{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T19:22:17Z","timestamp":1776885737743,"version":"3.51.2"},"reference-count":71,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2025,2,13]],"date-time":"2025-02-13T00:00:00Z","timestamp":1739404800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Recommendation systems (RS) have become integral to numerous digital platforms and applications, ranging from e-commerce to content streaming field. A critical problem in RS is that the ratings are missing not at random (MNAR), which is due to the users always giving feedback on items with self-selection. The biased selection of rating data results in inaccurate rating prediction for all user-item pairs. Doubly robust (DR) learning has been studied in many tasks in RS, which is unbiased when either a single imputation or a single propensity model is accurate. In addition, multiple robust (MR) has been proposed with multiple imputation models and propensity models, and is unbiased when there exists a linear combination of these imputation models and propensity models is correct. However, we claim that the imputed errors and propensity scores are miscalibrated in the MR method. In this paper, we propose a gradient-based calibrated multiple robust learning method to enhance the debiasing performance and reliability of the rating prediction model. Specifically, we propose to use bi-level optimization to solve the weights and model coefficients of each propensity and imputation model in MR framework. Moreover, we adopt the differentiable expected calibration error as part of the objective to optimize the model calibration quality directly. Experiments on three real-world datasets show that our method outperforms the state-of-the-art baselines.<\/jats:p>","DOI":"10.3390\/e27020196","type":"journal-article","created":{"date-parts":[[2025,2,13]],"date-time":"2025-02-13T09:05:42Z","timestamp":1739437542000},"page":"196","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Gradient-Based Multiple Robust Learning Calibration on Data Missing-Not-at-Random via Bi-Level Optimization"],"prefix":"10.3390","volume":"27","author":[{"given":"Shuxia","family":"Gong","sequence":"first","affiliation":[{"name":"Mogo Co., Ltd., Beijing 100000, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chen","family":"Ma","sequence":"additional","affiliation":[{"name":"Gaoling School of Artificial Intelligence, Renmin University of China, Beijing 100872, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,2,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Ricci, F., Rokach, L., and Shapira, B. (2010). Introduction to recommender systems handbook. Recommender Systems Handbook, Springer.","DOI":"10.1007\/978-0-387-85820-3"},{"key":"ref_2","unstructured":"Wang, X., Zhang, R., Sun, Y., and Qi, J. (2019, January 9\u201315). Doubly Robust Joint Learning for Recommendation on Data Missing Not at Random. Proceedings of the ICML, Long Beach, CA, USA."},{"key":"ref_3","unstructured":"Schnabel, T., Swaminathan, A., Singh, A., Chandak, N., and Joachims, T. (2016, January 19\u201324). Recommendations as Treatments: Debiasing Learning and Evaluation. Proceedings of the ICML, New York, NY, USA."},{"key":"ref_4","first-page":"1","article-title":"Bias and Debias in Recommender System: A Survey and Future Directions","volume":"41","author":"Chen","year":"2022","journal-title":"Acm Trans. Inf. Syst."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Wu, P., Li, H., Deng, Y., Hu, W., Dai, Q., Dong, Z., Sun, J., Zhang, R., and Zhou, X.H. (2022, January 23\u201329). On the Opportunity of Causal Learning in Recommendation Systems: Foundation, Estimation, Prediction and Challenges. Proceedings of the IJCAI, Vienna, Austria.","DOI":"10.24963\/ijcai.2022\/787"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Steck, H. (2010, January 25\u201328). Training and testing of recommender systems on data missing not at random. Proceedings of the KDD, Washington, DC, USA.","DOI":"10.1145\/1835804.1835895"},{"key":"ref_7","first-page":"1471","article-title":"Training and testing low-degree polynomial data mappings via linear SVM","volume":"11","author":"Chang","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Saito, Y., Yaginuma, S., Nishino, Y., Sakata, H., and Nakata, K. (2020, January 3\u20137). Unbiased recommender learning from missing-not-at-random implicit feedback. Proceedings of the WSDM, Houston, TX, USA.","DOI":"10.1145\/3336191.3371783"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Morgan, S.L., and Winship, C. (2015). Counterfactuals and Causal Inference: Methods and Principles for Social Research, Cambridge University Press. [2nd ed.].","DOI":"10.1017\/CBO9781107587991"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Saito, Y. (2020, January 22\u201326). Doubly robust estimator for ranking metrics with post-click conversions. Proceedings of the RecSys, Virtual Event.","DOI":"10.1145\/3383313.3412262"},{"key":"ref_11","unstructured":"Li, H., Dai, Q., Li, Y., Lyu, Y., Dong, Z., Zhou, X.H., and Wu, P. (2023, January 7\u201314). Multiple Robust Learning for Recommendation. Proceedings of the AAAI, Washington, DC, USA."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Kweon, W., and Yu, H. (2024, January 13\u201317). Doubly Calibrated Estimator for Recommendation on Data Missing Not At Random. Proceedings of the WWW, Singapore.","DOI":"10.1145\/3589334.3645617"},{"key":"ref_13","first-page":"100590","article-title":"A survey on causal inference for recommendation","volume":"5","author":"Luo","year":"2024","journal-title":"Innovation"},{"key":"ref_14","unstructured":"Li, M., and Sui, H. (March, January 25). Debiased Recommendation via Machine Unlearning. Proceedings of the AAAI Workshop on Artificial Intelligence with Causal Techniques, Philadelphia, PA, USA."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Wang, W., Zhang, Y., Li, H., Wu, P., Feng, F., and He, X. (2023, January 23\u201327). Causal Recommendation: Progresses and Future Directions. Proceedings of the SIGIR, Taipei, Taiwan.","DOI":"10.1145\/3539618.3594245"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Saito, Y., and Nomura, M. (2022, January 23\u201329). Towards Resolving Propensity Contradiction in Offline Recommender Learning. Proceedings of the IJCAI, Vienna, Austria.","DOI":"10.24963\/ijcai.2022\/307"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Wang, H., Yang, W., Yang, L., Wu, A., Xu, L., Ren, J., Wu, F., and Kuang, K. (2022, January 14\u201318). Estimating Individualized Causal Effect with Confounded Instruments. Proceedings of the KDD, Washington, DC, USA.","DOI":"10.1145\/3534678.3539335"},{"key":"ref_18","unstructured":"Zou, H., Wang, H., Xu, R., Li, B., Pei, J., Jian, Y.J., and Cui, P. (2023, January 11\u201314). Factual Observation Based Heterogeneity Learning for Counterfactual Prediction. Proceedings of the CCLR, T\u00fcbingen, Germany."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1758","DOI":"10.1109\/TKDE.2023.3312255","article-title":"Out-of-distribution generalization with causal feature separation","volume":"36","author":"Wang","year":"2024","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Wang, H., Kuang, K., Chi, H., Yang, L., Geng, M., Huang, W., and Yang, W. (2023, January 6\u201310). Treatment effect estimation with adjustment feature selection. Proceedings of the KDD, Long Beach, CA, USA.","DOI":"10.1145\/3580305.3599531"},{"key":"ref_21","unstructured":"Wu, A., Kuang, K., Xiong, R., Li, B., and Wu, F. (2023, January 23\u201329). Stable estimation of heterogeneous treatment effects. Proceedings of the ICML, Honolulu, HI, USA."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Saito, Y. (2020, January 25\u201330). Asymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedback. Proceedings of the SIGIR, Xi\u2019an, China.","DOI":"10.1145\/3397271.3401114"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Wang, H., Chang, T.W., Liu, T., Huang, J., Chen, Z., Yu, C., Li, R., and Chu, W. (2022, January 11\u201315). Escm2: Entire space counterfactual multi-task model for post-click conversion rate estimation. Proceedings of the SIGIR, Madrid, Spain.","DOI":"10.1145\/3477495.3531972"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Zhang, W., Bao, W., Liu, X.Y., Yang, K., Lin, Q., Wen, H., and Ramezani, R. (2020, January 20\u201324). Large-scale Causal Approaches to Debiasing Post-click Conversion Rate Estimation with Multi-task Learning. Proceedings of the WWW, Taipei, Taiwan.","DOI":"10.1145\/3366423.3380037"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Ding, S., Wu, P., Feng, F., He, X., Wang, Y., Liao, Y., and Zhang, Y. (2022, January 14\u201318). Addressing Unmeasured Confounder for Recommendation with Sensitivity Analysis. Proceedings of the KDD, Washington, DC, USA.","DOI":"10.1145\/3534678.3539240"},{"key":"ref_26","unstructured":"Li, H., Zheng, C., and Wu, P. (2023, January 1\u20135). StableDR: Stabilized Doubly Robust Learning for Recommendation on Data Missing Not at Random. Proceedings of the ICLR, Kigali, Rwanda."},{"key":"ref_27","unstructured":"Li, H., Lyu, Y., Zheng, C., and Wu, P. (2023, January 1\u20135). TDR-CL: Targeted Doubly Robust Collaborative Learning for Debiased Recommendations. Proceedings of the ICLR, Kigali, Rwanda."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Song, Z., Chen, J., Zhou, S., Shi, Q., Feng, Y., Chen, C., and Wang, C. (2023, January 21\u201325). CDR: Conservative Doubly Robust Learning for Debiased Recommendation. Proceedings of the CIKM, Birmingham, UK.","DOI":"10.1145\/3583780.3614805"},{"key":"ref_29","unstructured":"Li, H., Zheng, C., Ding, S., Feng, F., He, X., Geng, Z., and Wu, P. (2024, January 7\u201311). Be Aware of the Neighborhood Effect: Modeling Selection Bias under Interference for Recommendation. Proceedings of the ICLR, Vienna, Austria."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Zhang, H., Wang, S., Li, H., Zheng, C., Chen, X., Liu, L., Luo, S., and Wu, P. (2024, January 13\u201317). Uncovering the Propensity Identification Problem in Debiased Recommendations. Proceedings of the ICDE, Utrecht, The Netherlands.","DOI":"10.1109\/ICDE60146.2024.00056"},{"key":"ref_31","unstructured":"Li, H., Zheng, C., Wang, S., Wu, K., Wang, E., Wu, P., Geng, Z., Chen, X., and Zhou, X.H. (2024, January 21\u201327). Relaxing the Accurate Imputation Assumption in Doubly Robust Learning for Debiased Collaborative Filtering. Proceedings of the ICML, Vienna, Austria."},{"key":"ref_32","unstructured":"Li, H., Zheng, C., Xiao, Y., Wu, P., Geng, Z., Chen, X., and Cui, P. (2024, January 7\u201311). Debiased collaborative filtering with kernel-based causal balancing. Proceedings of the ICLR, Vienna, Austria."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3568030","article-title":"Debiased representation learning in recommendation via information bottleneck","volume":"1","author":"Liu","year":"2023","journal-title":"ACM Trans. Recomm. Syst."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Yang, M., Dai, Q., Dong, Z., Chen, X., He, X., and Wang, J. (2021, January 1\u20135). Top-n recommendation with counterfactual user preference simulation. Proceedings of the CIKM, Virtual Event.","DOI":"10.1145\/3459637.3482305"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Wang, J., Li, H., Zhang, C., Liang, D., Yu, E., Ou, W., and Wang, W. (2023, January 1\u20134). Counterclr: Counterfactual contrastive learning with non-random missing data in recommendation. Proceedings of the ICDM, Shanghai, China.","DOI":"10.1109\/ICDM58522.2023.00174"},{"key":"ref_36","unstructured":"Guo, C., Pleiss, G., Sun, Y., and Weinberger, K.Q. (2017, January 6\u201311). On calibration of modern neural networks. Proceedings of the ICML, Sydney, NSW, Australia."},{"key":"ref_37","unstructured":"Kull, M., Silva Filho, T., and Flach, P. (2017, January 20\u201322). Beta calibration: A well-founded and easily implemented improvement on logistic calibration for binary classifiers. Proceedings of the AISTATS, Fort Lauderdale, FL, USA."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Caruana, R., Lou, Y., Gehrke, J., Koch, P., Sturm, M., and Elhadad, N. (2015, January 10\u201313). Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. Proceedings of the KDD, Sydney, NSW, Australia.","DOI":"10.1145\/2783258.2788613"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"621","DOI":"10.1093\/jamia\/ocz228","article-title":"A tutorial on calibration measurements and calibration models for clinical prediction models","volume":"27","author":"Huang","year":"2020","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"ref_40","unstructured":"Bojarski, M. (2016). End to end learning for self-driving cars. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Chen, Z., and Huang, X. (2017, January 11\u201314). End-to-end learning for lane keeping of self-driving cars. Proceedings of the 2017 IEEE Intelligent Vehicles Symposium (IV), Los Angeles, CA, USA.","DOI":"10.1109\/IVS.2017.7995975"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"109","DOI":"10.1007\/s11147-021-09183-7","article-title":"Deep calibration of financial models: Turning theory into practice","volume":"25","author":"Kratochwil","year":"2022","journal-title":"Rev. Deriv. Res."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"734","DOI":"10.1137\/23M1606769","article-title":"Approximation rates for deep calibration of (rough) stochastic volatility models","volume":"15","author":"Biagini","year":"2024","journal-title":"SIAM J. Financ. Math."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Niculescu-Mizil, A., and Caruana, R. (2005, January 7\u201311). Predicting good probabilities with supervised learning. Proceedings of the ICML, Bonn, Germany.","DOI":"10.1145\/1102351.1102430"},{"key":"ref_45","unstructured":"Wang, C. (2023). Calibration in deep learning: A survey of the state-of-the-art. arXiv."},{"key":"ref_46","unstructured":"Zadrozny, B., and Elkan, C. (July, January 28). Obtaining calibrated probability estimates from decision trees and naive bayesian classifiers. Proceedings of the ICML, Williamstown, MA, USA."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Zadrozny, B., and Elkan, C. (2002, January 23\u201326). Transforming classifier scores into accurate multiclass probability estimates. Proceedings of the KDD, Edmonton, AB, Canada.","DOI":"10.1145\/775107.775151"},{"key":"ref_48","first-page":"61","article-title":"Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods","volume":"10","author":"Platt","year":"1999","journal-title":"Adv. Large Margin Classif."},{"key":"ref_49","unstructured":"Pereyra, G., Tucker, G., Chorowski, J., Kaiser, \u0141., and Hinton, G. (2017). Regularizing neural networks by penalizing confident output distributions. arXiv."},{"key":"ref_50","unstructured":"Liang, G., Zhang, Y., Wang, X., and Jacobs, N. (2020). Improved trainable calibration method for neural networks on medical imaging classification. arXiv."},{"key":"ref_51","unstructured":"Kumar, A., Sarawagi, S., and Jain, U. (2018, January 10\u201315). Trainable calibration measures for neural networks from kernel mean embeddings. Proceedings of the ICML, Stockholm, Sweden."},{"key":"ref_52","unstructured":"Bohdal, O., Yang, Y., and Hospedales, T. (2025, January 15). Meta-Calibration: Learning of Model Calibration Using Differentiable Expected Calibration Error. Trans. Mach. Learn. Res., Available online: https:\/\/openreview.net\/forum?id=R2hUure38l."},{"key":"ref_53","unstructured":"Blundell, C., Cornebise, J., Kavukcuoglu, K., and Wierstra, D. (2015, January 6\u201311). Weight uncertainty in neural network. Proceedings of the ICML, Lille, France."},{"key":"ref_54","unstructured":"Lakshminarayanan, B., Pritzel, A., and Blundell, C. (2017, January 4\u20139). Simple and scalable predictive uncertainty estimation using deep ensembles. Proceedings of the NeurIPS, Long Beach, CA, USA."},{"key":"ref_55","unstructured":"Gal, Y., and Ghahramani, Z. (2016, January 19\u201324). Dropout as a bayesian approximation: Representing model uncertainty in deep learning. Proceedings of the ICML, New York City, NY, USA."},{"key":"ref_56","unstructured":"Jang, E., Gu, S., and Poole, B. (2017, January 24\u201326). Categorical Reparametrization with Gumble-Softmax. Proceedings of the ICLR, Toulon, France."},{"key":"ref_57","unstructured":"Zhang, J., Kailkhura, B., and Han, T.Y.J. (2020, January 13\u201318). Mix-n-match: Ensemble and compositional methods for uncertainty calibration in deep learning. Proceedings of the ICML, Virtual Event."},{"key":"ref_58","unstructured":"Laves, M.H., Ihler, S., Kortmann, K.P., and Ortmaier, T. (2019). Well-calibrated model uncertainty with temperature scaling for dropout variational inference. arXiv."},{"key":"ref_59","unstructured":"Naeini, M.P., Cooper, G., and Hauskrecht, M. (2015, January 25\u201330). Obtaining well calibrated probabilities using bayesian binning. Proceedings of the AAAI, Austin, TX, USA."},{"key":"ref_60","doi-asserted-by":"crossref","unstructured":"Gao, C., Li, S., Lei, W., Chen, J., Li, B., Jiang, P., He, X., Mao, J., and Chua, T.S. (2022, January 17\u201321). KuaiRec: A Fully-observed Dataset and Insights for Evaluating Recommender Systems. Proceedings of the CIKM, Atlanta, GA, USA.","DOI":"10.1145\/3511808.3557220"},{"key":"ref_61","doi-asserted-by":"crossref","unstructured":"Chen, J., Dong, H., Qiu, Y., He, X., Xin, X., Chen, L., Lin, G., and Yang, K. (2021, January 11\u201315). AutoDebias: Learning to Debias for Recommendation. Proceedings of the SIGIR, Online.","DOI":"10.1145\/3404835.3462919"},{"key":"ref_62","unstructured":"Li, H., Wu, K., Zheng, C., Xiao, Y., Wang, H., Geng, Z., Feng, F., He, X., and Wu, P. (2024, January 10\u201315). Removing Hidden Confounding in Recommendation: A Unified Multi-Task Learning Approach. Proceedings of the NeurIPS, Vancouver, BC, Canada."},{"key":"ref_63","unstructured":"Li, H., Xiao, Y., Zheng, C., and Wu, P. (May, January 30). Balancing Unobserved Confounding with a Few Unbiased Ratings in Debiased Recommendations. Proceedings of the WWW, Austin, TX, USA."},{"key":"ref_64","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1109\/MC.2009.263","article-title":"Matrix factorization techniques for recommender systems","volume":"42","author":"Koren","year":"2009","journal-title":"Computer"},{"key":"ref_65","unstructured":"Swaminathan, A., and Joachims, T. (2015, January 7\u201312). The Self-Normalized Estimator for Counterfactual Learning. Proceedings of the NeurIPS, Montreal, QC, Canada."},{"key":"ref_66","doi-asserted-by":"crossref","unstructured":"Guo, S., Zou, L., Liu, Y., Ye, W., Cheng, S., Wang, S., Chen, H., Yin, D., and Chang, Y. (2021, January 11\u201315). Enhanced Doubly Robust Learning for Debiasing Post-Click Conversion Rate Estimation. Proceedings of the SIGIR, Online.","DOI":"10.1145\/3404835.3462917"},{"key":"ref_67","doi-asserted-by":"crossref","unstructured":"Dai, Q., Li, H., Wu, P., Dong, Z., Zhou, X.H., Zhang, R., Zhang, R., and Sun, J. (2022, January 14\u201318). A generalized doubly robust learning framework for debiasing post-click conversion rate prediction. Proceedings of the KDD, Washington, DC, USA.","DOI":"10.1145\/3534678.3539270"},{"key":"ref_68","unstructured":"Li, H., Xiao, Y., Zheng, C., Wu, P., and Cui, P. (2023, January 23\u201329). Propensity Matters: Measuring and Enhancing Balancing for Recommendation. Proceedings of the ICML, Honolulu, HI, USA."},{"key":"ref_69","doi-asserted-by":"crossref","first-page":"1145","DOI":"10.1016\/S0031-3203(96)00142-2","article-title":"The use of the area under the ROC curve in the evaluation of machine learning algorithms","volume":"30","author":"Bradley","year":"1997","journal-title":"Pattern Recognit."},{"key":"ref_70","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1145\/582415.582418","article-title":"Cumulated gain-based evaluation of IR techniques","volume":"20","year":"2002","journal-title":"ACM Trans. Inf. Syst. (TOIS)"},{"key":"ref_71","doi-asserted-by":"crossref","unstructured":"Lipton, Z.C., Elkan, C., and Naryanaswamy, B. (2014, January 15\u201319). Optimal thresholding of classifiers to maximize F1 measure. Proceedings of the Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2014, Nancy, France. Proceedings, Part II 14.","DOI":"10.1007\/978-3-662-44851-9_15"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/2\/196\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T16:33:07Z","timestamp":1760027587000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/2\/196"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,2,13]]},"references-count":71,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,2]]}},"alternative-id":["e27020196"],"URL":"https:\/\/doi.org\/10.3390\/e27020196","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,2,13]]}}}