SG-UMP: Sequence-Guided Universal Multimodal Prioritization Calculation Framework
2026-08-28 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors address a problem in multimodal sequential recommendation systems, which use different types of data like text and images to suggest items. Current methods struggle to adjust to different users' preferences and biases in various datasets. They propose SG-UMP, a framework that flexibly combines and orders different types of data based on user and dataset needs. Their experiments on four datasets show SG-UMP improves recommendation accuracy with various models. The framework can be added to existing systems easily.
multimodal sequential recommendationuser preference heterogeneitymodality biasmachine learning frameworkmodule combinermodule routerdataset adaptationrecommendation systemsdynamic module orderingflexible multimodal processing
Authors
Xinyi Zhang, Yutong Li, Peijie Sun
Abstract
Multimodal sequential recommendation (MSR) improves recommendation by incorporating heterogeneous information such as text, images, and user interactions. However, existing MSR methods often fail to capture user-level preference heterogeneity and dataset-level modality bias, limiting their adaptability across users and datasets. To address this issue, we propose \textbf{S}equence-\textbf{G}uided \textbf{U}niversal \textbf{M}ultimodal \textbf{P}rioritization Calculation Framework (\textbf{SG-UMP}), a plug-and-play plugin for enhancing multimodal information processing in MSR. SG-UMP includes a Module Combiner for flexible multimodal processing and a Module Router for dynamic module ordering, enabling adaptation to both user preferences and dataset characteristics. Experiments on four real-world datasets show that SG-UMP consistently improves recommendation performance across different backbones and multimodal settings. The code is available at https://github.com/esemsc-xz524/SG-UMP .