Recommender systems (RSs) often suffer from the feedback loop phenomenon, i.e., RSs are trained on data biased by their recommendations. This leads to the filter bubble effect that reinforces homogeneous content and reduces user satisfaction. To this end, serendipity recommendations, which offer unexpected yet relevant items, are proposed. Recently, large language models (LLMs) have shown potential in serendipity prediction due to their extensive world knowledge and reasoning capabilities. However, they still face challenges in aligning serendipity judgments with human assessments, handling long user behavior sequences, and meeting the latency requirements of industrial RSs. To address these issues, we propose SERAL (Serendipity Recommendations with Aligned Large Language Models), a framework comprising three stages: (1) Cognition Profile Generation to compress user behavior into multi-level profiles; (2) SerenGPT Alignment to align serendipity judgments with human preferences using enriched training data; and (3) Nearline Adaptation to integrate SerenGPT into industrial RSs pipelines efficiently. Online experiments demonstrate that SERAL improves exposure ratio (PVR), clicks, and transactions of serendipitous items by 5.7%, 29.56%, and 27.6%, enhancing user experience without much impact on overall revenue. Now, it has been fully deployed in ''Guess What You Like'' on the Taobao App homepage.

Recently, learning embeddings of nodes in graphs has attracted increasing research attention. There are two main kinds of graph embedding methods, i.e., transductive embedding methods and inductive embedding methods. The former focuses on directly optimizing the embedding vectors, and the latter tries to learn a mapping function for the given nodes and features. However, little work has focused on applying the learned model from one graph to another, which is a pervasive idea in Computer Vision or Natural Language Processing. Although some of the graph neural networks (GNNs) present a similar motivation, none of them considers graph biases between graphs. In this paper, we present a novel graph embedding problem called Adaptive Task (AT), and propose a unified framework for the adaptive task, which introduces two types of alignment to learn adaptive node embeddings across graphs. Then, based on the proposed framework, a novel Graph Adaptive Embedding network (GraphAE) is designed to address the adaptive task. Furthermore, we extend GraphAE to a multi-graph version to consider a more complex adaptive situation. The extensive experimental results demonstrate that our model significantly outperforms the state-of-the-art methods, and also show that our framework can make a great improvement over a number of existing GNNs.

Recently, Graph Neural Networks (GNNs) arouse a lot of research interest and achieve great success in dealing with graph-based data. The basic idea of GNNs is to aggregate neighbor information iteratively. After k iterations, a k-layer GNN can capture nodes' k-hop local structure. In this way, a deeper GNN can access much more neighbor information leading to better performance. However, when a GNN goes deeper, the exponential expansion of neighborhoods incurs expensive computations in batched training and inference. This takes the deeper GNN away from many applications, e.g., real-time systems. In this paper, we try to learn a small GNN (called TinyGNN), which can achieve high performance and infer the node representation in a short time. However, since a small GNN cannot explore as much local structure as a deeper GNN does, there exists a neighbor information gap between the deeper GNN and the small GNN. To address this problem, we leverage peer node information to model the local structure explicitly and adopt a neighbor distillation strategy to learn local structure knowledge from a deeper GNN implicitly. Extensive experimental results demonstrate that TinyGNN is empirically effective and achieves similar or even better performance compared with the deeper GNNs. Meanwhile, TinyGNN gains a 7.73x--126.59x speed-up on inference over all data sets.

The skyline of a data point set consists of the best points in the set, and is very important for multi-criteria decision making. One recent and important variant of the traditional skyline is group-based skyline, which aims to find the best groups of points in a given set. This paper brings forward an efficient approach, called minimum dominance search (MDS), to solve the g-skyline problem, a latest group-based skyline problem. MDS consists of two steps: In the first step, a novel g-skyline support structure, i.e., minimum dominance graph (MDG), is constructed to store all the points which may occur in g-skyline groups. In the second step, two searching algorithms are proposed to find g-skyline groups based on the MDG through two searching algorithms, and a skyline-combination based optimization strategy is employed to improve these two algorithms. The support for dynamic group sizes, i.e., a practical extension of the origin g-skyline problem, is provided through slightly modifying MDS. Comprehensive experiments are conducted on both synthetic and real-world data sets, and the results show that our algorithms are orders of magnitude faster than the state-of-the-art.

The skyline of a data point set is made up of the best points in the set, and is very important for multi-criteria decision making. In these years, the skyline problem attracts more and more attention, and many variants of the traditional skyline emerge in the database field. One recent and important variant is group-based skyline, which aims to find the best groups of points in a given set. In this paper, we bring forward an efficient approach, called minimum dominance search (MDS), to solve the g-skyline problem, a latest group-based skyline problem. MDS consists of two steps: In the first step, we construct a novel g-skyline support structure, i.e., minimum dominance graph (MDG), which proves to be a minimum g-skyline support structure. In the second step, we search for g-skyline groups based on the MDG through two searching algorithms, and a skyline-combination based optimization strategy is employed to improve these two algorithms. We conduct comprehensive experiments on both synthetic and real-world data sets, and show that our algorithms are orders of magnitude faster than the state-of-the-art in most cases.