Graph Neural Networks (GNNs) have demonstrated their superiority in collaborative filtering, where the user-item (U-I) interaction bipartite graph serves as the fundamental data format. However, when graph-structured side information (e.g., multimodal similarity graphs or social networks) is integrated into the U-I bipartite graph, existing graph collaborative filtering methods fall short of achieving satisfactory performance. We quantitatively analyze this problem from a spectral perspective. Recall that a bipartite graph possesses a full spectrum within the range of [-1, 1], with the highest frequency exactly achievable at -1 and the lowest frequency at 1; however, we observe as more side information is incorporated, the highest frequency of the augmented adjacency matrix progressively shifts rightward. This spectrum shift phenomenon has caused previous approaches built for the full spectrum [-1, 1] to assign mismatched importance to different frequencies. To this end, we propose Spectrum Shift Correction (dubbed SSC), incorporating shifting and scaling factors to enable spectral GNNs to adapt to the shifted spectrum. Unlike previous paradigms of leveraging side information, which necessitate tailored designs for diverse data types, SSC directly connects traditional graph collaborative filtering with any graph-structured side information. Experiments on social and multimodal recommendation demonstrate the effectiveness of SSC, achieving relative improvements of up to 23% without incurring any additional computational overhead. Our code is available at https://github.com/yhhe2004/SSC-KDD.
Multimodal click-through rate (CTR) prediction is a key technique in industrial recommender systems. It leverages heterogeneous modalities such as text, images, and behavioral logs to capture high-order feature interactions between users and items, thereby enhancing the system's understanding of user interests and its ability to predict click behavior. The primary challenge in this field lies in effectively utilizing the rich semantic information from multiple modalities while satisfying the low-latency requirements of online inference in real-world applications. To foster progress in this area, the Multimodal CTR Prediction Challenge Track of the WWW 2025 EReL@MIR Workshop formulates the problem into two tasks: (1) Task 1 of Multimodal Item Embedding: this task aims to explore multimodal information extraction and item representation learning methods that enhance recommendation tasks; and (2) Task 2 of Multimodal CTR Prediction: this task aims to explore what multimodal recommendation model can effectively leverage multimodal embedding features and achieve better performance. In this paper, we propose a novel model for Task 2, named Quadratic Interest Network (QIN) for Multimodal CTR Prediction. Specifically, QIN employs adaptive sparse target attention to extract multimodal user behavior features, and leverages Quadratic Neural Networks to capture high-order feature interactions. As a result, QIN achieved an AUC of 0.9798 on the leaderboard and ranked second in the competition. The model code, training logs, hyperparameter configurations, and checkpoints are available at https://github.com/salmon1802/QIN.
Recent works in multimodal recommendations, which leverage diverse modal information to address data sparsity and enhance recommendation accuracy, have garnered considerable interest. Two key processes in multimodal recommendations are modality fusion and representation learning. Previous approaches in modality fusion often employ simplistic attentive or pre-defined strategies at early or late stages, failing to effectively handle irrelevant information among modalities. In representation learning, prior research has constructed heterogeneous and homogeneous graph structures encapsulating user-item, user-user, and item-item relationships to better capture user interests and item profiles. Modality fusion and representation learning were considered as two independent processes in previous work. In this paper, we reveal that these two processes are complementary and can support each other. Specifically, powerful representation learning enhances modality fusion, while effective fusion improves representation quality. Stemming from these two processes, we introduce a COmposite grapH convolutional nEtwork with dual-stage fuSION for the multimodal recommendation, named COHESION. Specifically, it introduces a dual-stage fusion strategy to reduce the impact of irrelevant information, refining all modalities using ID embedding in the early stage and fusing their representations at the late stage. It also proposes a composite graph convolutional network that utilizes user-item, user-user, and item-item graphs to extract heterogeneous and homogeneous latent relationships within users and items. Besides, it introduces a novel adaptive optimization to ensure balanced and reasonable representations across modalities. Extensive experiments on three widely used datasets demonstrate the significant superiority of COHESION over various competitive baselines.
Incomplete scenario is a prevalent, practical, yet challenging setting in Multimodal Recommendations (MMRec), where some item modalities are missing due to various factors. Recently, a few efforts have sought to improve the recommendation accuracy by exploring generic structures from incomplete data. However, two significant gaps persist: 1) the difficulty in accurately generating missing data due to the limited ability to capture modality distributions; and 2) the critical but overlooked visibility bias, where items with missing modalities are more likely to be disregarded due to the prioritization of items' multimodal data over user preference alignment. This bias raises serious concerns about the fair treatment of items. To bridge these two gaps, we propose a novel Modality-Diffused Counterfactual (MoDiCF) framework for incomplete multimodal recommendations. MoDiCF features two key modules: a novel modality-diffused data completion module and a new counterfactual multimodal recommendation module. The former, equipped with a particularly designed multimodal generative framework, accurately generates and iteratively refines missing data from learned modality-specific distribution spaces. The latter, grounded in the causal perspective, effectively mitigates the negative causal effects of visibility bias and thus assures fairness in recommendations. Both modules work collaboratively to address the two aforementioned significant gaps for generating more accurate and fair results. Extensive experiments on three real-world datasets demonstrate the superior performance of MoDiCF in terms of both recommendation accuracy and fairness. The code and processed datasets are released at https://github.com/JinLi-i/MoDiCF.
Multimodal information (e.g., visual, acoustic, and textual) has been widely used to enhance representation learning for micro-video recommendation. For integrating multimodal information into a joint representation of micro-video, multimodal fusion plays a vital role in the existing micro-video recommendation approaches. However, the static multimodal fusion used in previous studies is insufficient to model the various relationships among multimodal information of different micro-videos. In this paper, we develop a novel meta-learning-based multimodal fusion framework called Meta Multimodal Fusion (MetaMMF), which dynamically assigns parameters to the multimodal fusion function for each micro-video during its representation learning. Specifically, MetaMMF regards the multimodal fusion of each micro-video as an independent task. Based on the meta information extracted from the multimodal features of the input task, MetaMMF parameterizes a neural network as the item-specific fusion function via a meta learner. We perform extensive experiments on three benchmark datasets, demonstrating the significant improvements over several state-of-the-art multimodal recommendation models, like MMGCN, LATTICE, and InvRL. Furthermore, we lighten our model by adopting canonical polyadic decomposition to improve the training efficiency, and validate its effectiveness through experimental results. Codes are available at https://github.com/hanliu95/MetaMMF.
Incorporating multi-modal features as side information has recently become a trend in recommender systems. To elucidate user-item preferences, recent studies focus on fusing modalities via concatenation, element-wise sum, or attention mechanisms. Despite having notable success, existing approaches do not account for the modality-specific noise encapsulated within each modality. As a result, direct fusion of modalities will lead to the amplification of cross-modality noise. Moreover, the variation of noise that is unique within each modality results in noise alleviation and fusion being more challenging. In this work, we propose a new Spectrum-based Modality Representation (SMORE) fusion graph recommender that aims to capture both uni-modal and fusion preferences while simultaneously suppressing modality noise. Specifically, SMORE projects the multi-modal features into the frequency domain and leverages the spectral space for fusion. To reduce dynamic contamination that is unique to each modality, we introduce a filter to attenuate and suppress the modality noise adaptively while capturing the universal modality patterns effectively. Furthermore, we explore the item latent structures by designing a new multi-modal graph learning module to capture associative semantic correlations and universal fusion patterns among similar items. Finally, we formulate a new modality-aware preference module, which infuses behavioral features and balances the uni- and multi-modal features for precise preference modeling. This empowers SMORE with the ability to infer both user modality-specific and fusion preferences more accurately. Experiments on three real-world datasets show the efficacy of our proposed model. The source code for this work has been made publicly available at https://github.com/kennethorq/SMORE.
Multimodal recommendation systems can learn users' preferences from existing user-item interactions as well as the semantics of multimodal data associated with items. Many existing methods model this through a multimodal user-item graph, approaching multimodal recommendation as a graph learning task. Graph Neural Networks (GNNs) have shown promising performance in this domain. Prior research has capitalized on GNNs' capability to capture neighborhood information within certain receptive fields (typically denoted by the number of hops, $K$) to enrich user and item semantics. We observe that the optimal receptive fields for GNNs can vary across different modalities. In this paper, we propose GNNs with Modality-Independent Receptive Fields, which employ separate GNNs with independent receptive fields for different modalities to enhance performance. Our results indicate that the optimal $K$ for certain modalities on specific datasets can be as low as 1 or 2, which may restrict the GNNs' capacity to capture global information. To address this, we introduce a Sampling-based Global Transformer, which utilizes uniform global sampling to effectively integrate global information for GNNs. We conduct comprehensive experiments that demonstrate the superiority of our approach over existing methods. Our code is publicly available at https://github.com/CrawlScript/MIG-GT.
Given the large volume of side information from different modalities, multimodal recommender systems have become increasingly vital, as they exploit richer semantic information beyond user-item interactions. Recent works highlight that leveraging Graph Convolutional Networks (GCNs) to explicitly model multimodal item-item relations can significantly enhance recommendation performance. However, due to the inherent over-smoothing issue of GCNs, existing models benefit only from shallow GCNs with limited representation power. This drawback is especially pronounced when facing complex and high-dimensional patterns such as multimodal data, as it requires large-capacity models to accommodate complicated correlations. To this end, in this paper, we investigate bypassing GCNs when modeling multimodal item-item relationship. More specifically, we propose a Topology-aware Multi-Layer Perceptron (TMLP), which uses MLPs instead of GCNs to model the relationships between items. TMLP enhances MLPs with topological pruning to denoise item-item relations and intra (inter)-modality learning to integrate higher-order modality correlations. Extensive experiments on three real-world datasets verify TMLP's superiority over nine baselines. We also find that by discarding the internal message passing in GCNs, which is sensitive to node connections, TMLP achieves significant improvements in both training efficiency and robustness against existing models.
While the mining of modalities is the focus of most multimodal recommendation methods, we believe that how to fully utilize both collaborative and multimodal information is pivotal in e-commerce scenarios where, as clarified in this work, the user behaviors are rarely determined entirely by multimodal features. In order to combine the two distinct types of information, some additional challenges are encountered: 1) Modality erasure: Vanilla graph convolution, which proves rather useful in collaborative filtering, however erases multimodal information; 2) Modality forgetting: Multimodal information tends to be gradually forgotten as the recommendation loss essentially facilitates the learning of collaborative information. To this end, we propose a novel approach named STAIR, which employs a novel STepwise grAph convolution to enable a co-existence of collaborative and multimodal Information in e-commerce Recommendation. Besides, it starts with the raw multimodal features as an initialization, and the forgetting problem can be significantly alleviated through constrained embedding updates. As a result, STAIR achieves state-of-the-art recommendation performance on three public e-commerce datasets with minimal computational and memory costs. Our code is available at https://github.com/yhhe2004/STAIR.
Most recommender systems adopt collaborative filtering (CF) and provide recommendations based on past collective interactions. Therefore, the performance of CF algorithms degrades when few or no interactions are available, a scenario referred to as cold-start. To address this issue, previous work relies on models leveraging both collaborative data and side information on the users or items. Similar to multimodal learning, these models aim at combining collaborative and content representations in a shared embedding space. In this work we propose a novel technique for multimodal recommendation, relying on a multimodal Single-Branch embedding network for Recommendation (SiBraR). Leveraging weight-sharing, SiBraR encodes interaction data as well as multimodal side information using the same single-branch embedding network on different modalities. This makes SiBraR effective in scenarios of missing modality, including cold start. Our extensive experiments on large-scale recommendation datasets from three different recommendation domains (music, movie, and e-commerce) and providing multimodal content information (audio, text, image, labels, and interactions) show that SiBraR significantly outperforms CF as well as state-of-the-art content-based RSs in cold-start scenarios, and is competitive in warm scenarios. We show that SiBraR's recommendations are accurate in missing modality scenarios, and that the model is able to map different modalities to the same region of the shared embedding space, hence reducing the modality gap.
Despite recent advancements in language and vision modeling, integrating rich multimodal knowledge into recommender systems continues to pose significant challenges. This is primarily due to the need for efficient recommendation, which requires adaptive and interactive responses. In this study, we focus on sequential recommendation and introduce a lightweight framework called full-scale Matryoshka representation learning for multimodal recommendation (fMRLRec). Our fMRLRec captures item features at different granularities, learning informative representations for efficient recommendation across multiple dimensions. To integrate item features from diverse modalities, fMRLRec employs a simple mapping to project multimodal item features into an aligned feature space. Additionally, we design an efficient linear transformation that embeds smaller features into larger ones, substantially reducing memory requirements for large-scale training on recommendation data. Combined with improved state space modeling techniques, fMRLRec scales to different dimensions and only requires one-time training to produce multiple models tailored to various granularities. We demonstrate the effectiveness and efficiency of fMRLRec on multiple benchmark datasets, which consistently achieves superior performance over state-of-the-art baseline methods. We make our code and data publicly available at https://github.com/yueqirex/fMRLRec.
In specific domains like fashion, music, and movie recommendation, the multi-faceted features characterizing products and services may influence each customer on online selling platforms differently, paving the way to novel multimodal recommendation models that can learn from such multimodal content. According to the literature, the common multimodal recommendation pipeline involves (i) extracting multimodal features, (ii) refining their high-level representations to suit the recommendation task, (iii) optionally fusing all multimodal features, and (iv) predicting the user-item score. While great effort has been put into designing optimal solutions for (ii-iv), to the best of our knowledge, very little attention has been devoted to exploring procedures for (i). In this respect, the existing literature outlines the large availability of multimodal datasets and the ever-growing number of large models accounting for multimodal-aware tasks, but (at the same time) an unjustified adoption of limited standardized solutions. This motivates us to explore more extensive techniques for the (i) stage of the pipeline. To this end, this paper settles as the first attempt to offer a large-scale benchmarking for multimodal recommender systems, with a specific focus on multimodal extractors. Specifically, we take advantage of two popular and recent frameworks for multimodal feature extraction and reproducibility in recommendation, Ducho and Elliot, to offer a unified and ready-to-use experimental environment able to run extensive benchmarking analyses leveraging novel multimodal feature extractors. Results, largely validated under different hyper-parameter settings for the chosen extractors, provide important insights on how to train and tune the next generation of multimodal recommendation algorithms.
Generally, items with missing modalities are dropped in multimodal recommendation. However, with this work, we question this procedure, highlighting that it would further damage the pipeline of any multimodal recommender system. First, we show that the lack of (some) modalities is, in fact, a widely-diffused phenomenon in multimodal recommendation. Second, we propose a pipeline that imputes missing multimodal features in recommendation by leveraging traditional imputation strategies in machine learning. Then, given the graph structure of the recommendation data, we also propose three more effective imputation solutions that leverage the item-item co-purchase graph and the multimodal similarities of co-interacted items. Our method can be plugged into any multimodal RSs in the literature working as an untrained pre-processing phase, showing (through extensive experiments) that any data pre-filtering is not only unnecessary but also harmful to the performance.
Recent advances in Large Language Models (LLMs) have demonstrated significant potential in the field of Recommendation Systems (RSs). Most existing studies have focused on converting user behavior logs into textual prompts and leveraging techniques such as prompt tuning to enable LLMs for recommendation tasks. Meanwhile, research interest has recently grown in multimodal recommendation systems that integrate data from images, text, and other sources using modality fusion techniques. This introduces new challenges to the existing LLM-based recommendation paradigm which relies solely on text modality information. Moreover, although Multimodal Large Language Models (MLLMs) capable of processing multi-modal inputs have emerged, how to equip MLLMs with multi-modal recommendation capabilities remains largely unexplored. To this end, in this paper, we propose the Multimodal Large Language Model-enhanced Multimodaln Sequential Recommendation (MLLM-MSR) model. To capture the dynamic user preference, we design a two-stage user preference summarization method. Specifically, we first utilize an MLLM-based item-summarizer to extract image feature given an item and convert the image into text. Then, we employ a recurrent user preference summarization generation paradigm to capture the dynamic changes in user preferences based on an LLM-based user-summarizer. Finally, to enable the MLLM for multi-modal recommendation task, we propose to fine-tune a MLLM-based recommender using Supervised Fine-Tuning (SFT) techniques. Extensive evaluations across various datasets validate the effectiveness of MLLM-MSR, showcasing its superior ability to capture and adapt to the evolving dynamics of user preferences.
Many recommender models have been proposed to investigate how to incorporate multimodal content information into traditional collaborative filtering framework effectively. The use of multimodal information is expected to provide more comprehensive information and lead to superior performance. However, the integration of multiple modalities often encounters the modal imbalance problem: since the information in different modalities is unbalanced, optimizing the same objective across all modalities leads to the under-optimization problem of the weak modalities with a slower convergence rate or lower performance. Even worse, we find that in multimodal recommendation models, all modalities suffer from the problem of insufficient optimization. To address these issues, we propose a Counterfactual Knowledge Distillation method that could solve the imbalance problem and make the best use of all modalities. Through modality-specific knowledge distillation, it could guide the multimodal model to learn modality-specific knowledge from uni-modal teachers. We also design a novel generic-and-specific distillation loss to guide the multimodal student to learn wider-and-deeper knowledge from teachers. Additionally, to adaptively recalibrate the focus of the multimodal model towards weaker modalities during training, we estimate the causal effect of each modality on the training objective using counterfactual inference techniques, through which we could determine the weak modalities, quantify the imbalance degree and re-weight the distillation loss accordingly. Our method could serve as a plug-and-play module for both late-fusion and early-fusion backbones. Extensive experiments on six backbones show that our proposed method can improve the performance by a large margin. The source code will be released at \url{https://github.com/CRIPAC-DIG/Balanced-Multimodal-Rec}
Multimodal recommendation systems (MMRS) have received considerable attention from the research community due to their ability to jointly utilize information from user behavior and product images and text. Previous research has two main issues. First, many long-tail items in recommendation systems have limited interaction data, making it difficult to learn comprehensive and informative representations. However, past MMRS studies have overlooked this issue. Secondly, users' modality preferences are crucial to their behavior. However, previous research has primarily focused on learning item modality representations, while user modality representations have remained relatively simplistic.To address these challenges, we propose a novel Graphs and User Modalities Enhancement (GUME) for long-tail multimodal recommendation. Specifically, we first enhance the user-item graph using multimodal similarity between items. This improves the connectivity of long-tail items and helps them learn high-quality representations through graph propagation. Then, we construct two types of user modalities: explicit interaction features and extended interest features. By using the user modality enhancement strategy to maximize mutual information between these two features, we improve the generalization ability of user modality representations. Additionally, we design an alignment strategy for modality data to remove noise from both internal and external perspectives. Extensive experiments on four publicly available datasets demonstrate the effectiveness of our approach.
Traditional recommender systems heavily rely on ID features, which often encounter challenges related to cold-start and generalization. Modeling pre-extracted content features can mitigate these issues, but is still a suboptimal solution due to the discrepancies between training tasks and model parameters. End-to-end training presents a promising solution for these problems, yet most of the existing works mainly focus on retrieval models, leaving the multimodal techniques under-utilized. In this paper, we propose an industrial multimodal recommendation framework named EM3: End-to-end training of Multimodal Model and ranking Model, which sufficiently utilizes multimodal information and allows personalized ranking tasks to directly train the core modules in the multimodal model to obtain more task-oriented content features, without overburdening resource consumption. First, we propose Fusion-Q-Former, which consists of transformers and a set of trainable queries, to fuse different modalities and generate fixed-length and robust multimodal embeddings. Second, in our sequential modeling for user content interest, we utilize Low-Rank Adaptation technique to alleviate the conflict between huge resource consumption and long sequence length. Third, we propose a novel Content-ID-Contrastive learning task to complement the advantages of content and ID by aligning them with each other, obtaining more task-oriented content embeddings and more generalized ID embeddings. In experiments, we implement EM3 on different ranking models in two scenario, achieving significant improvements in both offline evaluation and online A/B test, verifying the generalizability of our method. Ablation studies and visualization are also performed. Furthermore, we also conduct experiments on two public datasets to show that our proposed method outperforms the state-of-the-art methods.
With the development of multimedia systems, multimodal recommendations are playing an essential role, as they can leverage rich contexts beyond interactions. Existing methods mainly regard multimodal information as an auxiliary, using them to help learn ID features; However, there exist semantic gaps among multimodal content features and ID-based features, for which directly using multimodal information as an auxiliary would lead to misalignment in representations of users and items. In this paper, we first systematically investigate the misalignment issue in multimodal recommendations, and propose a solution named AlignRec. In AlignRec, the recommendation objective is decomposed into three alignments, namely alignment within contents, alignment between content and categorical ID, and alignment between users and items. Each alignment is characterized by a specific objective function and is integrated into our multimodal recommendation framework. To effectively train AlignRec, we propose starting from pre-training the first alignment to obtain unified multimodal features and subsequently training the following two alignments together with these features as input. As it is essential to analyze whether each multimodal feature helps in training and accelerate the iteration cycle of recommendation models, we design three new classes of metrics to evaluate intermediate performance. Our extensive experiments on three real-world datasets consistently verify the superiority of AlignRec compared to nine baselines. We also find that the multimodal features generated by AlignRec are better than currently used ones, which are to be open-sourced in our repository https://github.com/sjtulyf123/AlignRec_CIKM24.
In this work, we introduce Ducho 2.0, the latest stable version of our framework. Differently from Ducho, Ducho 2.0 offers a more personalized user experience with the definition and import of custom extraction models fine-tuned on specific tasks and datasets. Moreover, the new version is capable of extracting and processing features through multimodal-by-design large models. Notably, all these new features are supported by optimized data loading and storing to the local memory. To showcase the capabilities of Ducho 2.0, we demonstrate a complete multimodal recommendation pipeline, from the extraction/processing to the final recommendation. The idea is to provide practitioners and experienced scholars with a ready-to-use tool that, put on top of any multimodal recommendation framework, may permit them to run extensive benchmarking analyses. All materials are accessible at: \url{https://github.com/sisinflab/Ducho}.
With the increasing multimedia information, multimodal recommendation has received extensive attention. It utilizes multimodal information to alleviate the data sparsity problem in recommendation systems, thus improving recommendation accuracy. However, the reliance on labeled data severely limits the performance of multimodal recommendation models. Recently, self-supervised learning has been used in multimodal recommendations to mitigate the label sparsity problem. Nevertheless, the state-of-the-art methods cannot avoid the modality noise when aligning multimodal information due to the large differences in the distributions of different modalities. To this end, we propose a Multi-level sElf-supervised learNing for mulTimOdal Recommendation (MENTOR) method to address the label sparsity problem and the modality alignment problem. Specifically, MENTOR first enhances the specific features of each modality using the graph convolutional network (GCN) and fuses the visual and textual modalities. It then enhances the item representation via the item semantic graph for all modalities, including the fused modality. Then, it introduces two multilevel self-supervised tasks: the multilevel cross-modal alignment task and the general feature enhancement task. The multilevel cross-modal alignment task aligns each modality under the guidance of the ID embedding from multiple levels while maintaining the historical interaction information. The general feature enhancement task enhances the general feature from both the graph and feature perspectives to improve the robustness of our model. Extensive experiments on three publicly available datasets demonstrate the effectiveness of our method. Our code is publicly available at https://github.com/Jinfeng-Xu/MENTOR.