MIRAGE-SR: Missing-modality Invariant Representation And Generative Enrichment for multimodal Session-based Recommendation
and
*
Article Sidebar
Full Text: 
PDF
Main Article Content
Abstract
Multimodal session-based recommendation commonly assumes that ID, visual, and textual features are available for every item, but this assumption is difficult to satisfy in real-world e-commerce data. This paper proposes MIRAGE-SR, a session-graph neural network architecture designed to maintain recommendation quality when one or more modalities are missing. The method combines two-stage modality imputation over an item-item co-occurrence graph, modality-invariant and modality-specific representation learning, and a missing-modality-aware late-fusion gate trained with curriculum modality dropout. MIRAGE-SR is evaluated on two subsets of Amazon Reviews 2023 under two settings: simulated modality cold-start on Grocery and naturally missing modalities on Cellphones. In the cold-start scenario, the configurations combining modality imputation and curriculum modality dropout achieve higher P@20 and MRR@20 than the zero-imputation configuration without curriculum modality dropout. On Cellphones, the two-stage GMI configuration achieves a P@20 of 26.24% and an MRR@20 of 14.61%, improving upon zero imputation by 3.91 and 1.53 points, respectively. The results indicate that the configurations combining imputation and training-time modality masking improve performance over the zero-imputation configuration; the current evaluation does not isolate the contribution of each component. These findings provide initial evidence of the potential of MIRAGE-SR for incomplete multimodal sequential recommendation.
Article Details
Conflict of Interest
© 2026 The authors. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License.
References
Cam-Nhung, M. T., & Thai-Nghe, N. (2025). Graph neural networks with transformer for sequential recommendation. SN Computer Science, 6(7), 792.
Cam-Nhung, M. T., & Thai-Nghe, N. (2026). A fusion model of neural attention mechanism and matrix factorization for sequential recommendation systems. In The 14th Conference on Information Technology and its Applications (pp. 203–215). Springer Nature Switzerland.
Cam-Nhung, M. T., & Thai-Nghe, N. (2027). Mmlf-gnnt: A multimodal late-fusion graph neural network with transformer for sequential recommendation. In Advances and Trends in Artificial Intelligence. Theory and Applications (pp. 238–249). Springer Nature Singapore.
Chen, L., Yi, P., Hou, X., Yang, C., & Cai, X. (2026). When modalities go missing: Early fusion for multimodal recommendation. In Proceedings of the Nineteenth ACM International Conference on Web Search and Data Mining (pp. 1094–1098).
Ganhör, C., Moscati, M., Hausberger, A., Nawaz, S., & Schedl, M. (2024). A multimodal single-branch embedding network for recommendation in cold-start and missing modality scenarios. In Proceedings of the 18th ACM Conference on Recommender Systems (pp. 380–390).
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 770–778).
Hidasi, B., & Karatzoglou, A. (2018). Recurrent neural networks with top-k gains for session-based recommendations. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management (pp. 843–852).
Ho, T. D. H., & Nguyen, S. T. T. (2024). Self-attentive sequential recommendation models enriched with more features. In Proceedings of the 2024 8th International Conference on Deep Learning Technologies (pp. 49–55).
Hou, Y., Li, J., He, Z., Yan, A., Chen, X., & McAuley, J. (2024). Bridging language and items for retrieval and recommendation. arXiv preprint arXiv:2403.03952.
Jolliffe, I. T., & Cadima, J. (2016). Principal component analysis: A review and recent developments. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 374(2065), 20150202.
Li, J., Wang, S., Zhang, Q., Yu, S., & Chen, F. (2025). Generating with fairness: A modality-diffused counterfactual framework for incomplete multimodal recommendations. In Proceedings of the ACM on Web Conference 2025 (pp. 2787–2798).
Malitesta, D., Rossi, E., Pomo, C., Di Noia, T., & Malliaros, F. D. (2024). Do we really need to drop items with missing modalities in multimodal recommendation? In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (pp. 3943–3948).
McAuley Lab. (2026). Amazon Reviews 2023: Pure IDs (5-core). https://amazon-reviews-2023.github.io/data_processing/5core.html
Sentence Transformers. (2026). all-MiniLM-L6-v2: Model card. https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2.
Cam-Nhung, M., Thuy Anh, N., & Thai-Nghe, N. (2024). Sequential recommendation using graph neuron networks. In Future Data and Security Engineering (pp. 66–79). Springer Nature Singapore.
Wang, J., Zeng, Z., Wang, Y., Wang, Y., Lu, X., Li, T., Yuan, J., Zhang, R., Zheng, H. T., & Xia, S. T. (2023). Missrec: Pre-training and transferring multi-modal interest-aware sequence representation for recommendation. In Proceedings of the 31st ACM International Conference on Multimedia (pp. 6548–6557).