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}
Multimedia recommendation, which incorporates various modalities (e.g., images, texts, etc.) into user or item representation to improve recommendation quality, and self-supervised learning carries multimedia recommendation to a plateau of performance, because of its superior performance in aligning different modalities. However, more and more research finds that aligning all modal representations is suboptimal because it damages the unique attributes of each modal. These studies use subtraction and orthogonal constraints in geometric space to learn unique parts. However, our rigorous analysis reveals the flaws in this approach, such as that subtraction does not necessarily yield the desired modal-unique and that orthogonal constraints are ineffective in user and item high-dimensional representation spaces. To make up for the previous weaknesses, we propose Separate Learning (SEA) for multimedia recommendation, which mainly includes mutual information view of modal-unique and -generic learning. Specifically, we first use GNN to learn the representations of users and items in different modalities and split each modal representation into generic and unique parts. We employ contrastive log-ratio upper bound to minimize the mutual information between the general and unique parts within the same modality, to distance their representations, thus learning modal-unique features. Then, we design Solosimloss to maximize the lower bound of mutual information, to align the general parts of different modalities, thus learning more high-quality modal-generic features. Finally, extensive experiments on three datasets demonstrate the effectiveness and generalization of our proposed framework. The code is available at SEA and the full training record of the main experiment.
Multimedia recommender systems focus on utilizing behavioral information and content information to model user preferences. Typically, it employs pre-trained feature encoders to extract content features, then fuses them with behavioral features. However, pre-trained feature encoders often extract features from the entire content simultaneously, including excessive preference-irrelevant details. We speculate that it may result in the extracted features not containing sufficient features to accurately reflect user preferences. To verify our hypothesis, we introduce an attribution analysis method for visually and intuitively analyzing the content features. The results indicate that certain products' content features exhibit the issues of information drift}and information omission,reducing the expressive ability of features. Building upon this finding, we propose an effective and efficient general Behavior-driven Feature Adapter (BeFA) to tackle these issues. This adapter reconstructs the content feature with the guidance of behavioral information, enabling content features accurately reflecting user preferences. Extensive experiments demonstrate the effectiveness of the adapter across all multimedia recommendation methods. Our code is made publicly available on https://github.com/fqldom/BeFA.
In this paper, we focus on multimedia recommender systems using graph convolutional networks (GCNs) where the multimodal features as well as user-item interactions are employed together. Our study aims to exploit multimodal features more effectively in order to accurately capture users' preferences for items. To this end, we point out following two limitations of existing GCN-based multimedia recommender systems: (L1) although multimodal features of interacted items by a user can reveal her preferences on items, existing methods utilize GCN designed to focus only on capturing collaborative signals, resulting in insufficient reflection of the multimodal features in the final user/item embeddings; (L2) although a user decides whether to prefer the target item by considering its multimodal features, existing methods represent her as only a single embedding regardless of the target item's multimodal features and then utilize her embedding to predict her preference for the target item. To address the above issues, we propose a novel multimedia recommender system, named MONET, composed of following two core ideas: modality-embracing GCN (MeGCN) and target-aware attention. Through extensive experiments using four real-world datasets, we demonstrate i) the significant superiority of MONET over seven state-of-the-art competitors (up to 30.32% higher accuracy in terms of recall@20, compared to the best competitor) and ii) the effectiveness of the two core ideas in MONET. All MONET codes are available at https://github.com/Kimyungi/MONET.
Recommender systems (RSs) offer personalized navigation experiences on online platforms, but recommendation remains a challenging task, particularly in specific scenarios and domains. Multimodality can help tap into richer information sources and construct more refined user/item profiles for recommendations. However, existing literature lacks a shared and universal schema for modeling and solving the recommendation problem through the lens of multimodality. This work aims to formalize a general multimodal schema for multimedia recommendation. It provides a comprehensive literature review of multimodal approaches for multimedia recommendation from the last eight years, outlines the theoretical foundations of a multimodal pipeline, and demonstrates its rationale by applying it to selected state-of-the-art approaches. The work also conducts a benchmarking analysis of recent algorithms for multimedia recommendation within Elliot, a rigorous framework for evaluating recommender systems. The main aim is to provide guidelines for designing and implementing the next generation of multimodal approaches in multimedia recommendation.
Multimedia recommendation has received much attention in recent years. It models user preferences based on both behavior information and item multimodal information. Though current GCN-based methods achieve notable success, they suffer from two limitations: (1) Modality noise contamination to the item representations. Existing methods often mix modality features and behavior features in a single view (e.g., user-item view) for propagation, the noise in the modality features may be amplified and coupled with behavior features. In the end, it leads to poor feature discriminability; (2) Incomplete user preference modeling caused by equal treatment of modality features. Users often exhibit distinct modality preferences when purchasing different items. Equally fusing each modality feature ignores the relative importance among different modalities, leading to the suboptimal user preference modeling. To tackle the above issues, we propose a novel Multi-View Graph Convolutional Network for the multimedia recommendation. Specifically, to avoid modality noise contamination, the modality features are first purified with the aid of item behavior information. Then, the purified modality features of items and behavior features are enriched in separate views, including the user-item view and the item-item view. In this way, the distinguishability of features is enhanced. Meanwhile, a behavior-aware fuser is designed to comprehensively model user preferences by adaptively learning the relative importance of different modality features. Furthermore, we equip the fuser with a self-supervised auxiliary task. This task is expected to maximize the mutual information between the fused multimodal features and behavior features, so as to capture complementary and supplementary preference information simultaneously. Extensive experiments on three public datasets demonstrate the effectiveness of our methods.
The online emergence of multi-modal sharing platforms (eg, TikTok, Youtube) is powering personalized recommender systems to incorporate various modalities (eg, visual, textual and acoustic) into the latent user representations. While existing works on multi-modal recommendation exploit multimedia content features in enhancing item embeddings, their model representation capability is limited by heavy label reliance and weak robustness on sparse user behavior data. Inspired by the recent progress of self-supervised learning in alleviating label scarcity issue, we explore deriving self-supervision signals with effectively learning of modality-aware user preference and cross-modal dependencies. To this end, we propose a new Multi-Modal Self-Supervised Learning (MMSSL) method which tackles two key challenges. Specifically, to characterize the inter-dependency between the user-item collaborative view and item multi-modal semantic view, we design a modality-aware interactive structure learning paradigm via adversarial perturbations for data augmentation. In addition, to capture the effects that user's modality-aware interaction pattern would interweave with each other, a cross-modal contrastive learning approach is introduced to jointly preserve the inter-modal semantic commonality and user preference diversity. Experiments on real-world datasets verify the superiority of our method in offering great potential for multimedia recommendation over various state-of-the-art baselines. The implementation is released at: https://github.com/HKUDS/MMSSL.
Reorganizing implicit feedback of users as a user-item interaction graph facilitates the applications of graph convolutional networks (GCNs) in recommendation tasks. In the interaction graph, edges between user and item nodes function as the main element of GCNs to perform information propagation and generate informative representations. Nevertheless, an underlying challenge lies in the quality of interaction graph, since observed interactions with less-interested items occur in implicit feedback (say, a user views micro-videos accidentally). This means that the neighborhoods involved with such false-positive edges will be influenced negatively and the signal on user preference can be severely contaminated. However, existing GCN-based recommender models leave such challenge under-explored, resulting in suboptimal representations and performance. In this work, we focus on adaptively refining the structure of interaction graph to discover and prune potential false-positive edges. Towards this end, we devise a new GCN-based recommender model, \emph{Graph-Refined Convolutional Network} (GRCN), which adjusts the structure of interaction graph adaptively based on status of model training, instead of remaining the fixed structure. In particular, a graph refining layer is designed to identify the noisy edges with the high confidence of being false-positive interactions, and consequently prune them in a soft manner. We then apply a graph convolutional layer on the refined graph to distill informative signals on user preference. Through extensive experiments on three datasets for micro-video recommendation, we validate the rationality and effectiveness of our GRCN. Further in-depth analysis presents how the refined graph benefits the GCN-based recommender model.
Recent years have witnessed growing interests in multimedia recommendation, which aims to predict whether a user will interact with an item with multimodal contents. Previous studies focus on modeling user-item interactions with multimodal features included as side information. However, this scheme is not well-designed for multimedia recommendation. Firstly, only collaborative item-item relationships are implicitly modeled through high-order item-user-item co-occurrences. We argue that the latent semantic item-item structures underlying these multimodal contents could be beneficial for learning better item representations and assist the recommender models to comprehensively discover candidate items. Secondly, previous studies disregard the fine-grained multimodal fusion. Although having access to multiple modalities might allow us to capture rich information, we argue that the simple coarse-grained fusion by linear combination or concatenation in previous work is insufficient to fully understand content information and item relationships.To this end, we propose a latent structure MIning with ContRastive mOdality fusion method (MICRO for brevity). To be specific, we devise a novel modality-aware structure learning module, which learns item-item relationships for each modality. Based on the learned modality-aware latent item relationships, we perform graph convolutions that explicitly inject item affinities to modality-aware item representations. Then, we design a novel contrastive method to fuse multimodal features. These enriched item representations can be plugged into existing collaborative filtering methods to make more accurate recommendations. Extensive experiments on real-world datasets demonstrate the superiority of our method over state-of-the-art baselines.
Multimedia content is of predominance in the modern Web era. Investigating how users interact with multimodal items is a continuing concern within the rapid development of recommender systems. The majority of previous work focuses on modeling user-item interactions with multimodal features included as side information. However, this scheme is not well-designed for multimedia recommendation. Specifically, only collaborative item-item relationships are implicitly modeled through high-order item-user-item relations. Considering that items are associated with rich contents in multiple modalities, we argue that the latent semantic item-item structures underlying these multimodal contents could be beneficial for learning better item representations and further boosting recommendation. To this end, we propose a LATent sTructure mining method for multImodal reCommEndation, which we term LATTICE for brevity. To be specific, in the proposed LATTICE model, we devise a novel modality-aware structure learning layer, which learns item-item structures for each modality and aggregates multiple modalities to obtain latent item graphs. Based on the learned latent graphs, we perform graph convolutions to explicitly inject high-order item affinities into item representations. These enriched item representations can then be plugged into existing collaborative filtering methods to make more accurate recommendations. Extensive experiments on three real-world datasets demonstrate the superiority of our method over state-of-the-art multimedia recommendation methods and validate the efficacy of mining latent item-item relationships from multimodal features.
In this article, we introduce the ContentWise Impressions dataset, a collection of implicit interactions and impressions of movies and TV series from an Over-The-Top media service, which delivers its media contents over the Internet. The dataset is distinguished from other already available multimedia recommendation datasets by the availability of impressions, i.e., the recommendations shown to the user, its size, and by being open-source. We describe the data collection process, the preprocessing applied, its characteristics, and statistics when compared to other commonly used datasets. We also highlight several possible use cases and research questions that can benefit from the availability of user impressions in an open-source dataset. Furthermore, we release software tools to load and split the data, as well as examples of how to use both user interactions and impressions in several common recommendation algorithms.