{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-tale-of-two-graphs-freezing-and-denoising","title":"A Tale of Two Graphs: Freezing and Denoising Graph Structures for Multimodal Recommendation","arxiv_id":"2211.06924","date":"2022-11-13","proceeding":null,"authors":["Xin Zhou","Zhiqi Shen"],"abstract":"Multimodal recommender systems utilizing multimodal features (e.g., images and textual descriptions) typically show better recommendation accuracy than general recommendation models based solely on user-item interactions. Generally, prior work fuses multimodal features into item ID embeddings to enrich item representations, thus failing to capture the latent semantic item-item structures. In this context, LATTICE proposes to learn the latent structure between items explicitly and achieves state-of-the-art performance for multimodal recommendations. However, we argue the latent graph structure learning of LATTICE is both inefficient and unnecessary. Experimentally, we demonstrate that freezing its item-item structure before training can also achieve competitive performance. Based on this finding, we propose a simple yet effective model, dubbed as FREEDOM, that FREEzes the item-item graph and DenOises the user-item interaction graph simultaneously for Multimodal recommendation. Theoretically, we examine the design of FREEDOM through a graph spectral perspective and demonstrate that it possesses a tighter upper bound on the graph spectrum. In denoising the user-item interaction graph, we devise a degree-sensitive edge pruning method, which rejects possibly noisy edges with a high probability when sampling the graph. We evaluate the proposed model on three real-world datasets and show that FREEDOM can significantly outperform current strongest baselines. Compared with LATTICE, FREEDOM achieves an average improvement of 19.07% in recommendation accuracy while reducing its memory cost up to 6$\\times$ on large graphs. The source code is available at: https://github.com/enoche/FREEDOM.","url_abs":"https://arxiv.org/abs/2211.06924v3","url_pdf":"https://arxiv.org/pdf/2211.06924v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"a-tale-of-two-graphs-freezing-and-denoising","repo_url":"https://github.com/enoche/freedom","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"a-tale-of-two-graphs-freezing-and-denoising","repo_url":"https://github.com/sisinflab/formal-multimod-rec","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"graph-structure-learning","task_name":"Graph structure learning"},{"task_slug":"multi-modal-recommendation","task_name":"Multi-modal Recommendation"},{"task_slug":"multimodal-recommendation","task_name":"Multimodal Recommendation"},{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multi-modal-recommendation-on-amazon-baby","task":"Multi-modal Recommendation","dataset":"Amazon Baby","model":"FREEDOM","rank_in_archive_order":2,"of":10,"metrics":{"NDCG@20":"0.0424"},"uses_additional_data":false},{"leaderboard":"/sota/multi-modal-recommendation-on-amazon-clothing","task":"Multi-modal Recommendation","dataset":"Amazon Clothing","model":"FREEDOM","rank_in_archive_order":2,"of":10,"metrics":{"NDCG@20":"0.0416"},"uses_additional_data":false},{"leaderboard":"/sota/multi-modal-recommendation-on-amazon-sports","task":"Multi-modal Recommendation","dataset":"Amazon Sports","model":"FREEDOM","rank_in_archive_order":2,"of":10,"metrics":{"NGCG@20":"0.0481"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2211.06924","atlas_url":"https://app.syntology.ai/?focus=2211.06924","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}