and Technology  Shenzhen 518055 Guangdong P. R. China"},{"name":"Department of Mathematics and Theories Peng Cheng Laboratory  Shenzhen 518000 Guangdong P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao","family":"Xu","sequence":"additional","affiliation":[{"name":"BIC\u2010ESAT, ERE, and SKLTCS, College of Engineering Peking University  Beijing 100871 P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuntian","family":"Chen","sequence":"additional","affiliation":[{"name":"Ningbo Institute of Digital Twin Eastern Institute of Technology  Ningbo Zhejiang 315200 P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6930-5994","authenticated-orcid":false,"given":"Dongxiao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Mathematics and Theories Peng Cheng Laboratory  Shenzhen 518000 Guangdong P. R. China"},{"name":"Ningbo Institute of Digital Twin Eastern Institute of Technology  Ningbo Zhejiang 315200 P. R. China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2024,9,30]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1155\/2018\/7068349"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-81-322-3972-7_19"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aax0162"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1002\/rob.21918"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-023-00698-2"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2019.12.012"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2019.102003"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1038\/ncomms13890"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.artint.2018.07.007"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1037\/0022-3514.79.1.5"},{"key":"e_1_2_9_13_1","unstructured":"N.Garau N.Bisagno Z.Sambugaro N.Conci in2022 IEEECVF Conf. Comput. Vis. Pattern Recognit.CVPR IEEE New Orleans LA 2022 pp.13689\u201313698."},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2020.3027314"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2301842120"},{"key":"e_1_2_9_16_1","unstructured":"L. M.Zintgraf T. S.Cohen T.Adel M.Welling arXiv preprint arXiv:1702.045952017."},{"key":"e_1_2_9_17_1","unstructured":"A.Shrikumar P.Greenside A.Kundaje inProc. 34th Int. Conf. Mach. Learn. PMLR New York NY 2017pp.3145\u20133153."},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0130140"},{"key":"e_1_2_9_19_1","doi-asserted-by":"crossref","unstructured":"M. D.Zeiler R.Fergus inComput. Vision\u2013ECCV 2014: 13th Eur. Conf. Springer Zurich2014 pp.818\u2013833.","DOI":"10.1007\/978-3-319-10590-1_53"},{"key":"e_1_2_9_20_1","unstructured":"M.Sundararajan A.Taly Q.Yan inProc. 34th Int. Conf. Mach. Learn. PMLR New York NY 2017 3319\u20133328."},{"key":"e_1_2_9_21_1","unstructured":"B.Zhou A.Khosla A.Lapedriza A.Oliva A.Torralba in2016 IEEECVF Conf. Comput. Vis. Pattern Recognit. CVPR IEEE Las Vegas NV 2016 pp.2921\u20132929."},{"key":"e_1_2_9_22_1","doi-asserted-by":"crossref","unstructured":"R. R.Selvaraju M.Cogswell A.Das R.Vedantam D.Parikh D.Batra inProc. IEEE Int. Conf. Comput. Vision.ICCV IEEE Venice Italy 2017 pp.618\u2013626.","DOI":"10.1109\/ICCV.2017.74"},{"key":"e_1_2_9_23_1","unstructured":"A.Chattopadhyay A.Sarkar P.Howlader V. N.Balasubramanian in2018 IEEE Winter Conf. Appl Comput. Vision (WACV) Lake Tahoe NV IEEE2018."},{"key":"e_1_2_9_24_1","unstructured":"D.Bau B.Zhou A.Khosla A.Oliva A.Torralba in2017 IEEECVF Conf. Comput. Vis. Pattern Recognit.CVPR IEEE Honolulu HI 2017 pp.6541\u20136549."},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-00265-z"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-019-0048-x"},{"key":"e_1_2_9_27_1","first-page":"32","volume-title":"Advances in Neural Information Processing Systems","author":"Chen C.","year":"2019"},{"key":"e_1_2_9_28_1","unstructured":"J.Wang H.Liu X.Wang L.Jing inProc. IEEE\/CVF Int. Conf. Comput. Vision.ICCV IEEE Montreal Canada 2021 pp.895\u2013904."},{"key":"e_1_2_9_29_1","unstructured":"J.Donnelly A. J.Barnett C.Chen in2016 IEEECVF Conf. Comput. Vis. Pattern Recognit. CVPR IEEE New Orleans LA 2022 pp.10265\u201310275."},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1002\/advs.202204723"},{"key":"e_1_2_9_31_1","first-page":"30","volume-title":"Advances in Neural Information Processing Systems","author":"Lundberg S. M.","year":"2017"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-023-00711-8"},{"key":"e_1_2_9_33_1","unstructured":"K.Simonyan A.Zisserman arXiv preprint arXiv:1409.15562014."},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/0168-9002(95)01478-0"},{"key":"e_1_2_9_35_1","unstructured":"D. P.Kingma J.Ba arXiv preprint arXiv:1412.69802014."},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2011.282"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1098\/rsta.2015.0202"},{"key":"e_1_2_9_38_1","first-page":"13","volume-title":"Advances in Neural Information Processing Systems","author":"Lee D.","year":"2000"},{"key":"e_1_2_9_39_1","unstructured":"M.Lin R.Ji Y.Wang Y.Zhang B.Zhang Y.Tian L.Shao in2020 IEEECVF Conf. Comput. Vis. Pattern Recognit. CVPR IEEE Seattle WA 2020 pp.1529\u20131538."},{"key":"e_1_2_9_40_1","unstructured":"M.Graziani L. O.Mahony A.\u2010P.Nguyen H.M\u00fcller V.Andrearczyk arXiv preprint arXiv:2307.069132023."},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2857768"},{"key":"e_1_2_9_42_1","doi-asserted-by":"crossref","unstructured":"M.\u2010E.Nilsback A.Zisserman in2008 6th Indian Conf. Comput. Vision Graph. Image Process. IEEE Indian 2008 pp.722\u2013729.","DOI":"10.1109\/ICVGIP.2008.47"},{"key":"e_1_2_9_43_1","unstructured":"J.Yang K.Zhou Y.Li Z.Liu arXiv preprint arXiv:2110.113342021."},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.isci.2021.103581"},{"key":"e_1_2_9_45_1","unstructured":"R.Geirhos P.Rubisch C.Michaelis M.Bethge F. A.Wichmann W.Brendel arXiv preprint arXiv:1811.122312018."},{"key":"e_1_2_9_46_1","doi-asserted-by":"crossref","unstructured":"Y.Yue T.Finley F.Radlinski T.Joachims inProc. 30th Annu. Int. ACM SIGIR Conf. Res. Dev. Inf. Retrieval ACM Amsterdam 2007 pp.271\u2013278.","DOI":"10.1145\/1277741.1277790"},{"key":"e_1_2_9_47_1","unstructured":"B.Kim M.Wattenberg J.Gilmer C.Cai J.Wexler F.Viegas inInt. Conf. Mach. Learn.PMLR New York NY 2018 pp.2668\u20132677."},{"key":"e_1_2_9_48_1","first-page":"343","author":"Hacker P.","year":"2021","journal-title":"SSRN Electron. J."}],"container-title":["Advanced Intelligent Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/aisy.202400359","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,7]],"date-time":"2025-10-07T14:31:23Z","timestamp":1759847483000},"score":1,"resource":{"primary":{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/10.1002\/aisy.202400359"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,30]]},"references-count":47,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2024,12]]}},"alternative-id":["10.1002\/aisy.202400359"],"URL":"https:\/\/doi.org\/10.1002\/aisy.202400359","archive":["Portico"],"relation":{},"ISSN":["2640-4567","2640-4567"],"issn-type":[{"value":"2640-4567","type":"print"},{"value":"2640-4567","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,9,30]]},"assertion":[{"value":"2024-05-06","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-09-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"2400359"}}