Training-free Graph-based Imputation of Missing Modalities in Multimodal Recommendation

2026-02-19

AI Summary

This work addresses the challenge in multimodal recommendation systems where items lacking certain modalities—such as images or text—are often discarded, degrading overall performance. The paper formally characterizes this missing-modality problem and introduces a training-free graph propagation approach: it constructs an item co-purchase graph from user–item interactions and leverages graph signal interpolation to propagate available modality features to missing nodes. The method highlights the critical role of feature homophily over the item graph in enabling effective interpolation. It can be seamlessly integrated into existing recommender systems and consistently outperforms conventional imputation strategies across diverse missing-modality scenarios, while preserving—and often amplifying—the performance advantage of multimodal over unimodal recommendation models.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Multimodal recommender systems (RSs) represent items in the catalog through multimodal data (e.g., product images and descriptions) that, in some cases, might be noisy or (even worse) missing. In those scenarios, the common practice is to drop items with missing modalities and train the multimodal RSs on a subsample of the original dataset. To date, the problem of missing modalities in multimodal recommendation has still received limited attention in the literature, lacking a precise formalisation as done with missing information in traditional machine learning. In this work, we first provide a problem formalisation for missing modalities in multimodal recommendation. Second, by leveraging the user-item graph structure, we re-cast the problem of missing multimodal information as a problem of graph features interpolation on the item-item co-purchase graph. On this basis, we propose four training-free approaches that propagate the available multimodal features throughout the item-item graph to impute the missing features. Extensive experiments on popular multimodal recommendation datasets demonstrate that our solutions can be seamlessly plugged into any existing multimodal RS and benchmarking framework while still preserving (or even widen) the performance gap between multimodal and traditional RSs. Moreover, we show that our graph-based techniques can perform better than traditional imputations in machine learning under different missing modalities settings. Finally, we analyse (for the first time in multimodal RSs) how feature homophily calculated on the item-item graph can influence our graph-based imputations.

Problem

• multimodal recommendation
• missing modalities
• imputation
• graph-based methods
• feature homophily

Innovation

• training-free imputation
• graph-based interpolation
• multimodal recommendation
• missing modalities
• feature homophily