Bingyu Yan, Xiaoming Zhang, JinYu Hou, Chaozhuo Li, Ziyi Zhou, Yiming Hei, Litian Zhang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

News recommendation aims to offer potentially interesting news items to a specific user, guided by his historical browsing behaviors. Existing methods failed to effectively address the knowledge sparsity issue that the user may have sparse behaviors and the news may own sparse features. To address the problem, we propose a graph-based news recommendation model with intrinsic interest enhancement, named GIIE , leveraging intrinsic interests and neighbor information to enhance the representation of sparse users and news. Concretely, to fully take advantage of the intrinsic interests, we design an interest encoder based on an interest-type graph with a learnable structure and explore the interest embeddings from news types. Then, we inject the obtained interest embeddings into news and represent the user by aggregating the clicked news under the same interest and across different interests sequentially. These interests can build a bridge between users so users with sparse behaviors can implicitly share knowledge with other users, thereby enhancing their representation. To properly introduce the neighbor knowledge, we propose a graph-based neighbor enhancing mechanism. First, we design a news relation graph and a user relation graph in encoders. Then, based on these graphs, we take the attention module to aggregate additional knowledge from neighbors, enhancing sparse news and user representations. To avoid feature ambiguity, we adopt a way to represent the current item (user and news) and its neighbors separately and then do adaptive aggregation. We evaluate GIIE on the public news recommendation datasets MIND-Large and MIND-Small. Experimental results show that our model can solve the knowledge-sparse problem and outperforms current state-of-the-art models in four indicators.

Bipartite graph representation learning aims to obtain node embeddings by compressing sparse vectorized representations of interactions between two types of nodes, e.g., users and items. Incorporating structural attributes among homogeneous nodes, such as user communities, improves the identification of similar interaction preferences, namely, user/item embeddings, for downstream tasks. However, existing methods often fail to proactively discover and fully utilize these latent structural attributes. Moreover, the manual collection and labeling of structural attributes is always costly. In this article, we propose a novel approach called Dirichlet Max-margin Matrix Factorization (DM3F), which adopts a self-supervised strategy to discover latent structural attributes and model discriminative node representations. Specifically, in self-supervised learning, our approach generates pseudo group labels (i.e., structural attributes) as a supervised signal using the Dirichlet process without relying on manual collection and labeling, and employs them in a max-margin classification. Additionally, we introduce a Variational Markov Chain Monte Carlo algorithm (Variational MCMC) to effectively update the parameters. The experimental results on six real datasets demonstrate that, in the majority of cases, the proposed method outperforms existing approaches based on matrix factorization and neural networks. Furthermore, the modularity analysis confirms the effectiveness of our model in capturing structural attributes to produce high-quality user embeddings.

Android is undergoing unprecedented malicious threats daily, but the existing methods for malware detection often fail to cope with evolving camouflage in malware. To address this issue, we present Hawk, a new malware detection framework for evolutionary Android applications. We model Android entities and behavioral relationships as a heterogeneous information network (HIN), exploiting its rich semantic meta-structures for specifying implicit higher order relationships. An incremental learning model is created to handle the applications that manifest dynamically, without the need for reconstructing the whole HIN and the subsequent embedding model. The model can pinpoint rapidly the proximity between a new application and existing in-sample applications and aggregate their numerical embeddings under various semantics. Our experiments examine more than 80 860 malicious and 100 375 benign applications developed over a period of seven years, showing that Hawk achieves the highest detection accuracy against baselines and takes only 3.5 ms on average to detect an out-of-sample application, with the accelerated training time of 50× faster than the existing approach.

Event extraction (EE) is a crucial research task for promptly apprehending event information from massive textual data. With the rapid development of deep learning, EE based on deep learning technology has become a research hotspot. Numerous methods, datasets, and evaluation metrics have been proposed in the literature, raising the need for a comprehensive and updated survey. This article fills the research gap by reviewing the state-of-the-art approaches, especially focusing on the general domain EE based on deep learning models. We introduce a new literature classification of current general domain EE research according to the task definition. Afterward, we summarize the paradigm and models of EE approaches, and then discuss each of them in detail. As an important aspect, we summarize the benchmarks that support tests of predictions and evaluation metrics. A comprehensive comparison among different approaches is also provided in this survey. Finally, we conclude by summarizing future research directions facing the research area.

Event detection in power systems aims to identify triggers and event types, which helps relevant personnel respond to emergencies promptly and facilitates the optimization of power supply strategies. However, the limited length of short electrical record texts causes severe information sparsity, and numerous domain-specific terminologies of power systems makes it difficult to transfer knowledge from language models pre-trained on general-domain texts. Traditional event detection approaches primarily focus on the general domain and ignore these two problems in the power system domain. To address the above issues, we propose a Multi-Channel graph neural network utilizing Type information for Event Detection in power systems, named MC-TED , leveraging a semantic channel and a topological channel to enrich information interaction from short texts. Concretely, the semantic channel refines textual representations with semantic similarity, building the semantic information interaction among potential event-related words. The topological channel generates a relation-type-aware graph modeling word dependencies, and a word-type-aware graph integrating part-of-speech tags. To further reduce errors worsened by professional terminologies in type analysis, a type learning mechanism is designed for updating the representations of both the word type and relation type in the topological channel. In this way, the information sparsity and professional term occurrence problems can be alleviated by enabling interaction between topological and semantic information. Furthermore, to address the lack of labeled data in power systems, we built a Chinese event detection dataset based on electrical Power Event texts, named PoE . In experiments, our model achieves compelling results not only on the PoE dataset, but on general-domain event detection datasets including ACE 2005 and MAVEN.

Sharding blockchains are proposed to solve the scalability problem while maintaining security and decentralization. However, there are still many issues to be solved. First, the member selection and assignment process are not strictly analyzed, which might lead to an increase in the adversary proportion. Second, current intra-shard consensus algorithms are inefficient. Besides, cross-shard transaction processing costs expensive system overhead. Moreover, there is a lack of a formal security framework. In this article, we propose a secure and scalable hybrid consensus (SSHC). First, we propose a fair sharding selection scheme to select committee members, including mining processes and member lists confirmation by a reference committee. Second, a pipelined Byzantine fault tolerance for intra-shard consensus is designed, combining the pipelined technology with threshold signatures. Third, we propose a responsive sharding transaction batch processing mechanism to handle cross-shard transactions, which reduces the number of calls to Byzantine fault tolerance algorithms. Fourth, a secure committee reconfiguration method is designed to update shard members efficiently. Furthermore, we employ a formal security framework to design and analyze a sharding blockchain. For an adversary whose computational power fraction is less than$1/3$, by reasonably setting a corruption parameter and other related parameters, SSHC is proved to achieve consistency and liveness.

In permissionless blockchains, due to the corruption attack of an adversary, nodes participating the protocol need to be updated regularly. In the process of node selection and committee reconfiguration, there may exist some problems. First, a complicated secure randomness generation protocol is in need. Besides, an adversary might obtain a mining puzzle in advance and start mining in ahead of honest nodes. Moreover, an adversary usually has an advantage of network delay. In order to solve the above problems, we conduct the following research. Firstly, we propose a PoW solution withhold attack against PoW-based member selection methods. An adversary might withhold his mining results in an epoch to obtain the mining puzzle of the next epoch in advance of honest nodes. Secondly, a secure shard reconfiguration protocol is designed, which does not rely on any complicated randomness generation protocol. Our shard reconfiguration protocol is proved rigorously to be secure, which means that in each selected committee, the honest node fraction exceeds a predefined target value. Thirdly, we implement our shard reconfiguration protocol. By carefully setting related system parameters, our protocol could be applied easily to most sharding blockchains. To our best knowledge, the shard reconfiguration protocol proposed in this paper is the first protocol that could safely implement node selection and committee reconfiguration of a sharding blockchain without using a secure randomness, which greatly reduces the communication and time overhead caused by the generation of a randomness.

Mobile edge computing (MEC) is to process, analyse, store and calculate the network data at the edge of the network. When the ground infrastructure is damaged in an emergency, the unmanned aerial vehicle (UAV) formation can be rapidly deployed to undertake the task of MEC. However, there are some potential problems to be considered in UAV‐enabled MEC, such as the trust among UAVs from different sources and the stability of UAV formation network. In view of the problems existing, this study proposes a blockchain‐based architecture to build a system of mutual trust, fairness, openness, and stability in this scenario. Through the implementation of blockchain technology, key data such as device computing capacity, task allocation, and task execution process are recorded openly, transparently, and irrevocably. As multi‐party trust is built to reduce the occurrence of fraud, system participants can get a reasonable reward. On this basis, the smart contract is used to ensure that algorithms are accessible to the public, and the sub‐blockchain technology improves the stability of the system. In the case study, the simulation results show that the resource consumption and time cost of the proposed scheme is reasonable and feasible.