{"type":"ScholarlyArticle","title":"PDFlow: Popularity-Debiased Flow Matching for Sequential Recommendation","authors":[{"name":"Zican Yang","slug":"zican-yang","given":"Zican","family":"Yang","affiliation":""},{"name":"Zeyu Li","slug":"zeyu-li","given":"Zeyu","family":"Li","affiliation":""},{"name":"Cong He","slug":"cong-he","given":"Cong","family":"He","affiliation":""},{"name":"Xiaotao Wu","slug":"xiaotao-wu","given":"Xiaotao","family":"Wu","affiliation":""},{"name":"Guanfeng Liu","slug":"guanfeng-liu","given":"Guanfeng","family":"Liu","affiliation":""},{"name":"Pengpeng Zhao","slug":"pengpeng-zhao","given":"Pengpeng","family":"Zhao","affiliation":""}],"abstract":"Generative models have emerged as a powerful paradigm in sequential recommendation due to their superior distribution modeling. However, long-tail data distributions inevitably induce popularity bias, as iterative generation trajectories gravitate toward dense clusters of popular items. Current debiasing methods struggle to enforce consistent step-wise constraints, rendering them ineffective for the multi-step process of generative recommendation. To address this challenge, we propose a method called Popularity-Debiased Flow Matching for sequential recommendation (PDFlow). Specifically, grounded in the flow matching framework, PDFlow integrates a residual popularity vector field beyond the main flow to model the popularity debiasing flow at each step. This enables real-time trajectory intervention to adjust the generative path, preventing the model from overemphasizing popular items. Subsequently, we incorporate cross-popularity alignment loss, using co-occurrence patterns within the same sequence to align features of both popular and tail items, thereby mitigating popularity-driven distribution separation. Jointly, trajectory correction prevents popularity overfitting while cross-popularity alignment ensures latent fairness, effectively mitigating popularity bias. Extensive experiments on four real-world datasets demonstrate the effectiveness of PDFlow in improving long-tail coverage and overall recommendation performance. All codes and  datasets are available at https://github.com/eqmll/PDFlow.","doi":"10.24963/ijcai.2026/353","url":"https://ijcai.org/proceedings/2026/353","pdf":"https://ijcai.org/proceedings/2026/0353.pdf","markdown":"https://ijcai.org/proceedings/2026/353.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/353","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Data Mining","track":"Main Track","number":353,"pages":{"first":3178,"last":3186},"keywords":["Data Mining: Recommender systems"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}