The DOI matches its Crossref record, with a small difference in title or year; 47 of the 48 references it cites were found in a registry or an index.
Popularity Bias is not Always Evil: Disentangling Benign and Harmful Bias for Recommendation
Open accessAbstract
Recommender system usually suffers from severepopularity bias— the collected interaction data usually exhibits quite imbalanced or even long-tailed distribution over items. Such skewed distribution may result from the users’conformityto the group, which deviates from reflecting users’ true preference. Existing efforts for tackling this issue mainly focus on completely eliminating popularity bias. However, we argue that not all popularity bias is evil. Popularity bias not only results from conformity but alsoitem quality, which is usually ignored by existing methods. Some items exhibit higher popularity as they have intrinsic better property. Blindly removing the popularity bias would lose such important signal, and further deteriorate model performance. To sufficiently exploit such important information for recommendation, it is essential to disentangle the benign popularity bias caused by item quality from the harmful popularity bias caused by conformity. Although important, it is quite challenging as we lack an explicit signal to differentiate the two factors of popularity bias. In this paper, we propose to leverage temporal information as the two factors exhibit quite different patterns along the time: item quality revealing item inherent property is stable and static while conformity that depends on items’ recent clicks is highly time-sensitive. Correspondingly, we further propose a novelTime-awareDisEntangled framework (TIDE), where a click is generated from three components namely the static item quality, the dynamic conformity effect, as well as the user-item matching score returned by any recommendation model. Lastly, we conduct interventional inference so that the recommendation can benefit from the benign popularity bias while circumvent the harmful one. Extensive experiments on four real-world datasets demonstrated the effectiveness of TIDE.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“An appropriate proportion of popularity bias in the system can improve performance and user satisfaction [183].”
From A Deep Dive into Fairness, Bias, Threats, and Privacy in Recommender Systems: Insights and Future Research (2024) · cites this paper as [183]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Moreover, popularity bias can act as a double-edged sword [182, 183].”
From A Deep Dive into Fairness, Bias, Threats, and Privacy in Recommender Systems: Insights and Future Research (2024) · cites this paper as [183]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Auto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067
Neural Collaborative Filtering
He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T. (2017) · International World Wide Web Conference (WWW) · doi:10.1145/3038912.3052569 · Cited by 7,385 · cited 1×
“experiments, and it would be straightforward to replace it with more sophisticated models such as Factorization Machine [15], or Neural Network [5], [16].”
MethodSelf-citationBPR: Bayesian Personalized Ranking from Implicit FeedbackPreprint
Rendle, S., Freudenthaler, C., Gantner, Z., & Schmidt-Thieme, L. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1205.2618 · Cited by 6,168
He, X., Deng, K., Wang, X., Li, Y., Zhang, Y., & Wang, M. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401063 · Cited by 5,300 · cited 4×
“Amazon-CDs is increased to 8,192 for speed and the number of convolutional layers is searched in {2, 3, 4} as advised by [5].”
BackgroundMethodInfluentialSelf-citationPearson Correlation Coefficient
Benesty, J., Chen, J., Huang, Y. A., & Cohen, I. (2009) · Springer topics in signal processing · doi:10.1007/978-3-642-00296-0_5 · Cited by 3,540
Self-citationThe long tail: why the future of business is selling less of more
(2007) · Choice Reviews Online · doi:10.5860/choice.44-2783 · Cited by 2,869
Deep Interest Network for Click-Through Rate Prediction
Zhou, G., Zhu, X., Song, C., Fan, Y., Zhu, H., Ma, X., Yan, Y., Jin, J., Li, H., & Gai, K. (2018) · Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining · doi:10.1145/3219819.3219823 · Cited by 2,128 · cited 2×
“(3) ( U, I ) → M → Y : these links project user and item features ( e.g., IDs) into their matching scores m ui = f θ ( u, i ) . f θ ( u, i ) can be implemented by various recommendation models, such as MF [4], LightGCN [5], DIN [6], etc. Finally these three components are aggregated into a final prediction score for r…”
BackgroundFactorization Machines with libFM
Rendle, S. (2012) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/2168752.2168771 · Cited by 1,345 · cited 1×
“Matrix Factorization (MF) has been selected as the main backbone recommendation model for experiments, and it would be straightforward to replace it with more sophisticated models such as Factorization Machine [15], or Neural Network [5], [16].”
MethodBias and Debias in Recommender System: A Survey and Future Directions
Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., & He, X. (2023) · ACM Transactions on Information Systems · doi:10.1145/3564284 · Cited by 741
Self-citationDisentangled Graph Collaborative Filtering
Wang, X., Jin, H., Zhang, A., He, X., Xu, T., & Chua, T.-S. (2020) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3397271.3401137 · Cited by 737
Self-citation
Flagged first, then by citations
1 of the 48 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Bias and Debias in Recommender System: A Survey and Future Directions
Chen, J., Dong, H., Wang, X., Feng, F., Wang, M., & He, X. (2023) · ACM Transactions on Information Systems · doi:10.1145/3564284 · Cited by 741
“Zheng et al. [217] propose to model personalized conformity effect as users have different sensitivities to the public opinions; while Zhao et al. [215] model time-aware conformity effect by considering item dynamic popularity.”
BackgroundSelf-citationA survey on popularity bias in recommender systems
Klimashevskaia, A., Jannach, D., Elahi, M., & Trattner, C. (2024) · User Modeling and User-Adapted Interaction · doi:10.1007/s11257-024-09406-0 · Cited by 223
“[148], for example, found that “the observation data usually exhibits”
BackgroundA survey of sequential recommendation systems: Techniques, evaluation, and future directions
Boka, T. F., Niu, Z., & Neupane, R. B. (2024) · Information Systems · doi:10.1016/j.is.2024.102427 · Cited by 98
Distributionally Robust Graph-based Recommendation System
Wang, B., Chen, J., Li, C., Zhou, S., Shi, Q., Gao, Y., Feng, Y., Chen, C., & Wang, C. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645598 · Cited by 45
Self-citationLalor, J. P., Abbasi, A., Oketch, K., Yang, Y., & Forsgren, N. (2024) · ACM Transactions on Information Systems · doi:10.1145/3641276 · Cited by 40
“For instance, the recommender systems literature has explored the implications of user behavior biases on future model recommendations/performance for many years—namely, the effect of data/sampling biases pertaining to selection, position, exposure, and popularity [34, 64, 75, 108].”
BackgroundAdaptive Popularity Debiasing Aggregator for Graph Collaborative Filtering
Zhou, H., Chen, H., Dong, J., Zha, D., Zhou, C. E., & Huang, X. X. (2023) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3539618.3591635 · Cited by 33
“Causal graphs can also be applied to understand and mitigate bias [32, 36, 48, 51, 53].”
BackgroundLong tail service recommendation based on cross-view and contrastive learning
Yu, D., Yu, T., Wang, D., et al. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121957 · Cited by 32
Cui, Y., Liu, F., Wang, P., Wang, B., Tang, H., Wan, Y., Wang, J., & Chen, J. (2024) · ACM Conference on Recommender Systems (RecSys) · doi:10.1145/3640457.3688118 · Cited by 54
Self-citationHow graph convolutions amplify popularity bias for recommendation?
Chen, J., Wu, J., Chen, J., Xin, X., Li, Y., & He, X. (2023) · Frontiers of Computer Science · doi:10.1007/s11704-023-2655-2 · Cited by 30
“Directly eliminating all the bias is not reasonable in recommendation, which conforms with the finding of [21].”
Self-citationA review of causality-based fairness machine learning
Su, C., Yu, G., Wang, J., Yan, Z., & Cui, L. (2022) · Intelligence & Robotics · doi:10.20517/ir.2022.17 · Cited by 28
Collaborative Tag-Aware Graph Neural Network for Long-Tail Service Recommendation
Zhang, Z., Zhang, Y., Dong, M., et al. (2024) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2024.3349853 · Cited by 24
How Do Recommendation Models Amplify Popularity Bias? An Analysis from the Spectral Perspective
Lin, S., Gao, C., Chen, J., Zhou, S., Hu, B., Feng, Y., Chen, C., & Wang, C. (2025) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3701551.3703579 · Cited by 29
“This implies that ¯ v tends to gravitate towards the origin.”
Self-citationCDR: Conservative Doubly Robust Learning for Debiased Recommendation
Song, Z., Chen, J., Zhou, S., Shi, Q., Feng, Y., Chen, C., & Wang, C. (2023) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3583780.3614805 · Cited by 23
“There are various sources of bias found in RS data, such as selection bias [7, 34, 35], exposure bias [8, 9, 29], conformity bias [30, 43], position bias [22, 23] and popularity bias [1, 10, 48, 54].”
BackgroundSelf-citationInteract with the Explanations: Causal Debiased Explainable Recommendation System
Liu, X., Yu, T., Xie, K., Wu, J., & Li, S. (2024) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3616855.3635855 · Cited by 20
“[52]analyzebiasinrecommendationsystemsfromaconformityef-fectperspectiveandproposecorrespondingcausalinterventions onpre-defineditempopularitytoenablefairrecommendations.”
BackgroundCausal Distillation for Alleviating Performance Heterogeneity in Recommender Systems
Zhang, S., Jiang, Z., Yao, J., Feng, F., Kuang, K., Zhao, Z., Li, S., Yang, H., Chua, T., & Wu, F. Q. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2023.3290545 · Cited by 18
“One representative line of research to alleviate these biases is to explicitly model or utilize these biases [30].”
BackgroundSelf-citationEnhancing Recommendation with Search Data in a Causal Learning Manner
Si, Z., Sun, Z., Zhang, X., Xu, J., Song, Y., Zang, X., & Wen, J.-R. (2023) · ACM Transactions on Information Systems · doi:10.1145/3582425 · Cited by 17
LLM4DSR: Leveraging Large Language Model for Denoising Sequential Recommendation
Wang, B., Liu, F., Zhang, C., Chen, J., Wu, Y., Zhou, S., Lou, X., Wang, J., Feng, Y., Chen, C., et al. (2025) · ACM Transactions on Information Systems · doi:10.1145/3762182 · Cited by 46
Self-citationGeneral Debiasing for Graph-based Collaborative Filtering via Adversarial Graph Dropout
Zhang, A., Ma, W., Wei, P., Sheng, L., & Wang, X. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645667 · Cited by 15
Relation pruning and discriminative sampling over knowledge graph for long-tail recommendation
Zhang, Z., Wang, A., Zhang, Y., et al. (2024) · Information Sciences · doi:10.1016/j.ins.2024.120871 · Cited by 15
Transfer learning for collaborative recommendation with biased and unbiased data
Lin, Z., Liu, D., Pan, W., et al. (2023) · Artificial Intelligence · doi:10.1016/j.artint.2023.103992 · Cited by 14
Simulating News Recommendation Ecosystems for Insights and Implications
Zhang, G., Li, D., Gu, H., Lu, T., Shang, L., & Gu, N. (2024) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2024.3381329 · Cited by 18
“Regarding the relationship between popularity and quality, Zhao et al. [16] and Ciampaglia et al. [17] pointed out that under certain conditions, high quality leads to high popularity.”
BackgroundA general tail item representation enhancement framework for sequential recommendation
Cheng, M., Liu, Q., Zhang, W., Liu, Z., Zhao, H., & Chen, E. (2023) · Frontiers of Computer Science · doi:10.1007/s11704-023-3112-y · Cited by 13
Cai, M., Hou, M., Chen, L., et al. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3664931 · Cited by 13
Path-Specific Causal Reasoning for Fairness-aware Cognitive Diagnosis
Zhang, D., Zhang, K., Wu, L., Tian, M., Hong, R., & Wang, M. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3672049 · Cited by 12
Debiased Recommendation with User Feature Balancing
Yang, M., Cai, G., Liu, F., Jin, J., Dong, Z., He, X., Hao, J., Shao, W., Wang, Y.-X., & Chen, X. (2023) · ACM Transactions on Information Systems · doi:10.1145/3580594 · Cited by 13
“Traditionally, there are two types of strategies to build debiased recommender models [5, 7, 8, 42, 51, 51].”
BackgroundSelf-citationToward Bias-Agnostic Recommender Systems: A Universal Generative Framework
Wang, Z., Zou, L., Li, C., Wang, S., Chen, X., Yin, D., & Liu, W. (2024) · ACM Transactions on Information Systems · doi:10.1145/3655617 · Cited by 11
“— TIDE [63]: A state-of-the-art causal inference method that only removes the harmful effect of popularity bias while leveraging its benign part to make recommendations.”
Dual disentanglement of user–item interaction for recommendation with causal embedding
Wang, C., Ye, Y., Ma, L., Li, D., & Zhuang, L. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2023.103456 · Cited by 11
A Novel Popularity Extraction Method Applied in Session-Based Recommendation
Peng, Y., Xu, S., Chen, Q., Huang, W., & Huang, Y. (2024) · Tsinghua Science & Technology · doi:10.26599/tst.2023.9010061 · Cited by 10
Cao, T., Xu, Q., Yang, Z., & Huang, Q. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3327411 · Cited by 9
“Furthermore, [26] and [27] propose methods to preserve useful information when causally intervening the item popularity bias.”
MethodTowards a Causal Decision-Making Framework for Recommender Systems
Cavenaghi, E., Zanga, A., Stella, F. A., & Zanker, M. (2023) · ACM Transactions on Recommender Systems · doi:10.1145/3629169 · Cited by 12
Reproducing Popularity Bias in Recommendation: The Effect of Evaluation Strategies
Daniil, S., Cuper, M., Liem, C. C. S., van Ossenbruggen, J., & Laura Hollink, A. (2023) · ACM Transactions on Recommender Systems · doi:10.1145/3637066 · Cited by 9
“However, very popular items are often more likely to be already known, which renders the recommendation of mostly popular items to a user potentially not useful for the user and the system in general [1].”
BackgroundMultimodal recommender systems: A survey of representation, modeling, and optimization
Pan, Z., Pan, Z., Cai, F., & Chen, H. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103991 · Cited by 13
Multi-Cause Deconfounding for Recommender Systems with Latent Confounders
Zhi-rong, H., Hu, Y., Cheng, D., et al. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114345 · Cited by 8
Self-Adaptive Deep Asymmetric Network for Imbalanced Recommendation
Zhu, Y. Q., Geng, Y., Li, J., et al. (2023) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2023.3300740 · Cited by 8
“Zhao et al. proposed to leverage temporal information to distinguish benign and harmful popularity bias [41], and the final interventional inference is conducted in the recommendation for benefiting from the benign popularity bias.”
BackgroundCapturing Popularity Trends: A Simplistic Non-Personalized Approach for Enhanced Item Recommendation
Jing, J., Zhang, Y., Zhou, X., & Shen, Z. (2023) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3583780.3614801 · Cited by 9
“As a result, recent studies have been focusing on reducing popularity bias while maintaining the models’ performances [70, 71, 77].”
BackgroundWang, X., Rahmani, H. A., Liu, J., & Yılmaz, E. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.233 · Cited by 19
Modeling Dynamic Item Tendency Bias in Sequential Recommendation With Causal Intervention
Liao, Z., Wu, X., Shang, S., Wang, J., & Zhang, W. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3427719 · Cited by 8
Deconfounding User Preference in Recommendation Systems through Implicit and Explicit Feedback
Liang, Y., Yang, E., Guo, G., et al. (2024) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3673762 · Cited by 7
“However, historical interaction data may be confounded by various factors, resulting in biases, such as selection bias [15, 23], popularity bias [28, 29], and exposure bias [1, 26].”
Invariant debiasing learning for recommendation via biased imputation
Bai, T., Chen, W., Yang, C., & Shi, C. (2024) · Information Processing & Management · doi:10.1016/j.ipm.2024.104028 · Cited by 9
Popularity Debiasing from Exposure to Interaction in Collaborative Filtering
Liu, Y., Cao, Q., Shen, H., Wu, Y., Tao, S., & Cheng, X. (2023) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3539618.3591947 · Cited by 8
Pan, Q., Li, S., Zheng, T., Ma, F., & Huang, J. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.123227 · Cited by 7
Rankformer: A Graph Transformer for Recommendation based on Ranking Objective
Chen, S., Shen, H., Chen, J., et al. (2025) · doi:10.1145/3696410.3714547 · Cited by 5
Self-citationCoherence-guided Preference Disentanglement for Cross-domain Recommendations
Xiang, Z., Zhang, Y., Duan, L., Yin, H., & Tsang, I. W.-H. (2025) · ACM Transactions on Information Systems · doi:10.1145/3742855 · Cited by 6
“Therefore, CoPD achieve significant performance improvement compared with UniCDR in all CDR tasks.”
MSL: Not All Tokens Are What You Need for Tuning LLM as a Recommender
Wang, B., Liu, F., Chen, J., Lou, X., Zhang, C., Wang, J., Sun, Y., Feng, Y., Chen, C., & Wang, C. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730041 · Cited by 5
Self-citationTaming Recommendation Bias with Causal Intervention on Evolving Personal Popularity
Tan, S., Li, D., Jiang, R., et al. (2025) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3711896.3737146 · Cited by 5
Causal inference for alleviating confounding bias in multi-criteria rating recommendation
Guo, Z., Song, P., Feng, C., Yao, K., & Liang, J. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104364 · Cited by 5
CaDRec: Contextualized and Debiased Recommender Model
Wang, X., Fukumoto, F., Cui, J., Suzuki, Y., Li, J., & Yu, D. (2024) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3626772.3657799 · Cited by 5
Causal Inference and Recommendations
Zhu, Y., Ma, J., & Li, J. (2023) · Springer eBooks · doi:10.1007/978-3-031-35051-1_10 · Cited by 6
Learning-to-rank debias with popularity-weighted negative sampling and popularity regularization
Wang, C., Liu, A., & Qin, T. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127681 · Cited by 5
A neighborhood-based method for mining and fusing positive and negative false samples
Pan, Q., Zheng, T., Li, S., Zhang, Z., & Jiang, J. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.111345 · Cited by 4
Uncertainty Calibration for Counterfactual Propensity Estimation in Recommendation
Hu, W., Sun, X., Liu, Q., Wu, L., & Wang, L. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3552658 · Cited by 8
“Bias in recommendation systems is a significant concern in current research [66], [67], [68], impacting the fairness and diversity of recommendations.”
Learning instrumental variable representation for debiasing in recommender systems
Zhi-rong, H., Zhang, S., Cheng, D., Li, J., Liu, L., Lu, G., & Zhang, G. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107977 · Cited by 8
Disentangled Variational Autoencoder for Social Recommendation
Zhang, Y., Huang, J., & Yang, J. Y. (2024) · Neural Processing Letters · doi:10.1007/s11063-024-11607-y · Cited by 4
“Liu et al. [28] and Zhao et al. [29] individually disentangled biased and unbiased factors for recommendations.”
Model-Agnostic Dual-Side Online Fairness Learning for Dynamic Recommendation
Tang, H., Wu, S., Cui, Z., et al. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3544510 · Cited by 4
“There has been a growing awareness that recommendation systems are subject to algorithmic fairness considerations along different dimensions, including stakeholder (e.g., producer and consumer), type of benefit, context, and even morality [8], [9], [10], [11], [12].”
Causal intervention for knowledge graph denoising in recommender systems
Guo, Z., Song, P., Feng, C., Yao, K., Dang, C., & Liang, J. (2024) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-024-02500-0 · Cited by 3
“For instance, Zhao et al. [43] utilized the causal graph to analyze the effect of popularity bias in RSs. Chen et al. [44] proposed an unbiased recommendation framework for knowledge distillation.”
Counteracting Popularity Bias in Multimedia Web API Recommendation
Zhai, D., Yan, C., Zhong, W., Ding, S., Qi, L., & Zhou, X. (2025) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2024.3517601 · Cited by 3
“However, unbiased data needs to be ar-tificially achieved, which leads to the high price of unbiased data [41].”
Price-aware debiased learning model for recommendation
Wu, J., Yang, B., Zhu, Q., Mao, R., & Li, Q. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107872 · Cited by 3
Mitigating influence of selection and conformity biases for user interest in recommender systems
Zheng, T., Xu, X., Li, S., et al. (2026) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2026.113810 · Cited by 3
TEXT CAN BE FAIR: Mitigating Popularity Bias with PLMs by Learning Relative Preference
Tang, Z., Huan, Z., Li, Z., Hu, S., Zhang, X., Zhou, J., Zou, L., & Li, C. (2024) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3627673.3679581 · Cited by 4
CSLP: Collaborative Solution to Long-Tail Problem and Popularity Bias in Sequential Recommendation
Huang, Y., Yang, Z., Hu, W., Xu, B., & Zhang, Z. (2024) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc54092.2024.10831683 · Cited by 3
“Recent researches [3]– [6], [22] report significant progress in mitigating popularity bias.”
Causal Inference for Multi-Criteria Rating Recommender Systems
Guo, Z., Song, P., Feng, C., Yao, K., & Liang, J. (2025) · ACM Transactions on Information Systems · doi:10.1145/3757737 · Cited by 3
“For RSs, causal inference can effectively alleviate the spurious association and bias problems (such as popularity bias [74], location bias [23], exposure bias [48], etc.) caused by confounding variables [11].”
Popularity Bias in Correlation Graph-based API Recommendation for Mashup Creation
Yan, C., Zhong, W., Zhai, D., Khan, A. A., Gong, W., Xu, Y., & Xin, B. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3654445 · Cited by 2
“Researchers are increasingly aware of the problem of popularity bias in recommender systems and have conducted many research works [ 1 , 7 , 14 , 20 , 24 , 42 , 43 ].”
Causal Disentanglement for Tackling Popularity Bias in Sequential Recommendation
Liu, A.-A., Zhao, Y., Wen, X., Chang, R., & Nie, W. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3575554 · Cited by 2
“Additionally, with the development of causal inference technology, it has been widely applied in recommendation systems [36].”
Neural recommendation by user-item-user and item-user-item relation modeling
Xu, Y.-Y., Xiao, H., Zhang, H., Wang, D.-D., Wang, K., & Min, F. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112104 · Cited by 2
Huang, B., Wang, Z., Yan, F., et al. (2025) · Applied Soft Computing · doi:10.1016/j.asoc.2025.113928 · Cited by 2
Ding, H., Zhu, W., Hu, G., et al. (2025) · Systems · doi:10.3390/systems13111034 · Cited by 2
Graph-Structured Driven Dual Adaptation for Mitigating Popularity Bias
Cai, M., Chen, L., Wang, Y., Cheng, Z., Zhang, M., & Wang, M. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3638343 · Cited by 2
“However, popularity bias persists as a critical challenge, wherein highly-interacted items dominate recommendation lists, causing unpopular items to be increasingly marginalized [6], [7], [8].”
FMGAN: A Filter-Enhanced MLP Debias Recommendation Model Based on Generative Adversarial Network
Liu, Z., & Luo, W. (2023) · Applied Sciences · doi:10.3390/app13137975 · Cited by 2
Addressing popularity discrepancy in collaborative filtering
Yu, C., Li, D., Gu, H., Zhang, P., Gu, N., & Lu, T. (2025) · Knowledge and Information Systems · doi:10.1007/s10115-025-02426-1 · Cited by 2
“TIDE [17] is a novel time-aware disentangled framework, enabling interventional inference to leverage beneficial popularity bias while mitigating harmful bias.”
InfluentialUnbiased, Effective, and Efficient Distillation from Heterogeneous Models for Recommender Systems
Kang, S., Kweon, W., Lee, D., Lian, J., Xie, X., & Yu, H. (2024) · ACM Transactions on Recommender Systems · doi:10.1145/3649443 · Cited by 2
“There have been many attempts to tackle the popularity lift [ 32 , 33 , 52 , 63 ], including approaches such as invariant learning [ 62 ], specialized architectures [ 30 ], and popularity-robust loss functions [ 43 ].”
Learning Robust Sequential Recommenders through Confident Soft Labels
Wu, S., Xin, X., Ren, P., et al. (2024) · ACM Transactions on Information Systems · doi:10.1145/3700876 · Cited by 1
“For a deeper analysis, popularity bias is not always harmful for the recommender according to Zhao et al. [43], popular items sometimes do reflect the general user interests.”
Are Representation Disentanglement and Interpretability Linked in Recommendation Models?
Dervishaj, E., Ruotsalo, T., Maistro, M., & Lioma, C. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-88717-8_4 · Cited by 1
TDIC: Time-Aware Disentanglement of Interest and Conformity in Mobile App Recommendations
Li, Q., Zhao, Y., & Ma, Y. (2025) · IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws67624.2025.00011 · Cited by 2
“To enhance recommendation fairness and accuracy, we must better capture the distinct roles of temporal conformity and item quality within popularity [10], [11], using causal models that adapt to time-evolving user behavior.”
Popularity debiasing of multi-behavior recommendation via causal inference and data augmentation
Bin, C., Chen, H., & Zhang, F. (2025) · Applied Intelligence · doi:10.1007/s10489-025-07053-5 · Cited by 1
“For instance, Zhao et al. [ 28] mitigate the direct impact of item popularity by tracing the consistency effect and applying causal interventions.”
Self-citationEvolution and impact of AI in knowledge sharing: insights from a bibliometric analysis
Trinh, N., Nguyen, M., & Ngo, D.-T. (2025) · Evolutionary Intelligence · doi:10.1007/s12065-025-01129-4 · Cited by 1
“They are affected by the same algorithmic biases common in commercial recommendation systems, including popularity bias and exposure bias [53–55].”
SAGERec: Sampling and Gating for Enhanced Long-Tail Item Recommendations
Alshabanah, A., Yang, Y., & Annavaram, M. (2026) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3773966.3778004 · Cited by 1
Fairness-Aware Group Recommendation with Spatio-Temporal Feature Fusion
Jia, J., Wang, F., & Liu, S. (2026) · Measurement · doi:10.1016/j.measurement.2026.121344 · Cited by 1
PD-SRS: Personalized Diversity for a Fair Session-Based Recommendation System
Kermany, N. R., Pizzato, L. A. S., Yang, J. Y., et al. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20984-0_23 · Cited by 1
Calibrating Popularity Bias Based on Quality for Recommendation Fairness
Guo, Z., Zhu, Y., Wang, Z., & Jing, M. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-46677-9_26 · Cited by 1
Alleviating Confounding Effects with Contrastive Learning in Recommendation
You, D., & Lee, K. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-56060-6_30 · Cited by 1
“To name a few, [3] categorized common biases into seven types, with selection bias [12, 27], exposure bias [18, 20, 22, 48] , and popularity bias [37, 43, 44, 46] being the most discussed.”
BackgroundBiasRec: A General Bias-Aware Social Recommendation Model
Zhang, C., & Li, G. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-5572-1_7 · Cited by 1
Kralr: knowledge-enhanced representation aggregation for long-tail recommendation
Zhang, Z., Zhang, Y., Li, W., et al. (2024) · Annals of Operations Research · doi:10.1007/s10479-024-06208-w · Cited by 1
“Thus, popularity bias is widespread in RSs (Chen et al., 2023a; Zhao et al., 2021).”
Mitigating Confounding and Selection Biases in Personalized Recommendation: A Causal Approach
Huang, W. Q., Zhou, J., Wu, X., & Dou, D. (2023) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata59044.2023.10386496 · Cited by 1
“Both of them have been studied but still under-explored in causal recommendation literature [24], [26].”
BackgroundModeling Variational Anchoring Effect for Recommender Systems
Xiao, Y., Zhang, Y., & Li, X. (2024) · IEEE Conference on Artificial Intelligence (CAI) · doi:10.1109/cai59869.2024.00170
Chen, G.-M., Qiang, W., Ouyang, Y., et al. (2024) · doi:10.1109/ijcnn60899.2024.10651019
“Popularity bias refers to a common issue in recommender systems where certain popular items receive a disproportionately high level of attention compared to others [9], which means the distribution of interactions over items is quite imbalanced and even long-tailed [10].”
BackgroundMethodWang, Z., Zou, H., Liu, J., Wu, J., Tian, P., He, Y., & Cui, P. (2024) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3706637
“— AdaptSel outperforms both TIDE and PDA by adapting to the test environment.”
InfluentialDisentangled Causal Embedding with Unbiased Knowledge Distillation for Recommendation
Liu, N., Meng, S., Liu, X., Li, Q., & Jiang, Y. Y. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0850-8_9
Contrastive Disentangled Variational Autoencoder for Collaborative Filtering
Yun, W.-S., Kang, S.-M., & Cho, Y.-S. (2025) · IEEE Access · doi:10.1109/access.2025.3576445
Enhancing collaborative translational metric learning with causal graph network
Wang, J., Xie, H., Qin, S. J., Tao, X., Wang, F. L., & Xu, X. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130933
Modeling Social Behavior in Collaborative Filtering
Zhang, Y., & Hara, T. (2025) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3726302.3730039
SM-RS 2.0: User-perceived Qualities of Single- and Multi-Objective Recommender Systems
Dokoupil, P., & Peška, L. (2025) · ACM Transactions on Recommender Systems · doi:10.1145/3754459
“This is in line with, e.g., Zhao et al. [67] who state that a portion of popularity bias simply reflects better item qualities, rather than users’ conformity patterns.”
Wu, J., Chen, G., Pang, C., Zhao, J., & See-To, E. W. K. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.108028
Xu, H., Yang, R., & Geng, R. (2025) · INFORMS journal on computing · doi:10.1287/ijoc.2024.0682
KGRL: Knowledge graph-enhanced representation learning for long-tail recommendation
Zhang, Z., Zhang, Y., Wang, M., et al. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.115063
Accurate and Diverse Recommendations via Propensity-Weighted Linear Autoencoders
Onishi, K., Hayashi, K., & Kamigaito, H. (2025) · Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (SIGIR-AP) · doi:10.1145/3767695.3769512
KGNS: Knowledge graph-driven neighbor selection for long-tail recommendations
Zhang, Z., Zhang, Y., & Ren, Y. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104561
Adaptive disentangled learning recommendation via similarity popularity
Ye, J., Wang, H., Deng, J., Zhang, Z., Wang, Y., Xu, Z., & Khushvakhtzoda, K. K. (2025) · Applied Intelligence · doi:10.1007/s10489-025-07050-8
“In recent years, causal theory has been widely applied in the recommendation systems, which improves the performances of various recommendation tasks [30, 31].”
The Unfairness of Multifactorial Bias in Recommendation
Mansoury, M., Huang, J., Pechenizkiy, M., van Hoof, H., & de Rijke, M. (2026) · ACM Transactions on Information Systems · doi:10.1145/3841477
“Notably, Zhao et al. [92] show that conformity is also a key driver of popularity bias, as users tend to interact with already popular items regardless of their actual preferences, further inflating the popularity of these items.”
Metadata
- Authors
- Zhao, Z., Chen, J., Zhou, S., He, X., Cao, X., Zhang, F., & Wu, W.
- Affiliations
- University of Science and Technology of China (CN)
- Zhejiang University (CN)
- Meituan (China) (CN)
- Venue
- IEEE Transactions on Knowledge and Data Engineering
- Published
- 2023-10-01
- Volume
- 35
- Issue
- 10
- Pages
- 9920-9931
- ISSN
- 1041-4347 (ISSN-L)
- 1558-2191
- 2326-3865
- OpenAlex
- W4312551924
- Semantic Scholar
- 3d361ead1582071a5ed26cff391504746a1210a6
- Cited by
- 117
- References
- 48
- Citations by year
- 2026: 21 · 2025: 36 · 2024: 35 · 2023: 21
- FWCI
- 23.75
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Knowledge and Data Engineering (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- National Key Research and Development Program of China (2021ZD0111802)
- National Natural Science Foundation of China (62102382, 61972372, U21B2026)
- Meituan Inc.
- CCCD Key Lab of Ministry of Culture and Tourism
- Starry Night Science Fund of Zhejiang University Shanghai Institute for Advanced Study (SN-ZJU-SIAS-001)
- Topics
- Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Social Sciences › Decision Sciences › Management Science and Operations Research › Advanced Bandit Algorithms Research
- Keywords
- popularity bias, recommender systems, item quality, temporal modeling, disentangled representations, click data
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
