Ideological isolation in online social networks, including selective exposure, echo chambers, filter bubbles, tunnel vision, and polarization, has become a central concern for computational and neural modeling of information ecosystems. With the rapid adoption of graph learning, representation learning, and feedback-driven recommender systems, a growing body of work has proposed diverse metrics and models to quantify and mitigate these phenomena. However, existing studies and surveys rely on heterogeneous definitions and incompatible measurements, making empirical findings difficult to compare and obscuring how different forms of ideological isolation arise in learning-based systems. This survey provides a computationally grounded and comprehensive review of existing approaches to defining, analyzing, measuring, and mitigating ideological isolation in online social networks. We examine the mechanisms underlying content personalization, user behavior, and network structure that drive exposure concentration and attention narrowing. We then systematically review methodological approaches for detecting and quantifying ideological isolation, covering network-, content-, and behavior-based metrics, and synthesize empirical findings across platforms to assess their applicability and limitations. We further organize computational mitigation strategies, including network-topological interventions and recommendation-level controls, compare mitigation families and their trade-offs, and examine the dual role of large language models. The key ethical considerations in the design and deployment of diversity-aware systems are also discussed. By resolving the definition-metric-intervention mismatch that characterizes existing work, this survey provides a principled foundation for the design, evaluation, and deployment of neural and learning-based systems aimed at diagnosing and mitigating ideological isolation in online social networks.

The influence maximization (IM) problem seeks to identify a set of influential seed nodes to maximise information diffusion in a network. While most existing approaches focus solely on maximizing influence spread, they often neglect fairness in the diffusion of diverse aspects of information across different communities. This oversight can lead to a biased public understanding or the exclusion of minority interests in real-world applications, such as public health messaging, political discourse, or content recommendations. To address these challenges, we define the aspect-aware fair multiobjective influence maximization (AFMOIM) problem that jointly considers three objectives: influence coverage, intercommunity fairness, and the equitable dissemination of multiple information aspects. We propose a multiobjective discrete tree seed algorithm (MODTSA) to solve the AFMOIM problem effectively. Extensive experiments on real-world networks validate the effectiveness of MODTSA, demonstrating its ability to achieve well-balanced Pareto-optimal solutions that deliver both high diffusion performance and fairness across communities and information aspects.

Symbolic music, represented as MIDI files, encapsulates intricate performance details and complex temporal and structural dependencies. Effectively modelling ultra-long MIDI sequences is essential for understanding sophisticated compositions and advancing tasks like music generation, classification, and performance analysis. However, these long MIDI sequences pose significant challenges due to their complexity, multi-track concurrency, and extensive temporal relationships. This paper introduces a novel model named LongMIDI-Net that enhances the capability of bar-level pretrained large-scale MIDI sequence understanding models, extending their effectiveness to handle complete and long MIDI sequences. The proposed approach integrates structure-sensitive models for processing bar-level segments with temporal-sensitive models to capture global relationships across entire sequences. This hierarchical design significantly reduces sequence length while maintaining high model performance. Comprehensive experiments on classification tasks across multiple datasets demonstrate the superior effectiveness of the proposed model, achieving consistently strong results. Furthermore, ablation studies highlight the advantages of bar-level segmentation over random slicing, showcasing its ability to provide a more effective and structurally coherent representation of MIDI sequences. These findings underline the importance of combining local and global information for advancing symbolic music understanding.

Detecting social media bots has become increasingly critical due to their detrimental impact on online environments. With the emergence of sophisticated large language models (LLM) such as ChatGPT, bot detection faces new challenges. These bots based on LLMs exhibit human-like behaviors, and it is difficult for traditional detection approaches to identify them effectively. Such conventional methods struggle with the advanced features associated with LLM-driven bots, which possess contextual understanding and mimic human interaction patterns. The significance of detecting LLM-driven bots lies in their increased difficulty of detection and their potential to inflict more covert harm compared with traditional bots. To address these challenges, we propose LLM-BotGuard, a novel detection model that is capable of capturing the unique features of LLM-driven bots alongside other bot characteristics through three key modules, i.e., pattern-informed feature extraction module, mixture of experts module, and graph module with graph sample and aggregation networks. Extensive experiments have been conducted to evaluate the performance of the proposed LLM-BotGuard. The results demonstrate that LLM-BotGuard significantly outperforms baseline methods in detecting LLM-driven bots. The proposed LLM-BotGuard offers a robust solution for identifying sophisticated LLM-driven bots in online social networks.

While preference-based recommendation algorithms effectively enhance user engagement by recommending personalized content, they often result in the creation of “filter bubbles.” These bubbles restrict the range of information users interact with, inadvertently reinforcing their existing viewpoints. Many studies have been dedicated to improving the recommendation algorithms to tackle this issue. Yet, approaches that maintain the integrity of the original algorithms remain largely unexplored. This article introduces the agent-based adaptive information neutrality (AAIN) model, grounded in Yin-Yang theory. The proposed novel approach targets the imbalance in information perception within existing recommendation systems. It is designed to integrate with these preference-based systems, ensuring the delivery of recommendations with neutral information. Our empirical evaluation of this model proved its effectiveness, showcasing its capacity to expand information diversity while respecting user preferences. Therefore, AAIN proves to be an effective model in reducing the adverse impact of filter bubbles on how information is consumed.

Nowadays, most music generation models are limited to accepting conditions from a single modality, whether text-based or neural signal-based. For most users, text represents the most accessible and intuitive input modality. However, this often dilutes the text's emotional characteristics, misaligning the emotional depth in the final music output. This article presents comprehensive research and improvement in the text-conditioned music generation model, named Emotion-Conditioned MusicLM. Building upon the existing text-conditioned music generation model, MusicLM, ECMusicLM is designed to generate music with a deeper emotional resonance while maintaining the high quality of musical output. Our research shows that combining text and emotional elements in music generation leads to the creation of emotionally resonant music. Through the experiments, ECMusicLM showed a notable capability in capturing implicit Valence-Arousal features from text prompts, significantly enhancing the emotional depth of the generated music. This study not only pushes the boundaries of AI in artistic creation but also opens avenues for future research in multi-modal emotional synthesis.

Abstract Recently, diffusion models have been proven to perform remarkably well in text-to-image synthesis tasks in a number of studies, immediately presenting new study opportunities for image generation. Google’s Imagen follows this research trend and outperforms DALLE2 as the best model for text-to-image generation. However, Imagen merely uses a T5 language model for text processing, which cannot ensure learning the semantic information of the text. Furthermore, the Efficient UNet leveraged by Imagen is not the best choice in image processing. To address these issues, we propose the Swinv2-Imagen, a novel text-to-image diffusion model based on a Hierarchical Visual Transformer and a Scene Graph incorporating a semantic layout. In the proposed model, the feature vectors of entities and relationships are extracted and involved in the diffusion model, effectively improving the quality of generated images. On top of that, we also introduce a Swin-Transformer-based UNet architecture, called Swinv2-Unet, which can address the problems stemming from the CNN convolution operations. Extensive experiments are conducted to evaluate the performance of the proposed model by using three real-world datasets, i.e. MSCOCO, CUB and MM-CelebA-HQ. The experimental results show that the proposed Swinv2-Imagen model outperforms several popular state-of-the-art methods.

Due to the presence of noise in crowdsourced labels, label aggregation (LA) has become a standard procedure for post-processing these labels. LA methods estimate true labels from crowdsourced labels by modeling worker quality. However, most existing LA methods are iterative in nature. They require multiple passes through all crowdsourced labels, jointly and iteratively updating true labels and worker qualities until a termination condition is met. As a result, these methods are burdened with high space and time complexities, which restrict their applicability in scenarios where scalability and online aggregation are essential. Furthermore, defining a suitable termination condition for iterative algorithms can be challenging. In this article, we view LA as a dynamic system and represent it as a Dynamic Bayesian Network. From this dynamic model, we derive two lightweight and scalable algorithms: LAonepassand LAtwopass. These algorithms can efficiently and effectively estimate worker qualities and true labels by traversing all labels at most twice, thereby eliminating the need for explicit termination conditions and multiple traversals over the crowdsourced labels. Due to their dynamic nature, the proposed algorithms are also capable of performing label aggregation online. We provide theoretical proof of the convergence property of the proposed algorithms and bound the error of the estimated worker qualities. Furthermore, we analyze the space and time complexities of our proposed algorithms, demonstrating their equivalence to those of majority voting. Through experiments conducted on 20 real-world datasets, we demonstrate that our proposed algorithms can effectively and efficiently aggregate labels in both offline and online settings, even though they traverse all labels at most twice. The code is on https://github.com/yyang318/LA_onepass .

Abstract As online social media platforms continue to proliferate, users are faced with an overwhelming amount of information, making it challenging to filter and locate relevant information. While personalized recommendation algorithms have been developed to help, most existing models primarily rely on user behavior observations such as viewing history, often overlooking the intricate connection between the reading content and the user’s prior knowledge and interest. This disconnect can consequently lead to a paucity of diverse and personalized recommendations. In this paper, we propose a novel approach to tackle the multifaceted issue of recommendation. We introduce the Dual-Observation-based approach for the Recommendation (DOR) system, a novel model leveraging dual observation mechanisms integrated into a deep neural network. Our approach is designed to identify both the core theme of an article and the user’s unique engagement with the article, considering the user’s belief network, i.e., a reflection of their personal interests and biases. Extensive experiments have been conducted using real-world datasets, in which the DOR model was compared against a number of state-of-the-art baselines. The experimental results explicitly demonstrate the reliability and effectiveness of the DOR model, highlighting its superior performance in news recommendation tasks.

As a key task of fine-grained sentiment analysis, aspect-based sentiment analysis aims to analyse people’s opinions at the aspect level from user-generated texts. Various sub-tasks have been defined according to different scenarios, extracting aspect terms, opinion terms, and the corresponding sentiment. However, most existing studies merely focus on a specific sub-task or a subset of sub-tasks, having many complicated models designed and developed. This hinders the practical applications of aspect-based sentiment analysis. Therefore, some unified frameworks are proposed to handle all the subtasks, but most of them suffer from two limitations. First, the syntactic features are neglected, but such features have been proven effective for aspect-based sentiment analysis. Second, very few efficient mechanisms are developed to leverage important syntactic features, e.g., dependency relations, dependency relation types, and part-of-speech tags. To address these challenges, in this paper, we propose a novel unified framework to handle all defined sub-tasks for aspect-based sentiment analysis. Specifically, based on the graph convolutional network, a multi-layer semantic model is designed to capture the semantic relations between aspect and opinion terms. Moreover, a multi-layer syntax model is proposed to learn explicit dependency relations from different layers. To facilitate the sub-tasks, the learned semantic features are propagated to the syntax model with better semantic guidance to learn the syntactic representations comprehensively. Different from the conventional syntactic model, the proposed framework introduces two attention mechanisms. One is to model dependency relation and type, and the other is to encode part-of-speech tags for detecting aspect and opinion term boundaries. Extensive experiments are conducted to evaluate the proposed novel unified framework, and the experimental results on four groups of real-world datasets explicitly demonstrate the superiority of the proposed framework over a range of baselines.

AI recommendation techniques provide users with personalized services, feeding them the information they may be interested in. The increasing personalization raises the hypotheses of the "filter bubble" and "echo chamber" effects. To investigate these hypotheses, in this paper, we inspect the impact of recommendation algorithms on forming two types of ideological isolation, i.e., the individual isolation and the topological isolation, in terms of the filter bubble and echo chamber effects, respectively. Simulation results show that AI recommendation strategies severely facilitate the evolution of the filter bubble effect, leading users to become ideologically isolated at an individual level. Whereas, at a topological level, recommendation algorithms show eligibility in connecting individuals with dissimilar users or recommending diverse topics to receive more diverse viewpoints. This research sheds light on the ability of AI recommendation strategies to temper ideological isolation at a topological level.

Abstract Aspect terms are opinion targets for people to express and understand opinions in reviews. Aspect terms extraction is an essential subtask in aspect-level sentiment analysis. To extract aspect terms from a sentence, existing methods mainly focus on context features generated by pre-trained models. However, these models either neglect the crucial implicit linguistic features, e.g., post-of-tag, head, and head dependency, or fail to explore sufficient valuable features for aspect term extraction, which lead to the deficiency in aspect term extraction task. To address the challenges, in this paper, we propose a novel and effective framework for aspect term extraction by integrating both contextual and linguistic features with the artificial bee colony-based feature selection method. Firstly, a novel variant of artificial bee colony is designed to identify the most valuable linguistic features to reduce the high sparsity and dimensionality of the raw dataset. Next, the selected features and context embeddings are integrated to improve the performance of aspect extraction. Finally, extensive experiments are conducted on real-world datasets, and the results exhibit that our proposed framework can outperform the competitive baselines. Compared with the latest baselines, the proposed framework achieves the comparatively higherF1 scores of 80.7%, 84.7%, 72.2%, and 74.8% on the four groups of datasets. Furthermore, the ablation study shows that the proposed method with the designed feature selection module significantly outperforms the method with the original artificial bee colony, having 4.15%, 4.4%, 4.4%, and 3.2% improvements inF1 score on all the four datasets, respectively.

Influence diffusion modelling, analysis and applications in the preference-aware context draw tremendous attention to both researchers and practitioners. Most contemporary studies typically model the influence-diffusion pheromone from a centralized perspective. In this paper, we model the bi-directional influence propagation in directed weighted networks in a distributed manner with the consideration of user preference by facilitating Agent-Based Modelling. In the proposed model, each individual’s personalized features and the social context are modelled based on the underlying social theories, i.e. social influence and the homophily effect. In addition, the model is capable of not only producing a certain range of dynamical behaviours based on different parameter constellation but also analyzing the evolutionary trends of a social network and capturing the dynamics in the environment. Another attractive feature is the training capability of agents, which enables them to adapt the personalized features. Comparing with traditional approaches, the proposed model is more suitable for handling the complex nature of influence diffusion, and demonstrates the advantages in simulating the real-world influence diffusion. Furthermore, we propose a novel seeding algorithm for influence maximization, named Enhanced Evolution-Based Backward selection. The algorithm utilizes the advantages offered by the proposed agent-based model. The experimental results reveal that the algorithm is superior to those state-of-the-art algorithms for influence maximization.

With the widespread of the Internet, the on-line social network with big data is rapidly developing over time. Many enterprises attempt to develop their business by utilizing the power of on-line social networking platforms. A considerable amount of work has focused on how to select a set of influential users to maximize a kind of positive influence in static social networks. However, networks evolve, and the topological structure changes over time. How to mine and adapt the influencers in a dynamic and large-scale environment becomes a challenging issue. In this paper, a collective intelligence model, i.e., stigmergy-based influencers miner, is proposed to investigate influential nodes in a fully dynamic environment. The proposed model is capable of analysing influential relationships in a social network in decentralized manners and identifying the influencers more efficiently than traditional seed selection algorithms. Moreover, it is capable of adapting the solutions in complex dynamic environments without any interruptions or recalculations. Experimental results show that the proposed model achieves better performance than other traditional models in both static and dynamic social networks by considering both efficiency and effectiveness.

Social influence modelling and maximization appear significant in various domains, such as e-business, marketing, and social computing. Most existing studies focus on how to maximize positive social impact to promote product adoptions based on static network snapshots. Such approaches can only increase influence in a social network in short-term, but cannot generate sustainable or long-term effects. In this research work, we study how to maintain long-term influence in a social network and propose an agent-based influence maintenance model, which can select influential nodes based on the current status in dynamic social networks in multiple times. Within the context of our investigation, the experimental results indicate that multiple-time seed selection is capable of achieving more constant impact than that of one-shot selection. We claim that influence maintenance is crucial for supporting, enhancing, and assisting long-term goals in business development. The proposed approach can automatically maintain long-lasting impact and achieve influence maintenance.