{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T05:07:59Z","timestamp":1778908079893,"version":"3.51.4"},"reference-count":67,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100003725","name":"National Research Foundation of Korea","doi-asserted-by":"publisher","award":["RS-2024-00451435 (15%)"],"award-info":[{"award-number":["RS-2024-00451435 (15%)"]}],"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-2024-00413957 (15%)"],"award-info":[{"award-number":["RS-2024-00413957 (15%)"]}],"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-2025-16072709 (15%)"],"award-info":[{"award-number":["RS-2025-16072709 (15%)"]}],"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-2026-25494040 (15%)"],"award-info":[{"award-number":["RS-2026-25494040 (15%)"]}],"id":[{"id":"10.13039\/501100003725","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Institute of Information and Communications Technology Planning and Evaluation"},{"name":"Ministry of Science and Information and Communications Technology (ICT)","award":["RS-2025-02305453 (10%)"],"award-info":[{"award-number":["RS-2025-02305453 (10%)"]}]},{"name":"Ministry of Science and Information and Communications Technology (ICT)","award":["RS-2025-02273157 (10%)"],"award-info":[{"award-number":["RS-2025-02273157 (10%)"]}]},{"name":"Ministry of Science and Information and Communications Technology (ICT)","award":["RS-2025-25442149 (10%)"],"award-info":[{"award-number":["RS-2025-25442149 (10%)"]}]},{"name":"Ministry of Science and Information and Communications Technology (ICT)","award":["RS-2021-II211343 (10%)"],"award-info":[{"award-number":["RS-2021-II211343 (10%)"]}]},{"name":"Institute of New Media and Communications"},{"name":"Brain Korea 21 (BK21) Fostering Outstanding Universities for Research (FOUR) Program of the Education, Artificial Intelligence Graduate School Program"},{"name":"Research Program for Future ICT Pioneers, Seoul National University, in 2026"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. on Image Process."],"published-print":{"date-parts":[[2026]]},"DOI":"10.1109\/tip.2026.3687070","type":"journal-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:54:23Z","timestamp":1777492463000},"page":"5092-5106","source":"Crossref","is-referenced-by-count":0,"title":["Bias Alleviation Through Network Pruning for Sparse and Debiased Models"],"prefix":"10.1109","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0270-2781","authenticated-orcid":false,"given":"Sangwoo","family":"Hong","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering, Konkuk University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-2818-4892","authenticated-orcid":false,"given":"Sehwan","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hyungjun","family":"Joo","sequence":"additional","affiliation":[{"name":"Samsung Electronics, Suwon-si, Gyeonggi-do, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hyeonggeun","family":"Han","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiyoon","family":"Shin","sequence":"additional","affiliation":[{"name":"College of AI Convergence, Korea Aerospace University, Goyang, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-2546-8884","authenticated-orcid":false,"given":"Yoav","family":"Wald","sequence":"additional","affiliation":[{"name":"Faculty of Data and Decision Sciences, Technion, Haifa, Israel"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6804-980X","authenticated-orcid":false,"given":"Jungwoo","family":"Lee","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"598","article-title":"Optimal brain damage","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"LeCun"},{"key":"ref2","first-page":"164","article-title":"Second order derivatives for network pruning: Optimal brain surgeon","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Hassibi"},{"key":"ref3","article-title":"The lottery ticket hypothesis: Finding sparse, trainable neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Frankle"},{"key":"ref4","article-title":"Pruning neural networks at initialization: Why are we missing the mark?","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Frankle"},{"key":"ref5","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","author":"Han","year":"2015","journal-title":"arXiv:1510.00149"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2020.03.034"},{"key":"ref7","article-title":"SNIP: Single-shot network pruning based on connection sensitivity","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Lee"},{"key":"ref8","first-page":"6682","article-title":"Proving the lottery ticket hypothesis: Pruning is all you need","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","volume":"119","author":"Malach"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01152"},{"key":"ref10","article-title":"Pruning convolutional neural networks for resource efficient transfer learning","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Molchanov","year":"2017"},{"key":"ref11","first-page":"6377","article-title":"Pruning neural networks without any data by iteratively conserving synaptic flow","volume-title":"Proc. Int. Conf. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Tanaka"},{"key":"ref12","article-title":"What do compressed deep neural networks forget?","author":"Hooker","year":"2020"},{"key":"ref13","article-title":"Characterising bias in compressed models","author":"Hooker","year":"2020","journal-title":"arXiv:2010.03058"},{"key":"ref14","first-page":"17652","article-title":"Pruning has a disparate impact on model accuracy","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Tran"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2374"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.findings-emnlp.282"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-37745-7_9"},{"key":"ref18","article-title":"Last layer re-training is sufficient for robustness to spurious correlations","volume-title":"Proc. 11st Int. Conf. Learn. Represent.","author":"Kirichenko"},{"key":"ref19","article-title":"Distributionally robust neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Sagawa"},{"key":"ref20","first-page":"25407","article-title":"Improving out-of-distribution robustness via selective augmentation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Yao"},{"key":"ref21","first-page":"26484","article-title":"Correct-n-contrast: A contrastive approach for improving robustness to spurious correlations","volume-title":"Proc. 39th Int. Conf. Mach. Learn.","volume":"162","author":"Zhang"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2733"},{"key":"ref23","first-page":"28448","article-title":"Simple and fast group robustness by automatic feature reweighting","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Qiu"},{"key":"ref24","first-page":"20673","article-title":"Learning from failure: De-biasing classifier from biased classifier","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Nam"},{"key":"ref25","article-title":"Mitigating dataset bias by using per-sample gradient","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Ahn"},{"key":"ref26","first-page":"6781","article-title":"Just train twice: Improving group robustness without training group information","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","volume":"139","author":"Liu"},{"key":"ref27","first-page":"336","article-title":"Simple data balancing achieves competitive worst-group-accuracy, \"Simple data balancing achieves competitive worst-group-accuracy","volume-title":"Proc. 1st Conf. Causal Learn. Reasoning","volume":"177","author":"Idrissi"},{"key":"ref28","first-page":"528","article-title":"Learning de-biased representations with biased representations","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","volume":"119","author":"Bahng"},{"key":"ref29","first-page":"25123","article-title":"Learning debiased representation via disentangled feature augmentation","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Lee"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00456"},{"key":"ref31","first-page":"26449","article-title":"Unbiased classification through bias-contrastive and bias-balanced learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Hong"},{"key":"ref32","first-page":"27203","article-title":"Model agnostic sample reweighting for out-of-distribution learning","volume-title":"Proc. 39th Int. Conf. Mach. Learn.","volume":"162","author":"Zhou"},{"key":"ref33","first-page":"19339","article-title":"No subclass left behind: Fine-grained robustness in coarse-grained classification problems","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Sohoni"},{"key":"ref34","first-page":"2189","article-title":"Environment inference for invariant learning","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","volume":"139","author":"Creager"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/72.248452"},{"key":"ref36","article-title":"Lookahead: A far-sighted alternative of magnitude-based pruning","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Park"},{"key":"ref37","first-page":"9908","article-title":"Sparse training via boosting pruning plasticity with neuroregeneration","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","volume":"34","author":"Liu"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevA.39.6600"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1080\/09540098908915626"},{"key":"ref40","first-page":"1135","article-title":"Learning both weights and connections for efficient neural network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"28","author":"Han"},{"key":"ref41","first-page":"18098","article-title":"WoodFisher: Efficient second-order approximation for neural network compression","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Singh"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.280"},{"key":"ref43","first-page":"4860","article-title":"Learning to prune deep neural networks via layer-wise optimal brain surgeon","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Dong"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00958"},{"key":"ref45","first-page":"6566","article-title":"EigenDamage: Structured pruning in the Kronecker-factored Eigenbasis","volume-title":"Proc. Int. Conf. Mach. Learn. (PMLR)","author":"Wang"},{"key":"ref46","first-page":"25668","article-title":"The combinatorial brain surgeon: Pruning weights that cancel one another in neural networks","volume-title":"Proc. 39th Int. Conf. Mach. Learn.","volume":"162","author":"Yu"},{"key":"ref47","first-page":"12356","article-title":"Can subnetwork structure be the key to out-of-distribution generalization?","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","volume":"139","author":"Zhang"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00766"},{"key":"ref49","article-title":"A fair loss function for network pruning","volume-title":"Proc. Workshop Trustworthy Socially Responsible Mach. Learn., NeurIPS","author":"Meyer"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19778-9_24"},{"key":"ref51","first-page":"19141","article-title":"Implicit Jacobian regularization weighted with impurity of probability output","volume-title":"Proc. 40th Int. Conf. Mach. Learn.","volume":"202","author":"Lee"},{"key":"ref52","article-title":"Empirical analysis of the Hessian of over-parametrized neural networks","author":"Sagun","year":"2018","journal-title":"arXiv:1706.04454"},{"key":"ref53","first-page":"6405","article-title":"Simple and scalable predictive uncertainty estimation using deep ensembles","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Lakshminarayanan"},{"key":"ref54","article-title":"Evaluating and improving robustness of self-supervised representations to spurious correlations","volume-title":"Proc. Workshop Spurious Correlations, Invariance Stability (ICML)","author":"Hamidieh"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00780"},{"key":"ref56","article-title":"Spurious feature diversification improves out-of-distribution generalization","volume-title":"Proc. The 12th Int. Conf. Learn. Represent.","author":"Yong"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.425"},{"key":"ref58","article-title":"MetaShift: A dataset of datasets for evaluating contextual distribution shifts and training conflicts","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Liang"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00264"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref61","first-page":"4175","article-title":"Balanced meta-softmax for long-tailed visual recognition","volume-title":"Proc. NIPS","author":"Ren"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00656"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00159"},{"key":"ref64","article-title":"Avoiding spurious correlations via logit correction","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Liu"},{"key":"ref65","first-page":"841","article-title":"RUBi: Reducing unimodal biases for visual question answering","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Cadene"},{"key":"ref66","article-title":"Unbiased supervised contrastive learning","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Barbano"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01501"}],"container-title":["IEEE Transactions on Image Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/83\/11355710\/11501194.pdf?arnumber=11501194","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T04:20:22Z","timestamp":1778905222000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11501194\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":67,"URL":"https:\/\/doi.org\/10.1109\/tip.2026.3687070","relation":{},"ISSN":["1057-7149","1941-0042"],"issn-type":[{"value":"1057-7149","type":"print"},{"value":"1941-0042","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}