Click-through Rate (CTR) prediction is a crucial task in online advertising and recommender systems. Theoretically, proper ensemble of multiple different CTR prediction models can improve the prediction effectiveness. Unfortunately, most of the existing ensemble learning methods for CTR prediction are designed for cross-sectional data and neglect user historical behavior sequence which is important for predicting user's future click behavior. To address the above issues, we propose a Hypergraph Diffusion-based Sequential Ensemble framework for CTR prediction (HDSE). Specifically, considering the inherent capability of diffusion models in space exploration, we design a generalized conditional diffusion model, which adaptively identifies critical information from diverse sequential models, to capture user's dynamic interest evolution across diverse contexts. To avoid corrupting the item dependencies caused by isotropic Gaussian noise used in traditional diffusion models, we construct a behavior hypergraph and design an anisotropic hypergraph-based smoothing operator to inject structure-aware noise for better exploring the user's interest space. For large-scale application scenarios, we propose a computationally efficient approximation method for estimating hypergraph propagation matrix in the smoothing operator. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed model.

The current paradigm of training large language models (LLMs) on public available Web data is becoming unsustainable as high-quality data sources in specialized domains near exhaustion. Federated Learning (FL) emerges as a practical solution for the next generation of AI on a decentralized Web, enabling privacy-preserving collaborative fine-tuning on decentralized private data. While Low-Rank Adaptation (LoRA) is standard for efficient fine-tuning, its federated application faces a critical bottleneck: communication overhead under heterogeneous network conditions. Structural redundancy in LoRA parameters increases communication costs and causes aggregation conflicts. To address this, we propose FedSRD, a Sparsify-Reconstruct-Decompose framework for communication-efficient federated LLM fine-tuning. We introduce importance-aware sparsification to reduce the upload parameter count while preserving the structural integrity of LoRA updates. The server aggregates updates in full-rank space to mitigate conflicts, then decomposes the global update into a sparse low-rank format for broadcast, ensuring a symmetrically efficient cycle. We also propose an efficient variant, FedSRD-e, to reduce computational overhead. Experiments on 10 benchmarks show our framework significantly reduces communication costs by up to 90% while improving performance on heterogeneous client data.

The development of artificial intelligence and medical informatics has empowered the medication recommendation systems with enhanced capabilities. However, existing methods struggle with the data imbalance problem in Electronic Health Records (EHRs), where the majority of records are concentrated on a limited subset of common diagnoses, procedures, and medications. It hampers the models’ ability to recommend appropriate medications when dealing with uncommon or multifaceted cases. In addition, existing approaches often fail to adequately model the complex relationships inherent in heterogeneous medical data sources, especially medication molecular structure information. This gap restricts the potential for uncovering meaningful associations among diverse clinical entities. To address these issues, we design a hierarchical attention-based pretraining strategy, leveraging the semantic hierarchies of medical entity codes to facilitate knowledge transfer, so as to alleviate the challenge of data imbalance. Furthermore, we design a novel bilateral hierarchical heterogeneous hypergraph convolution network for medication recommendation. Specifically, we construct specialized hypergraphs for both EHR data and medication molecular structure data, enabling hypergraph convolution to capture high-order relationships while promoting bilateral knowledge enhancement between these heterogeneous data sources. This comprehensive integration allows the model to effectively capture the relationships among clinical and molecular information. Experimental results on different hospital departments of MIMIC-III and MIMIC-IV datasets demonstrate the superior performance of our model compared to state-of-the-art methods. Our source code is released at: https://github.com/LusiaZ/BH3-MedRec .

Distributed databases are fundamental infrastructures of today's large-scale software systems such as cloud systems. Detecting anomalies in distributed databases is essential for maintaining software availability. Existing approaches, predominantly developed using Loghub-a comprehensive collection of log datasets from various systems-lack datasets specifically tailored to distributed databases, which exhibit unique anomalies. Additionally, there's a notable absence of datasets encompassing multi-anomaly, multi-node logs. Consequently, models built upon these datasets, primarily designed for standalone systems, are inadequate for distributed databases, and the prevalent method of deeming an entire cluster anomalous based on irregularities in a single node leads to a high false-positive rate. This paper addresses the unique anomalies and multivariate nature of logs in distributed databases. We expose the first open-sourced, comprehensive dataset with multivariate logs from distributed databases. Utilizing this dataset, we conduct an extensive study to identify multiple database anomalies and to assess the effectiveness of state-of-the-art anomaly detection using multivariate log data. Our findings reveal that relying solely on logs from a single node is insufficient for accurate anomaly detection on distributed database. Leveraging these insights, we propose MultiLog, an innovative multivariate log-based anomaly detection approach tailored for distributed databases. Our experiments, based on this novel dataset, demonstrate MultiLog's superiority, outperforming existing state-of-the-art methods by approximately 12%.

Session-based recommendation (SBR) aims to predict user behaviors based on anonymous sessions. Compared with traditional user-based recommendation, SBR has a wider range of applications, but also suffers from more severe data sparsity problems because of the absence of user-profiles and limited short-term interactions. Furthermore, both users and items in the real world have a multifaceted nature. Users may exhibit multiple intents within a session, while items may have different semantics in different contexts. Unfortunately, existing approaches often overlook or only consider one aspect of them. To address these issues, we propose a novel hypergraph-based framework for session-based recommendation, called Hyperedge Interactional Convolution Network (HICN). Each session is represented as a sequential hyperedge, and multiple modules are designed to model and make use of the multifaceted nature of items and users. In addition, two inter-hyperedge modeling modules are designed to leverage related auxiliary information from other sessions with consideration of the existence of noise, which can help alleviate the data sparsity problem. Extensive experiments on three real-world datasets demonstrate the effectiveness of the proposed model HICN.

Sequential recommendation seeks to predict users' next behaviors and recommend related items over time. Existing research has mainly focused on modeling users' dynamic preferences from their sequential behaviors. However, most of these studies have ignored the negative effects of noise behaviors in the given sequences, which may mislead the recommender. In addition, users' behavior data is always sparse, which makes it difficult to effectively learn users' preferences purely from their historical behaviors. Most recently, knowledge graphs (KGs) have been exploited by few researchers for sequential recommendation. However, they always assume all information in KGs or KG paths with limited length are useful for recommendation, which may bring irrelevant information from KGs into the recommender and further mislead the recommender. To address these issues, we propose a novel KG-based behavior denoising and preference learning model named KGDPL for sequential recommendation. We argue that the paths in KGs that reflect semantic relations between entities can not only help to remove noise behaviors and recommend successive items for users, but also provide relevant explanations. Therefore, we first devise a supervised knowledge path selection module to select effective paths between items from KGs for behavior prediction, which aims to filter out irrelevant information from KGs for the given recommendation task. Then, we design a knowledge-enhanced behavior denoising module to mitigate the negative effects of the noise behaviors contained in historical sequences by using the knowledge path information. After that, we propose a knowledge-enhanced preference learning module to better learn users' personalized and dynamic preferences from their historical behavior sequences and related knowledge information, which can also help tag users and provide explanations for recommendation results. Experimental results on four real-world datasets demonstrate the effectiveness and interpretability of the proposed model KGDPL.

Reciprocal recommendation is the core of many social websites like online recruitment and online dating. Most recently, graph neural networks have been exploited by few researchers for reciprocal recommendation. However, they tend to oversimplify the interactions between users, treating them as simple pairwise relationships, which overlooks the multidimensional relationships among users. Additionally, these methods fail to consider users’ historical interaction sequences and feedback behaviors, which makes it challenging to effectively capture the changes of user preferences over time.To address these issues, this study proposes a novel bilateral recommendation model based on sequential hypergraphs for reciprocal scenarios. Firstly, to capture the complex multidimensional relationships between bilateral users, we design a new data structure called bilateral sequential hypergraphs to capture the diverse relationships among users and to mine collaborative signals at the sequential level. Secondly, we propose corresponding bilateral sequential hypergraph convolution structures to learn the embedded representations of bilateral users. To adequately capture the changes in user preferences, the model incorporates position modeling and feedback behavior modeling within the proposed convolution strategy. Extensive experiments on several real-world datasets demonstrate that the proposed method consistently outperforms state-of-the-art methods.

With the increasing complexity of modern software systems, it is essential yet hard to detect anomalies and diagnose problems precisely. Existing log-based anomaly detection approaches rely on a few key assumptions on system logs and perform well in some experimental systems. However, real-world industrial systems are often with poor logging quality, in which system logs are noisy and often violate the assumptions of existing approaches. This makes these approaches inefficient. This paper first conducts a comprehensive study on the system logs of three large-scale industrial software systems. Through the study, we identify four typical anti-patterns that affect the detection results the most. Based on these patterns, we propose HiLog, an effective human-in-the-loop log-based anomaly detection approach that integrates human knowledge to augment anomaly detection models. With little human labeling effort, our approach can significantly improve the effectiveness of existing models. Experiment results on three large-scale industrial software systems show that our method improves over 50% precision rate on average.

Traditional conversational systems can only access the given query during the response generation, leading to meaningless responses. To this end, researchers proposed to enhance dialogue generation by integrating external knowledge. Although such methods have achieved remarkable gains, the use of only single-source knowledge often makes existing knowledge-enhanced methods degenerate into traditional models in real scenarios because of the insufficient knowledge coverage of single-source knowledge. To improve the applicability of knowledge-enhanced methods, we propose two novel frameworks to use heterogeneous knowledge from multiple sources. We first propose an MHKD-Seq2Seq framework, which can use different heterogeneous knowledge by identifying abstract-level knowledge behaviors; meanwhile, a Diffuse-Aggregate scheme is used to process multiple knowledge simultaneously and produce a unified result. The next framework MHKD-ARPLM can leverage the advantages of pretrained language models with Knowledge Linearization techniques. In experiments, we collected dialogues from previously open-released datasets and built a multi-source knowledge-aligned dataset TriKE-Weibo, which involves three knowledge sources: commonsense, texts, and infobox tables. Extensive evaluations demonstrate the performance leadership of our approaches against competitive baseline models.

Social recommendation, which leverages social connections to construct Recommender Systems (RS), plays an important role in alleviating information overload. Recently, Graph Neural Networks (GNNs) have received increasing attention due to their great capacity for graph data. Since data in RS is essentially in the structure of graphs, GNN-based RS is flourishing. However, existing works lack in-depth thinking of social recommendations. These methods contain implicit assumptions that are not well analyzed in practice. To tackle these problems, we conduct statistical analyses on widely used social recommendation datasets. We design metrics to evaluate the social information, which can provide guidance about whether and how we should use this information in the RS task. Based on these analyses, we propose a Distillation Enhanced SocIal Graph Network (DESIGN). We train a model that integrates information from the user-item interaction graph and the user-user social graph and train two auxiliary models that only use one of the above graphs respectively. These models are trained simultaneously, where the knowledge distillation technique restricts the training process and makes them learn from each other. Our extensive experiments show that our model significantly and consistently outperforms the state-of-the-art competitors on real-world datasets.

Online reviews, which contain the quality information and user experience about products, always affect the consumption decisions of customers. Unfortunately, quite a number of spammers attempt to mislead consumers by writing fake reviews for some intents. Existing methods for detecting spam reviews mainly focus on constructing discriminative features, which heavily depend on experts and may miss some complex but effective features. Recently, some models attempt to learn the latent representations of reviews, users, and items. However, the learned embeddings usually lack interpretability. Moreover, most of existing methods are based on single classification model while ignoring the complementarity of different classification models.

Precise prediction of users' behavior is critical for users' satisfaction and platforms' benefit. A user's behavior heavily depends on the user's general preference and contextual information (current location, weather etc.). In this paper, we propose a succinct hierarchical framework named Hierarchical Hybrid Feature Model (HHFM). It combines users' general taste and diverse contextual information into a hybrid feature representation to profile users' dynamic preference w.r.t context. Meanwhile, we propose an n-way concatenation pooling strategy to capture the non-linear and complex inherent structures of real-world data, which were ignored by most existing methods like Factorization Machines. Conceptually, our model subsumes several existing methods when choosing proper concatenation and pooling strategies. Extensive experiments show our model consistently outperforms state-of-the-art methods on three real-world data sets.

Recommendation based on heterogeneous information network(HIN) is attracting more and more attention due to its ability to emulate collaborative filtering, content-based filtering, context-aware recommendation and combinations of any of these recommendation semantics. Random walk based methods are usually used to mine the paths, weigh the paths, and compute the closeness or relevance between two nodes in a HIN. A key for the success of these methods is how to properly set the weights of links in a HIN. In existing methods, the weights of links are mostly set heuristically. In this paper, we propose a Bayesian Personalized Ranking(BPR) based machine learning method, called HeteLearn, to learn the weights of links in a HIN. In order to model user preferences for personalized recommendation, we also propose a generalized random walk with restart model on HINs. We evaluate the proposed method in a personalized recommendation task and a tag recommendation task. Experimental results show that our method performs significantly better than both the traditional collaborative filtering and the state-of-the-art HIN-based recommendation methods.

A combination of multiple retrieval systems can outperform its individual component systems, but it remains a challenging problem to predict whether two systems can be beneficially combined and, if so, the optimal means by which they should be merged. The performance of combined systems is affected by many factors, including the performance of individual systems, the diversity between a pair of systems, and the method for combination. In this paper, we undertake the study of these issues using combinatorial fusion algorithm (CFA) utilizing the rank-score characteristic (RSC) function and the notion of a weighted cognitive diversity. Using the selected eight TREC datasets, we demonstrated that: (a) the combination of two retrieval systems performs better than each individual system only when the individual systems have relatively good performance and they are diverse, (b) a dynamic combination method, using rank vs. score combination based on cognitive diversity which does not display a tight correlation with other statistical diversity measures, can improve the performance of the combined system, even when performance of each individual system is not known or in the context of an unsupervised learning environment. Within the TREC datasets, the proposed dynamic approach offers a potential for substantial improvement with no significant risk. Our results provide a new paradigm of dynamic fusion to the study of the combination of multiple retrieval systems.