Artificial Intelligence (AI) technology has been increasingly integrated into Radio Access Networks (RAN), thereby boosting the intelligence capabilities of the air interface in mobile communications. Specifically, AI-driven beam management technology achieves multiple performance breakthroughs. These include spectral efficiency improvement and coverage quality enhancement, enabled by dynamic optimization of beam direction and intelligent resource multiplexing. This paper proposes a differentiated service datadriven intelligent adjustment method for 5G-A broadcast beams. By considering service-level performance requirements and traffic popularity, the method dynamically and intelligently adjusts broadcast beam parameters. The method aims to provide differential network coverage guarantee for different services under fixed resources and enhance the precision of broadcast beam adjustment.

With the rapid development of communication technologies, the need for accurate and efficient complaint auxiliary analysis (CAA) among mobile network optimization personnel is growing. However, most existing research solutions focus only on text data and structured data, with few incorporating user complaint speech data. To address this, the authors propose a CAA scheme for mobile network based on multi-modal generative Large Language Model (LLM). By integrating speech, text, and other multi-modal data, the proposed scheme aims to accurately understand and efficiently analyze user complaint information. The proposed approach consists of three key components. First, the Whisper model is employed to transform user complaint speech data into structured textual representations. Subsequently, a self-constructed domain-specific dictionary is integrated with an Attention-based mechanism to train a Key Information Extraction (KIE) model, thereby enhancing its semantic comprehension performance. Finally, keyword-based knowledge derived from knowledge graphs is combined with expert-defined rules to train a Generative Programs and Recommendations (GPR) model, enabling the system to deliver more accurate and professional Customer Assistance Automation (CAA) solutions. Experimental evaluations demonstrate that the proposed framework exhibits superior generalization capabilities.

The integration of Large Language Models (LLMs) with Graph Neural Networks (GNNs) has recently been explored to enhance the capabilities of Text Attribute Graphs (TAGs). Most existing methods feed textual descriptions of the graph structure or neighbouring nodes’ text directly into LLMs. However, these approaches often cause LLMs to treat structural information simply as general contextual text, thus limiting their effectiveness in graph-related tasks. In this paper, we introduce LanSAGNN (Language Semantic Anisotropic Graph Neural Network), a framework that extends the concept of anisotropic GNNs to the natural language level. This model leverages LLMs to extract tailor-made semantic information for node pairs, effectively capturing the unique interactions within node relationships. In addition, we propose an efficient dual-layer LLMs finetuning architecture to better align LLMs’ outputs with graph tasks. Experimental results demonstrate that LanSAGNN significantly enhances existing LLM-based methods without increasing complexity while also exhibiting strong robustness against interference.

The dissemination of misinformation on social media is likely to lead to severe social conflicts, therefore, the prediction of individual responses to news events becomes a significant task of social importance. The existing belief-centered method overlooks the fact that people tend to respond similarly to certain types of news. To tackle this issue, we propose a Multi-View framework with Dynamic Evidential Fusion(MVDEF) for Response Forecasting. To carry out the task, we first utilize Large Language Models to extract news topics and user beliefs from the news content and user profiles. Then, we construct three single-view graphs: a user-news interaction graph, a belief-aware graph, and a topic-aware graph. We dynamically evaluate the reliability of each view for different samples and then integrate the results based on their corresponding uncertainty mass. Additionally, we introduce a pseudo-view to enhance the interaction between these three views. Extensive experiments demonstrate that our model achieves excellent performance on real-world Twitter data. Further analysis reveals the model’s capability in unseen user scenarios, underscoring its practical applicability in real-world problems.

Recently, Deepfake has posed a significant threat to our digital society. This technology allows for the modification of facial identity, expression, and attributes in facial images and videos. The misuse of Deepfake can invade personal privacy, damage individuals’ reputations, and have serious consequences. To counter this threat, researchers have proposed active defense methods using adversarial perturbation to distort Deepfake products which can hinder the dissemination of false information. However, the existing methods are primarily based on image-specific approaches, which are inefficient for large-scale data. To address these issues, we propose an end-to-end approach to generate universal perturbations for combating Deepfake. To further cope with diverse Deepfakes, we introduce an adaptive balancing strategy to combat multiple models simultaneously. Specifically, for different scenarios, we propose two types of universal perturbations. Disrupting Universal Perturbation (DUP) leads Deepfake models to generate distorted outputs. In contrast, Lapsing Universal Perturbation (LUP) tries to make the output consistent with the original image, allowing the correct information to continue propagating. Experiments demonstrate the effectiveness and better generalization of our proposed perturbation compared with state-of-the-art methods. Consequently, our proposed method offers a powerful and efficient solution for combating Deepfake, which can help preserve personal privacy and prevent reputational damage.

Facial manipulation techniques pose a significant threat to society due to the prevalence of deepfake content on the internet. While previous efforts have focused on developing accurate deepfake detection models, these models may be limited in real-world scenarios due to the lack of confidence that human analysts have in their results. Therefore, this study presents a novel approach to improve the practicality of deepfake detection models by incorporating human understanding. We propose a human prompt based deepfake detection framework that overlays Human-enhanced artifacts attention onto image artifact attention, which utilizes vision prompts to improve the model’s responsiveness and feedback ability while preserving its precision and generalizability. The deepfake detection model achieves an AUC score of 0.99 on the FaceForensics++ dataset and exhibits graceful generalization when evaluated on the Celeb-DF dataset. Furthermore, the model generates "possible area of manipulation" that provides an intuitive signal to facilitate interpretation of the detection process, bridging the gap between machine and human perception of "fake". Our proposed approach can potentially mitigate the harm caused by deepfakes and provide a more reliable solution for real-world applications.

Leveraging user behavior has become an increasingly valuable resource for modeling and personalizing systems based on the unique characteristics of each user. While recent studies have recently been conducted along these lines, there is still a lack of understanding of the relationship between students’ Myers-Briggs Type Indicators and their behavior within educational systems. Facing this problem, we conducted a long-term study (15 weeks) with 96 students, analyzing how their engagement metrics and communication frequency in a Moodle Learning Management System are related to their Myers-Briggs personality types (i.e., extroversion/introversion, sensing/intuition, thinking/feeling, and judging/perceiving). The primary findings indicate that i) participants identified as extroverted demonstrated heightened activity levels throughout more weeks of the course, and ii) participants characterized by judging and thinking traits engaged in a greater number of activities over the course duration. The results contribute to the field of educational technologies by providing valuable insights into the relationships between different characteristics associated with the Myers-Briggs Type Indicator and students’ behavior when using an educational system.

Gamification is commonly employed to support the formation of positive psychological states and learning outcomes, with one of the primary psychological factors chiefly relevant to learning being the flow state. However, the effects of gamification on students’ flow experience are still little known. Filling this gap, we conducted a between-subjects controlled experiment (N = 65) to analyze the effects of gamification on students’ flow experience. Using descriptive and inferential statistical techniques, we compared the flow experience between participants who used a gamified version of an educational system (experimental group) and a group that used the same system without gamification (control group). The main results indicate that the employed gamification design did not affect students’ flow experience. Our study contributes especially to educational technologies and gamification fields, demonstrating that gamification may not affect students’ flow experience.

Deepfake detectors often encounter performance degradation when tested on unseen forgery methods. Existing literature tries to capture common features among multiple source forgery domains. However, we show that conflict arises in the shared feature space when each domain expresses domain-specific bias. If left unresolved, this conflict might mislead the model to learn domain-specific features and lead to inferior generalization. In this paper, we propose a new learning approach, Conflict Resolving (ConfR), designed to minimize conflict and learn features that generalize across forgeries. ConfR incorporates two key elements: the Intra-Domain Consistency Preserving (ICP) loss ensures updating consistency within forgery types, and the Inter-Domain Conflict Resolving (ICR) Module resolves updating conflicts between different forgery types. Extensive experiments demonstrate that ConfR significantly improves upon the state-of-the-art method, highlighting its potential for more generalizable deepfake detection.

Facial manipulation techniques pose a significant societal threat due to the widespread dissemination of deepfake content on the internet. Existing efforts for deepfake detection exhibit inadequate generalization performance when encountering unseen or degraded samples. We attribute this limitation to the overfitting of minor forgery patterns and variations in data distribution among disparate datasets. To tackle this issue, we introduce an innovative human-perception-centric incremental deepfake detection framework to enhance the generalization capabilities of deepfake detection models through continuous learning from a limited set of new samples. Firstly, the model leverages human perceptual salience to discern and comprehend significant artifacts, thereby mitigating overfitting to minor features. Subsequently, in the incremental learning process, we utilize multi-perspective knowledge distillation and a replay strategy to maintain the performance of the old model and minimize the feature distance between old and new samples. This comprehensive approach mitigates feature-level overfitting and addresses distribution differences among various datasets in the incremental phase. We conducted thorough experiments on four benchmark datasets (FF++, DFDC-P, CDF2, and DFD), and the experimental results demonstrate the superior performance of our method.

Facial manipulation techniques pose a significant societal threat due to the prevalent presence of deepfake content online. Current deepfake detection methods demonstrate subpar generalization performance when applied to unseen samples. The cause of this limitation lies in the overfitting of minor forgery patterns and variations in data distribution across different datasets. To tackle this issue, we introduce an innovative Human-perception-centric Deepfake Detection Adapter, namely HDDA, to enhance the generalization ability of deepfake detection models. This adaptation primarily involves two stages. During the pre-training stage, the model utilizes human perception salience to spot significant artifacts, thus reducing overfitting to minor features. In the subsequent fine-tuning stage, we introduce an efficient parameter tuning module named Deepfake Detection Adapter. The Adapter introduces two types of lightweight yet specialized adapter modules to the pre-trained model while keeping the backbone network frozen. It fine-tunes the pre-trained model through the adapter to adapt new and unseen datasets, thereby enhancing generalization. We conducted comprehensive experiments on various standard deepfake detection benchmarks to validate the effectiveness of our approach, particularly in showcasing a compelling advantage under cross-dataset and cross-manipulation settings.

Knowledge distillation enhances the performance of the student model by transferring knowledge from the teacher model. Moreover, the attention mechanism has been introduced recently to enable each layer of the student to learn knowledge from all teacher layers, which brings about considerable optimization. However, noted that features from different layers, such as shallow and deep layers, might have a big semantic gap, and compulsively aligning one student layer to all teacher layers would mislead the learning process. To tackle this problem, an effective framework called Semantic Stage-Wise learning for Knowledge Distillation (SSWKD) is presented in this paper. We divide all layers into shallow and deep stages, and only allow feature alignment within the same stage to alleviate semantic mismatch. In addition, with the observation that the performance of deep networks relies more on some key features rather than evenly on all of them, a crucial feature enhancement method based on KL divergence is then proposed for SSWKD, forcing the student to pay more attention to critical features of the teacher. Extensive experiments and visualizations show that our SSWKD outperforms other distillation methods on CIFAR-100 and COCO2017 datasets for image classification, object detection, and instance segmentation tasks.

Abstract The continuous application of artificial intelligence (AI) technologies in online education has led to significant progress, especially in the field of Intelligent Tutoring Systems (ITS), online courses and learning management systems (LMS). An important research direction of the field is to provide students with customised learning trajectories via student modelling. Previous studies have shown that customisation of learning trajectories could effectively improve students’ learning experiences and outcomes. However, training an ITS that can customise students’ learning trajectories suffers from cold-start, time-consumption, human labour-intensity, and cost problems. One feasible approach is to simulate real students’ behaviour trajectories through algorithms, to generate data that could be used to train the ITS. Nonetheless, implementing high-accuracy student modelling methods that effectively address these issues remains an ongoing challenge. Traditional simulation methods, in particular, encounter difficulties in ensuring the quality and diversity of the generated data, thereby limiting their capacity to provide intelligent tutoring systems (ITS) with high-fidelity and diverse training data. We thus propose Sim-GAIL, a novel student modelling method based on generative adversarial imitation learning (GAIL). To the best of our knowledge, it is the first method using GAIL to address the challenge of lacking training data, resulting from the issues mentioned above. We analyse and compare the performance of Sim-GAIL with two traditional Reinforcement Learning-based and Imitation Learning-based methods using action distribution evaluation, cumulative reward evaluation, and offline-policy evaluation. The experiments demonstrate that our method outperforms traditional ones on most metrics. Moreover, we apply our method to a domain plagued by the cold-start problem, knowledge tracing (KT), and the results show that our novel method could effectively improve the KT model’s prediction accuracy in a cold-start scenario.

Recently, methods enabling humans and Artificial Intelligent (AI) agents to collaborate towards improving the efficiency of Reinforcement Learning - also called Collaborative Reinforcement Learning (CRL) - have been receiving increasing attention. In this paper, we provide a long-term, in-depth survey, investigating human-AI collaborative methods based on both interactive reinforcement learning algorithms and human-AI collaborative frameworks, between 2011 and 2020. We elucidate and discuss synergistic analysis methods of both the growth of the field and the state-of-the-art; we suggest novel technical directions and new collaboration design ideas. Specifically, we provide a new CRL classification taxonomy, as a systematic modelling tool for selecting and improving new CRL designs. Furthermore, we propose generic CRL challenges providing the research community with a guide towards effective implementation of human-AI collaboration. The aim is to empower researchers to develop more efficient and natural human-AI collaborative methods that could utilise the different strengths of humans and AI.

Face reenactment is a challenging task, as it is difficult to maintain accurate expression, pose and identity simultaneously. Most existing methods directly apply driving facial landmarks to reenact source faces and ignore the intrinsic gap between two identities, resulting in the identity mismatch issue. Besides, they neglect the entanglement of expression and pose features when encoding driving faces, leading to inaccurate expressions and visual artifacts on large-pose reenacted faces. To address these problems, we propose a Large-pose Identity-preserving face reenactment network, LI-Net. Specifically, the Landmark Transformer is adopted to adjust driving landmark images, which aims to narrow the identity gap between driving and source landmark images. Then the Face Rotation Module and the Expression Enhancing Generator decouple the transformed landmark image into pose and expression features, and reenact those attributes separately to generate identity-preserving faces with accurate expressions and poses. Both qualitative and quantitative experimental results demonstrate the superiority of our method.

DeepFake videos are widely distributed on social media platforms, which has seriously affected the authenticity of digital media content, calling for robust DeepFake detection methods. Although numerous detection methods are formulated as frame-based binary classification, less attention has been paid to aggregate the features over individual frames to get a video-based judgement. We observed that for the detection of DeepFake videos, three different level forgery features from frame, clip and video can complement each other. We also found that discrete, large interval sampling strategy is more suitable for DeepFake detection, which can sample more complex video scenes, including multiple subjects, diverse facial expressions and head poses. In this work, we propose a hierarchical framework, using 2D convolutional neural networks for frame-level features extraction followed by a 1D convolutional aggregator to extract clip-level and video-level features, which can comprehensively exploit three different levels of features to make decisions. Evaluation was performed on four datasets, including DFDC, Celeb-DF, FaceForensics++ and UADFV, which provides competitive results compared to other methods. Experimental results of cross-test demonstrate that our hierarchical framework has excellent generalization performance in the face of unknown datasets.

Limited aperture inverse scattering problems arise in many important applications. In this paper, we propose a new method combining the extended sampling method (ESM) and the Bayesian approach for the inverse acoustic scattering problem to reconstruct the shape of a sound-soft obstacle using the limited aperture data. The problem is formulated as a statistical model using the Bayes formula. The well-posedness is proved in the sense of the Hellinger metric. A modified ESM is proposed to obtain the obstacle location, which is critical to the convergence of the MCMC algorithm. An extensive numerical study is presented to illustrate the performance of the method.

The unforeseen mobile data explosion poses a major challenge to the performance of today's cellular networks, and cellular network is in urgent need of original solutions to handle such voluminous mobile data. Obviously, data offloading through third-party WiFi access points (APs) can effectively alleviate the issue of overload in the cellular networks with a low operational and capital expenditure. In this paper, we study the network selection problem in operator-initiate offloading in ultradense wireless networks. To enhance the mobile data offloading, a dynamic and self-adaptive method for network selection is proposed, using the attractor selection mechanism described in biological system. In our proposed algorithm, the operator enables users to dynamically select an appropriate APs according to the dynamic conditions of various available networks. Simulation results show that the proposed algorithm decreases the service delay and achieve a high offloading efficiency in delay offloading.

Twitter has become a target platform for both promoters and spammers to disseminate their messages, which are more harmful than traditional spamming methods, such as email spamming. Recently, large amounts of campaigns that contain lots of spam or promotion accounts have emerged in Twitter. The campaigns cooperatively post unwanted information, and thus they can infect more normal users than individual spam or promotion accounts. Organizing or participating in campaigns has become the main technique to spread spam or promotion information in Twitter. Since traditional solutions focus on checking individual accounts or messages, efficient techniques for detecting spam and promotion campaigns in Twitter are urgently needed. In this article, we propose a framework to detect both spam and promotion campaigns. Our framework consists of three steps: the first step links accounts who post URLs for similar purposes; the second step extracts candidate campaigns that may be for spam or promotion purposes; and the third step classifies the candidate campaigns into normal, spam, and promotion groups. The key point of the framework is how to measure the similarity between accounts' purposes of posting URLs. We present two measure methods based on Shannon information theory: the first one uses the URLs posted by the users, and the second one considers both URLs and timestamps. Experimental results demonstrate that the proposed methods can extract the majority of the candidate campaigns correctly, and detect promotion and spam campaigns with high precision and recall.

Snow courses that measure snow water equivalent (SWE) are clustered and limited in areal coverage in Idaho. This study used a cell-based geographic information system and multiple regression models to construct SWE surfaces from the snow course data by month (January to May) and by watershed. SWE was the dependent variable and location and topographic variables derived from a digital elevation model were used as the independent variables. Multiple regression performed better than the traditional interpolation methods for SWE estimation. The estimated SWE surface can be displayed at different spatial scales through neighbourhood operations, or used directly as a map layer for hydrologic modelling.