To balance effectiveness and efficiency in recommender systems, multi-stage pipelines commonly use lightweight two-tower models for large-scale candidate retrieval. However, the isolated two-tower architecture restricts representation capacity, embedding-space alignment, and cross-feature interactions. Existing solutions such as late interaction and knowledge distillation can mitigate these issues, but often increase latency or are difficult to deploy in online learning settings. We propose Capability Synergy (CS3), an efficient online framework that strengthens two-tower retrievers while preserving real-time constraints. CS3 introduces three mechanisms: (1) Cycle-Adaptive Structure for self-revision via adaptive feature denoising within each tower; (2) Cross-Tower Synchronization to improve alignment through lightweight mutual awareness between towers; and (3) Cascade-Model Sharing to enhance cross-stage consistency by reusing knowledge from downstream models. CS3 is plug-and-play with diverse two-tower backbones and compatible with online learning. Experiments on three public datasets show consistent gains over strong baselines, and deployment in a large-scale advertising system yields up to 8.36% revenue improvement across three scenarios while maintaining ms-level latency.

Recommender systems are increasingly prevalent to provide personalized suggestions and enhance user satisfaction. Typical recommendation models encode users and items as embeddings, and generate recommendations by assessing the similarity between these embeddings. Despite their effectiveness, these embedding-based models struggle with modeling user uncertainty and capturing diverse user interests using a single fixed user embedding. Recent studies have begun to explore a user-distribution paradigm to learn distributions for users. However, this approach employs a single distribution per user, which fails to effectively delineate semantic boundaries, resulting in sub-optimal recommendations. To this end, we propose GCDR, a Guided Conditional Diffusion Recommender model, to learn multiple distributions for each user in this paper. Specifically, GCDR addresses two major challenges: 1) learning disentangled distributions, and 2) learning personalized distributions. GCDR captures inter-user and intra-user distribution properties through conditional and guided diffusion, respectively. It maintains user-specific embeddings to encode long-term interests for conditional diffusion, while for guided diffusion, it incorporates short-term interests encoded from recent interactions with category preferences. To align the diffusion model with the recommendation task, we train GCDR with three loss functions, included the user loss, the recommendation loss and the diffusion loss. Extensive experiments on four real-world datasets show that GCDR is able to learn effective user distributions and is superior to thirteen state-of-the-art baseline methods.

Semantic IDs are discrete identifiers generated by quantizing the Multi-modal Large Language Models embeddings, enabling efficient multi-modal content integration in recommendation systems. However, their lack of collaborative signals results in a misalignment with downstream discriminative and generative recommendation objectives. Recent studies have introduced various alignment mechanisms to address this problem, but their two-stage framework design still leads to two main limitations: (1) inevitable information loss during alignment, and (2) inflexibility in applying adaptive alignment strategies, consequently constraining the mutual information maximization during the alignment process.

Providing reasonable explanations for a specific suggestion given by the recommender can help users trust the system more. As logic rule-based inference is concise, transparent, and aligned with human cognition, it can be adopted to improve the interpretability of recommendation models. Previous work that interprets user preference with logic rules merely focuses on the construction of rules while neglecting the usage of feature embeddings. This limits the model in capturing implicit relationships between features. In this article, we aim to improve both the effectiveness and explainability of recommendation models by simultaneously representing logic rules and feature embeddings. We propose a novel model-intrinsic explainable recommendation method named Feature-Enhanced Neural Collaborative Reasoning (FENCR) . The model automatically extracts representative logic rules from massive possibilities in a data-driven way. In addition, we utilize feature interaction-based neural modules to represent logic operators on embeddings. Experiments on two large public datasets show our model outperforms state-of-the-art neural logical recommendation models. Further case analyses demonstrate that FENCR can derive reasonable rules, indicating its high robustness and expandability. 1

Recommender systems have emerged as an indispensable mean to meet personalized interests of users and alleviate information overload. Despite the great success, accuracy-oriented recommendation models are creating information cocoons, i.e., it is becoming increasingly difficult for users to see other items they might be interested in. Although recent studies start paying attention to enhancing recommendation diversity, models based on point embedding fail to describe the range of user preferences and item features well, which is essential for diversified matching. To this end, we propose LCD-UC , a novel List-Check-Decide framework with UnCertainty masking based on box embedding to improve recommendation diversity with recommendation accuracy maintained. Specifically, LCD-UC creates hypercubes to represent users and items using box embedding for high model flexibility and expressiveness. Then, a hypercube similarity scoring function is designed to measure the similarity between hypercubes representing users and items. To make a balance between the accuracy and diversity of recommendations and achieve personalized diversity needs, we further develop a user-item pairwise attention mechanism as well as a user uncertainty masking mechanism in LCD-UC. Besides, we present two new metrics for better evaluation on recommendation diversity, which address the issue that existing metrics only consider the coverage of categories while ignore the frequency of categories. The extensive experiments on three real-world datasets show that LCD-UC can improve both recommendation accuracy and diversity over three base models, and is superior to six state-of-the-art recommendation models. An online 10-day AB test also demonstrates that LCD-UC can improve the performance of a real-world advertising system.

Graphs can represent relational information among entities and graph structures are widely used in many intelligent tasks such as search, recommendation, and question answering. However, most of the graph-structured data in practice suffer from incompleteness, and thus link prediction becomes an important research problem. Though many models are proposed for link prediction, the following two problems are still less explored: (1) Most methods model each link independently without making use of the rich information from relevant links, and (2) existing models are mostly designed based on associative learning and do not take reasoning into consideration. With these concerns, in this paper, we propose Graph Collaborative Reasoning (GCR), which can use the neighbor link information for relational reasoning on graphs from logical reasoning perspectives. We provide a simple approach to translate a graph structure into logical expressions so that the link prediction task can be converted into a neural logic reasoning problem. We apply logical constrained neural modules to build the network architecture according to the logical expression and use backpropagation to efficiently learn the model parameters, which bridges differentiable learning and symbolic reasoning in a unified architecture. To show the effectiveness of our work, we conduct experiments on graph-related tasks such as link prediction and recommendation based on commonly used benchmark datasets, and our graph collaborative reasoning approach achieves state-of-the-art performance.

Although neural networks have achieved great successes in various machine learning tasks, people can hardly know what neural networks learn from data due to their black-box nature. The lack of such explainability is one of the limitations of neural networks when applied in domains, e.g., healthcare and finance, that demand transparency and accountability. Moreover, explainability is beneficial for guiding a neural network to learn the causal patterns that can extrapolate out-of-distribution (OOD) data, which is critical in real-world applications and has surged as a hot research topic.

News recommendation plays an indispensable role in acquiring daily news for users. Previous studies make great efforts to model high-order feature interactions between users and items, where various neural models are applied (e.g., RNN, GNN). However, we find that seldom efforts are made to get better representations for news. Most previous methods simply adopt pre-trained word embeddings to represent news and also suffer from cold-start users.

Existing Collaborative Filtering (CF) methods are mostly designed based on the idea of matching, i.e., by learning user and item embeddings from data using shallow or deep models, they try to capture the associative relevance patterns in data, so that a user embedding can be matched with relevant item embeddings using designed or learned similarity functions. However, as a cognition rather than a perception intelligent task, recommendation requires not only the ability of pattern recognition and matching from data, but also the ability of cognitive reasoning in data.

Recent years have witnessed the success of deep neural networks in many research areas. The fundamental idea behind the design of most neural networks is to learn similarity patterns from data for prediction and inference, which lacks the ability of cognitive reasoning. However, the concrete ability of reasoning is critical to many theoretical and practical problems. On the other hand, traditional symbolic reasoning methods do well in making logical inference, but they are mostly hard rule-based reasoning, which limits their generalization ability to different tasks since difference tasks may require different rules. Both reasoning and generalization ability are important for prediction tasks such as recommender systems, where reasoning provides strong connection between user history and target items for accurate prediction, and generalization helps the model to draw a robust user portrait over noisy inputs.

This paper focuses on tree-based modeling for the sentence classification task.In existing works, aggregating on a syntax tree usually considers local information of sub-trees.In contrast, in addition to the local information, our proposed Modularized Syntactic Neural Network (MSNN) utilizes the syntax category labels and takes advantage of the global context while modeling sub-trees.In MSNN, each node of a syntax tree is modeled by a label-related syntax module.Each syntax module aggregates the outputs of lower-level modules, and finally, the root module provides the sentence representation.We design a tree-parallel mini-batch strategy for efficient training and predicting.Experimental results on four benchmark datasets show that our MSNN significantly outperforms previous state-of-the-art tree-based methods on the sentence classification task.

Modeling large scale and rare-interaction users are the two major challenges in recommender systems, which derives big gaps between researches and applications. Facing to millions or even billions of users, it is hard to store and leverage personalized preferences with a user embedding matrix in real scenarios. And many researches pay attention to users with rich histories, while users with only one or several interactions are the biggest part in real systems. Previous studies make efforts to handle one of the above issues but rarely tackle efficiency and cold-start problems together.

Most recommendation algorithms mainly make use of user history interactions in the model, while these methods often suffer from the cold-start problem (user/item has no history information). On the other sides, content features help on cold-start scenarios for modeling new users or items. So it is essential to utilize content features to enhance different recommendation models. To take full advantage of content features, feature interactions such as cross features are used by some models and outperform than using raw features. However, in real-world systems, many content features are incomplete, e.g., we may know the occupation and gender of a user, but the values of other features (location, interests, etc.) are missing. This missing-feature-value (MFV) problem is harmful to the model performance, especially for models that rely heavily on rich feature interactions. Unfortunately, this problem has not been well studied previously.

Nowadays, recommender systems provide essential web services on the Internet. There are mainly two categories of traditional recommendation algorithms: Content-Based (CB) and Collaborative Filtering (CF). CF methods make recommendations mainly according to the historical feedback information. They usually perform better when there is sufficient feedback information but less successful on new users and items, which is called the "cold-start'' problem. However, CB methods help in this scenario because of using content information. To take both advantages of CF and CB, how to combine them is a challenging issue. To the best of our knowledge, little previous work has been done to solve the problem in one unified recommendation model. In this work, we study how to integrate CF and CB, which utilizes both types of information in model-level but not in result-level and makes recommendations adaptively. A novel attention-based model named Attentional Content&Collaborate Model (ACCM) is proposed. Attention mechanism helps adaptively adjust for each user-item pair from which source information the recommendation is made. Especially, a "cold sampling'' learning strategy is designed to handle the cold-start problem. Experimental results on two benchmark datasets show that the ACCM performs better on both warm and cold tests compared to the state-of-the-art algorithms.

Research on Computer-Assisted Language Learning (CALL) has shown that the use of multimedia materials such as images and videos can facilitate interpretation and memorization of new words and phrases by providing richer cues than text alone. We present ViVo, a novel video-augmented dictionary that provides an inexpensive, convenient, and scalable way to exploit huge online video resources for vocabulary learning. ViVo automatically generates short video clips from existing movies with the target word highlighted in the subtitles. In particular, we apply a word sense disambiguation algorithm to identify the appropriate movie scenes with adequate contextual information for learning. We analyze the challenges and feasibility of this approach and describe our interaction design. A user study showed that learners were able to retain nearly 30% more new words with ViVo than with a standard bilingual dictionary days after learning. They preferred our video-augmented dictionary for its benefits in memorization and enjoyable learning experience.