Deepfake technology, leveraging advanced artificial intelligence (AI) algorithms, has emerged as a powerful tool for generating hyper‐realistic synthetic human faces, presenting both innovative opportunities and significant challenges. Meanwhile, the development of Deepfake detectors represents another branch of models striving to recognize AI‐generated fake faces and protect people from the misinformation of Deepfake. This ongoing cat‐and‐mouse game between generation and detection has spurred a dynamic evolution in the landscape of Deepfake. This survey comprehensively studies recent advancements in Deepfake generation and detection techniques, focusing particularly on the utilization of generative adversarial networks (GANs) and diffusion models (DMs). For both GAN‐based and DM‐based Deepfake generators, we categorize them based on whether they synthesize new content or manipulate existing content. Correspondingly, we examine various strategies employed to identify synthetic and manipulated Deepfake, respectively. Finally, we summarize our findings by discussing the unique capabilities and limitations of GANs and DM in the context of Deepfake. We also identify promising future directions for research, including the development of hybrid approaches that leverage the strengths of both GANs and DM, the exploration of novel detection strategies utilizing advanced AI techniques, and the ethical considerations surrounding the development of Deepfake. This survey paper serves as a valuable resource for researchers, practitioners, and policymakers seeking to understand the state‐of‐the‐art in Deepfake technology, its implications, and potential avenues for future research and development.

Accurate traffic flow prediction is crucial for improving transportation efficiency. To improve the accuracy of traffic flow prediction, we developed a traffic flow prediction framework—namely, traffic flow multicomponent network—that appropriately processes the noise, volatility, and nonlinearity in traffic flow data. This framework comprises three components: a factor selection component, traffic flow decomposition component, and traffic flow prediction component. The factor selection component considers the dynamic effects of weather‐related, environmental, and spatiotemporal factors on traffic flow; it then extracts and analyzes factors exhibiting strong correlations with traffic flow. The traffic flow decomposition component optimizes the parameters of variational mode decomposition on the basis of the envelope entropy by using the sparrow search algorithm; it then transforms traffic flow into multiple intrinsic mode functions to enable accurate traffic flow prediction. Finally, the traffic flow prediction component constructs dynamic feature matrices by using a bidirectional gated recurrent unit model to identify relationships within the data. Moreover, it uses an attention mechanism to assign different weights to different features on the basis of the importance of these features to traffic flow prediction, thereby enabling the efficient processing of a large volume of data. The performance of the proposed framework was examined in experiments conducted on large volumes of traffic flow data with different time granularities. The results indicated that the proposed framework achieved high prediction accuracy and stability for various time granularities, data samples, dataset sizes, and noise conditions. Moreover, it generally outperformed existing traffic flow prediction models under all experimental conditions.

Bipolar neutrosophic soft sets are powerful tools for modeling data under conditions of uncertainty and imprecision due to their rich parametric structure and the useful mathematical properties of the operations defined on them. In this paper, motivated by the limitations of existing decision‐making algorithms, we introduce a new numerical characteristic, the energy of a bipolar neutrosophic soft set defined using singular values, analogous to the graph energy and nuclear norm. Our goal is to develop an efficient decision‐making algorithm that successfully identifies the optimal alternative even in cases where other algorithms provide inaccurate or inconsistent results. Our research is motivated by the need for more reliable decision‐making methods in complex soft environments and the potential of the energy‐based approach to overcome the weaknesses of existing methods, which we demonstrate through a comparative analysis using concrete examples.

The rapid development of deepfake technology has led to the generation of a large amount of tampered video and image content, posing a major challenge to content authenticity verification. In particular, detecting deepfakes in image sequences (e.g., agricultural product packaging) is particularly difficult because the anomalies introduced by the tampering techniques are often subtle and temporally continuous. In this paper, we propose a new deepfake detection method based on time series, combining independent component analysis (FastICA) with anomaly detection techniques. We first apply FastICA to extract independent components from image sequences to identify anomalous visual patterns that are unique to deepfake tampering. In addition, we use an efficient anomaly detection algorithm, LSHiforest, to achieve scalable and accurate identification of suspicious sequences. Experimental results show that the proposed method can still detect deepfake content with high accuracy in challenging scenarios with complex temporal dynamics. Our work provides a promising solution for real‐time and large‐scale detection of deepfake content in dynamic media.

In e‐commerce, customer reviews wield significant influence over business strategies. This study proposes a robust sentiment analysis (SA) model tailored to e‐commerce recommendations. It aims to address the key limitations of existing methods, including challenges in generalizability, feature extraction, class imbalance, and hyperparameter tuning. Our process uses an autoencoder (AE) to extract key features from the input sentence. We employ DistilBERT for word embedding, which performs faster than the standard BERT model (bidirectional encoder representations from transformers). The proposed architecture integrates an AE with a transductive long short‐term memory (TLSTM) unit, which is trained with a modified generative adversarial network (GAN). TLSTM leverages transductive learning to emphasize training samples that closely resemble those in the test distribution, enhancing the flexibility and predictive accuracy of the model. Within the GAN, the generator is designed to exclude gradients from dominant batch instances, encouraging greater output diversity and generalization. Once the AE is trained, its compressed feature representations are fed into a multilayer perceptron (MLP) classifier. To tackle class imbalance issues during classification, we implement a reinforcement learning (RL) mechanism. This strategy prioritizes the minority class by applying a reward mechanism to balance the classification outcomes. Moreover, we use the Bayesian optimization hyperband (BOHB) algorithm to fine‐tune the hyperparameters of the model. Experimental results on the AIV, AA, and Yelp datasets demonstrate superior performance, with F‐measure scores of 91.603%, 89.504%, and 90.397%, respectively. These outcomes validate the robustness of the model and its potential to significantly enhance recommendation quality in dynamic e‐commerce environments.

As Smart City (SC) infrastructures evolve rapidly, securing critical systems like smart water management (SWM) becomes paramount to protecting against cyber threats. Enhancing the security, sustainability and execution of conventional schemes is considered significant in developing smart environments. Intrusion detection systems (IDS) can be effectively leveraged to realise this security objective in an Internet of Things (IoT)‐based smart environment. This research addresses this need by proposing a novel IDS model called TabNet architecture optimised using Sailfish Optimisation (SFO). The TabNet‐SFO model was specifically developed for SWM in SC applications. The proposed IDS model includes data collection, preprocessing, feature selection and classification processes. For training the model, this research used the CIC‐DDoS‐2019 dataset, and for evaluation, real‐time data collected using an IoT‐based smart water metre are used. The preprocessing step eliminates unnecessary features, cleans the data, encodes labels and normalises the applied datasets. After preprocessing, the TabNet model selects significant features in the dataset. The TabNet architecture was optimised using the SFO algorithm, which allows hyperparameter tuning and model optimisation. The proposed model demonstrated improved detection accuracy and efficiency on both the simulated and real‐time datasets. The model attained a 98.90% accuracy, a 98.85% recall, a 98.80% precision, a 98.82% specificity and a 98.78% f1 score on the CIC‐DDoS dataset and a 99.21% accuracy, a 99.02% recall, a 99.05% precision, a 99.10% specificity and a 99.18% f1 score on real‐time data. Compared to existing models, the TabNet‐SFO model outperformed all existing models in terms of performance metrics and validated its efficiency in detecting attacks.

Lithium batteries are one class of key components in new‐energy vehicles, and surface defects are easily generated during production, causing serious threats to safety. Most deep learning methods of surface defect detection heavily rely on lots of high‐quality labeled samples. Unfortunately, it is very difficult and expensive to prepare defect datasets of lithium batteries in practice. To deal with this issue, this paper presents cross‐object transfer learning (COTL)–based few‐shot surface defect detection of lithium batteries by resort to massive defect samples of other objects. The COTL model is composed of image preprocessing, feature extraction, feature fusion, and contrastive learning‐based defect detection modules. The ResNeXt‐101 network is used as the backbone to enhance feature extraction capability. The path aggregation feature pyramid network (PAFPN) is used to realize multiscale feature fusion. The contrastive learning branch is added to improve the discrimination ability among different categories of region proposals under few defect samples and increase the generalization ability. Then, experiments are done to testify the proposed method, where base‐class defect dataset from other objects and new‐class defect dataset from soft‐pack lithium batteries are adopted for training and testing. Furthermore, model comparison and ablation studies are performed. The results show that the recall rate, the AP50, the mAP, and the F1 values of the COTL model are much better than those of other existing models when only using few defect samples. In particular, when there are only 30 new‐class defect samples, the above four metrics of the COTL model are already larger than 0.90. The results testify that the proposed COTL model provides a more effective solution for few‐shot surface defect detection of lithium batteries.

A great challenge to steganography has arisen with the wide application of steganalysis methods based on convolutional neural networks (CNNs). To this end, embedding cost learning frameworks based on generative adversarial networks (GANs) has been proposed and achieved success for spatial image steganography. However, the application of GAN to JPEG steganography is still in the prototype stage; its antidetectability and training efficiency should be improved. In conventional steganography, research has shown that the side information calculated from the precover can be used to enhance security. However, it is hard to calculate the side information without the spatial domain image. In this work, an embedding cost learning framework for JPEG image steganography via a GAN (JS–GAN) has been proposed, the learned embedding cost can be further adjusted asymmetrically according to the estimated side information (ESI). Experimental results have demonstrated that the proposed method can automatically learn a content‐adaptive embedding cost function, and using the ESI properly can effectively improve the security performance. For example, under the attack of a classic steganalyzer GFR with a quality factor of 75 and 0.4 bpnzAC, the proposed JS–GAN can increase the detection error by 2.58% over J‐UNIWARD, and the ESI–aided version JS–GAN (ESI) can further increase the security performance by 11.25% over JS–GAN.

Renewable energy forecasting is crucial for pollution prevention, management, and long‐term sustainability. In response to the challenges associated with energy forecasting, the simultaneous deployment of several data‐processing approaches has been used in a variety of studies in order to improve the energy–time‐series analysis, finding that, when combined with the wavelet analysis, deep learning techniques can achieve high accuracy in energy forecasting applications. Consequently, we investigate the implementation of various wavelets within the structure of a long short‐term memory neural network (LSTM), resulting in the new LSTM wavelet (LSTMW) neural network. In addition, and as an improvement phase, we modeled the uncertainty and incorporated it into the forecast so that systemic biases and deviations could be accounted for (LSTMW with luster: LSTMWL). The models were evaluated using data from six renewable power generation plants in Chile. When compared to other approaches, experimental results show that our method provides a prediction error within an acceptable range, achieving a coefficient of determination ( R 2 ) between 0.73 and 0.98 across different test scenarios, and a consistent alignment between forecasted and observed values, particularly during the first 3 prediction steps.

The exploration of embodied intelligence has garnered widespread consensus in the field of artificial intelligence (AI), aiming to achieve artificial general intelligence (AGI). Classical AI models, which rely on labeled data for learning, struggle to adapt to dynamic, unstructured environments due to their offline learning paradigms. Conversely, embodied intelligence emphasizes interactive learning, acquiring richer information through environmental interactions for training, thereby enabling autonomous learning and action. Early embodied tasks primarily centered on navigation. With the surge in popularity of large language models (LLMs), the focus shifted to integrating LLMs/multimodal large models (MLM) with robots, empowering them to tackle more intricate tasks through reasoning and planning, leveraging the prior knowledge imparted by LLM/MLM. This work reviews initial embodied tasks and corresponding research, categorizes various current embodied intelligence schemes deployed in robotics within the context of LLM/MLM, summarizes the perception–planning–action (PPA) paradigm, evaluates the performance of MLM across different schemes, and offers insights for future development directions in this domain.