Jia Liu, Jiaxin Luo, Weiwen Xu, Jonathan M. Garibaldi, Xiao-Kun Wu, Yixue Hao, Min Chen. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

The proliferation of hate speech on social media presents an escalating threat to both public discourse and individual mental well-being. Traditional strategies that prioritize detection and removal often neglect to engage directly with hate speakers or address the underlying causes of their hostility. This article proposesHateMediator, a dialogue-based intervention framework that fine-tunes large language models (LLMs) to generate persuasive, context-aware counter-hate speech. The framework emphasizes two core aspects: the generation of effective counter-hate responses and their evaluation through multiturn dialogues. Our fine-tuning approach integrates tutorial-based learning with critical token guidance, enabling LLMs to recognize and reproduce strategic rhetorical patterns observed in expert interventions. To support training and evaluation, we introduce theMedHatedataset, grounded in social science theory, comprising complete dialogue records from 85 real-world hate incidents (including 255 dialogues), expert-crafted counter-responses, and feedback from the original hate speakers. Experimental results show thatHateMediatorconsistently outperforms baseline LLMs across multiple evaluation dimensions. This study advances both the technical frontier of hate speech intervention and the ethical deployment of LLMs in addressing complex social issues.

In a socially tense environment with rising emotional pressure, understanding the spread patterns of group emotions-particularly negative emotions-is crucial for identifying social risks. Extensive research has explored emotion contagion, often using propagation models where node state transitions rely on preset probabilities. However, these methods introduce randomness, making them less reflective of real-world dynamics by failing to capture individual node behaviors and interactions in emotional networks. To address this, our study introduces a novel approach integrating text-based emotion recognition with propagation models, reconstructing emotion contagion at an individual level. This model enhances traditional nodes with multihop agents driven by text emotion analysis, where agents record and respond to neighbors' emotional states. As a result, emotion spread becomes a deterministic process, with individualized infection rates reflecting node variability. We categorized nodes based on emotional states, creating corresponding agent types to form the dynamic agent-based emotion model (AEmo). Tests on real-world and scale-free networks show this method effectively predicts group negative emotion spread and provides insight into individual emotion evolution, validating the model's effectiveness.

Sleep staging is a crucial method for the evaluation of sleep quality and the diagnosis of sleep disorders. In recent years, rapid progress has been made in sleep research through the application of fabric computing and neural networks. Flexible fabric sensors introduced by fabric computing minimize the discomfort of data collection devices on individuals, while neural network-based algorithms can automatically perform sleep staging based on the collected signals. However, there are two key challenges hinder the integration of automatic sleep staging networks with fabric computing: (1) signals in fabric-based environments exhibit strong heterogeneity due to the wide range of individuals, and (2) interactions between individuals and the fabric space introduce behavioral dynamics to the system. In this paper, we propose a masked autoencoder-based sleep staging neural networks (MAESleepNet), designed to integrate automatic sleep staging algorithm with fabric space. Specifically, MAESleepNet addresses the challenge of signal heterogeneity by learning fine-grained representations from local signals. Furthermore, MAESleepNet tackle the challenge of behavioral dynamics through stochastic masking and reconstruction pre-training. Experiments were conducted on three public datasets: (1) Sleep-EDF-20, (2) Sleep-EDF-78 and (3) SHHS. MAESleepNet achieves overall accuracies of 88.9%, 85.5%, and 87.3%, respectively, outperforming other state-of-the-art models. Furthermore, feature visualization and reconstruction visualization experiments were also conducted. The results demonstrates that MAESleepNet is an effective solution to the aforementioned challenges, paving the way for seamless integration into the fabric space.

The formation of information cocoons, driven by limited disclosure and individual preferences, has resulted in the polarization of society. However, the underlying mechanisms and pathways to escape these cocoons remain unresolved. This article aims to solve it by developing an adaptive imitation process. In this process, the measurement of information cocoons across the population is based on Shannon’s information entropy, taking into account neighborhood information. Incorporating the Dirac function to formulate information distribution over networks, theoretical results are validated by numerical simulation experiments. Results show that individual backgrounds and preferences are crucial factors in the formation of information cocoons, and the severity of information cocoon production increases with an individual capacity to stick to oneself. Encouraging connections among diverse communities can effectively mitigate the intensity of information cocoons. This research contributes to the advancement of computational communication systems and offers insights toward dismantling informational boundaries.

Emerging infectious diseases pose a growing threat to human society and have sparked extensive public discussions on social media. Although numerous efforts have been made in health data mining on social media, there is a lack of focus on quantitative comparisons across multiple platforms, despite their crucial role in the holistic social communication system. This study addresses this gap by developing a generalized regression model that considers the distinct attributes of social media platforms, including short-text, long-text, and Eastern or Western orientation. Using Monkeypox as an application case, this study examines differences among platforms based on four factors: user characteristics, text topics, text emotion, and text quality. The modeling and regression results reveal significant heterogeneity in public opinion expressions across different platforms, particularly between long-text and short-text platforms. Users on short-text platforms are more exposed to diverse information and tend to be susceptible to emotionally provocative content. On the other hand, users on long-text platforms prefer in-depth discussions and show greater receptivity to content infused with positive emotions. This study reveals the information bias brought by platform differences and contributes to data-driven modeling in social communication systems.

Online social media have aroused much research interest in recent years. In contrast to previous work that focused on the detection of emerging topics, this article undertakes the prediction of active users in online social events, which is so far rarely explored. This prediction task is formulated as a binary classification problem that built on real-world tweet streams, taking Ferguson event and New York Chockhold event as examples. Then, a comprehensive user feature system is designed to characterize the events’ online participants, which includes not only basic statistical characteristics and image-pixel-level features, but also some emotional features and personality features. Next, the Weighted Random Forest (Weighted-RF) classifier is adopted to solve the classification problem. Based on the user feature system and the classifier, the experience of a previous event can be archived and applied to the prediction of later similar events. Experimental results show that the Weighted-RF trained by samples of Ferguson event can effectively predict active users in NYC event, with an AUC value around 0.8392. Besides, the image-content based personality model provides a new tool for depicting user portraits, which further contributes to the quantitative analysis of online social events.

The control of epidemics taking place in complex networks has been an increasingly active topic in public health management. In this article, we propose an efficient networked epidemic control system, where a modified susceptible-exposed-infected-vigilant (SEIV) model is first built to simulate epidemic spreading. Then, different from existing continuous resource models which abstractly map resources to parameters of epidemic models, a concrete resource description model is built to simulate real-world goods/services and their allocation. Based on the two models, a cost-constraint subset selection problem in epidemic control is identified. To solve the problem, a swarm-based stochastic optimization policy is proposed, where each particle in the swarm can determine its own solutions according to the guidance of its superior peers and historical searching experience of the whole swarm, without extra problem-relative information. Theoretical proof about system equilibrium is provided, which is consistent with experimental observations. The competitive performance of the proposed optimizer is validated by theoretical analysis and comparison experiments. Finally, an application case is provided to illustrate the practicability.

The allocation of epidemic-control resources has been an increasingly active topic in the physical world. Most existing studies focus on the allocation of abstract and continuous epidemic control resources, and then formulate differentiable convex programming problems. However, real-world resources are usually discrete materials, goods, or services, so that resource allocation problems become non-convex. As a complementary study, this paper builds three discrete resource allocation problems based on an improved Susceptible-Exposed-Infectious- Vigilant (SEIV) spread model: the cost-constraint optimization problem (CCOP), rate-constraint optimization problem (RCOP), and eradication optimization problem (EOP). Then, existing swarm-based metaheuristic algorithms are adapted to effectively solve the problems. Thereinto, the Heuristic Majority-Voting Binary Particle Swarm Optimizer (HMV-BPSO) is present, which introduces a heuristic factor which concerns the probability distribution of resources to guide the evolution of particles and helps improve the performance of original MV-BPSO. Numerical experiments are developed to verify the effectiveness of swarm- based metaheuristic algorithms on epidemic control. Results show that HMV-BPSO can produce higher-quality solutions than other algorithms.