The emergence of the Internet of Agents (IoA) introduces critical challenges for communication privacy in sensitive, high-stakes domains. While standard Agent-to-Agent (A2A) protocols secure message content, they are not designed to protect the act of communication itself, leaving agents vulnerable to surveillance and traffic analysis. We find that the rich, event-driven nature of agent dialogues provides a powerful, yet untapped, medium for covert communication. To harness this potential, we introduce and formalize the Covert Event Channel, the first unified model for agent covert communication driven by three interconnected dimensions, which consist of the Storage, Timing, and Behavioral channels. Based on this model, we design and engineer Pi-CCAP, a novel protocol that operationalizes this event-driven paradigm. Our comprehensive evaluation demonstrates that Pi-CCAP achieves high capacity and robustness while remaining imperceptible to powerful LLM-based wardens, establishing its practical viability. By systematically engineering this channel, our work provides the foundational understanding essential for developing the next generation of monitoring systems and defensive protocols for a secure and trustworthy IoA.

Kaibo Huang, Jin Tan, Yukun Wei, Wanling Li, Zipei Zhang, Hui Tian, Zhongliang Yang, Linna Zhou. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

The core challenge of steganalysislies in detecting extremely weak signals. While deeper networks improve accuracy, they exacerbate the risks of overfitting and poor generalization. The root cause is attention misalignment, where models inherently focus on semantic content rather than the texture rich regions containing steganographic signals. To address this, we propose a general and lightweight Attention Misalignment Correction (AttMC) mechanism. Leveraging the natural tendency of filter detection to focus on textural regions, our mechanism guides the model's attention via a knowledge-sharing strategy. Specifically, we design an auxiliary task of random block-wise filter detection to explicitly teach the model to prioritize textural areas. This texture-centric attention is then transferred to the main task through a shared network backbone. This process not only corrects the model's attention misalignment but also steers the decision boundary toward a more robust and generalizable solution. Extensive experiments demonstrate that our plug-and play AttMC mechanism significantly improves the detection accuracy and generalization performance of existing models with a negligible increase in parameters.

This article addresses the data-driven asynchronous dynamic event-triggeredH∞tracking control problem for singularly perturbed systems (SPSs) with unknown slow dynamics and unknown bounded disturbances. First of all, considering the two-time-scale characteristic of SPSs, anH∞tracking control problem for the slow subsystem and an asymptotic stability problem for the fast subsystem are formulated via time-scale decomposition. Secondly, an asynchronous dynamic event-triggered scheme (ETS) based on dual-rate sampling is proposed to reduce the communication burden. Then, a data-based parameterized model of the augmented system is given, which consists of the reference system and the slow subsystem with unknown dynamics. Further, by combining the data-based parameterized model and employing the full-block S-procedure, the data-based co-design method of tracking controller and event-triggered matrix is developed. The overall stability analysis of the full system under the composite controller is given. Finally, the proposed scheme is verified by a Chua’s circuit and a networked DC motor control system.

Existing image steganography schemes always introduce obvious modification traces to the cover image, resulting in the risk of secret information leakage. To address this issue, an end-to-end framework for joint makeup style transfer and image steganography is proposed in this paper to achieve imperceptible higher-capacity data hiding. In the scheme, a Parsing-guided Semantic Feature Alignment (PSFA) module is designed to transfer the style of a makeup image to an object non-makeup image, thereby generating a content-style integrated feature matrix. Meanwhile, a Multi-Scale Feature Fusion and Data Embedding (MFFDE) module was devised to encode the secret image into its latent features and fuse them with the generated content-style integrated feature matrix, as well as the non-makeup image features across multiple scales, to achieve the makeup-stego image. As a result, the style of the makeup image is well transformed and the secret image is imperceptibly embedded simultaneously without directly modifying the pixels of the original non-makeup image. Additionally, a Residual-aware Information Compensation Network (RICN) is developed to compensate the loss of the secret image arising from the multilevel data embedding, thereby further enhancing the quality of the reconstructed secret image. Experimental results show that the proposed scheme achieves superior steganalysis resistance capability and visual quality in both makeup-stego images and recovered secret images, compared with other state-of-the-art schemes.

Flexible and efficient wide-load tracking in coal-fired power generation systems (CPGSs) is crucial for integrating renewable energy. To address the challenges arising from the dynamic characteristics and task distribution differences during the wide-load operation of thermal power units, this article proposes a novel hierarchical model-agnostic meta reinforcement learning (HMAMRL) framework. This framework combines inner meta-learning for quick adaptation within task categories and outer meta-learning for sharing general task knowledge, ensuring robust generalization under different load conditions. Meanwhile, an adaptive multicriterion reward function design method is proposed to dynamically balance load tracking costs, coal consumption costs, and input fluctuation costs. Moreover, a truncated proximal policy optimization (TPPO) algorithm ensures precise load control within physical constraints. Experimental results on the 160 and 1000 MW CPGSs demonstrate the effectiveness and superiority of the proposed algorithm.

Language steganography in social networks primarily focuses on embedding secret information into social media text efficiently to achieve covert communication. The misuse of such techniques could pose significant potential threats to public cyberspace, such as the spread of malicious code, commands, or viruses. Existing social text steganalysis techniques mainly focus on the analysis of individual social media texts. However, the information content in a single text is very limited, leading to poor detection performance in practical applications. To address this challenge, this paper proposes a social text steganalysis method that combines large-scale language models with common-sense knowledge graphs (STLC-KG). This method first uses knowledge graphs to expand the knowledge contained in the text under investigation, enriching its linguistic expression, and then utilizes large-scale language models to extract the linguistic features of the social text. The results of tests conducted on three mainstream social media platforms demonstrate that the proposed method significantly improves the performance of social text steganalysis.

The rapid advancement of diffusion models has significantly improved high-quality image generation, making generated content increasingly challenging to distinguish from real images and raising concerns about potential misuse. In this paper, we observe that diffusion models struggle to accurately reconstruct mid-band frequency information in real images, suggesting the limitation could serve as a cue for detecting diffusion model generated images. Motivated by this observation, we propose a novel method called Frequency-guIded Reconstruction Error (FIRE), which, to the best of our knowledge, is the first to investigate the influence of frequency decomposition on reconstruction error. FIRE assesses the variation in reconstruction error before and after the frequency decomposition, offering a robust method for identifying diffusion model generated images. Extensive experiments show that FIRE generalizes effectively to unseen diffusion models and maintains robustness against diverse perturbations.

As one of the prominent AI-generated content, Deepfake has raised significant safety concerns. Although it has been demonstrated that temporal consistency cues offer better generalization capability, existing methods based on CNNs inevitably introduce spatial bias, which hinders the extraction of intrinsic temporal features. To address this issue, we propose a novel method called Spatial Dependency Reduction (SDR), which integrates common temporal consistency features from multiple spatially-perturbed clusters, to reduce the dependency of the model on spatial information. Specifically, we design multiple Spatial Perturbation Branch (SPB) to construct spatially-perturbed feature clusters. Subsequently, we utilize the theory of mutual information and propose a Task-Relevant Feature Integration (TRFI) module to capture temporal features residing in similar latent space from these clusters. Finally, the integrated feature is fed into a temporal transformer to capture long-range dependencies. Extensive benchmarks and ablation studies demonstrate the effectiveness and rationale of our approach.

In recent years, research on information-hiding techniques based on network streaming media has focused on how to covertly embed secret information within real-time transmissions to achieve clandestine communication. The misuse of such technologies poses significant security risks, such as the dissemination of malicious codes, commands, viruses, and more. The existing methods for steganalysis of network voice streams generally face challenges in universality, exhibiting poor adaptability to steganographic detection scenarios with non-identity distributions. To address these issues, we introduce a framework named the Domain-Aware Enhanced Framework for VoIP Steganalysis (DAEF-VS), which harnesses the CutMix technology to enhance the shared steganographic domain features and employs the Domain-Aware Learning Model to fine-tune these features, thereby significantly improving generalization capabilities. Extensive experimental results demonstrate that our approach vastly surpasses existing advanced methods in terms of universality across a variety of steganographic detection scenarios.

With the rise and maturation of neural network technology, generative text steganography based on language models is gradually becoming the mainstream technique in text steganography. However, homomorphic extraction attacks and text modification attacks from third parties pose serious threats to the usability of generative text steganography. To address this issue, this paper proposes a generative text steganography algorithm framework based on error correction codes. This framework enhances the robustness and security of steganography by encoding the secret information. Experimental results verify that the proposed framework achieves the expected outcomes and exhibits a certain degree of generality.

Linguistic steganography is a key information hiding technique but faces challenges like abrupt content shifts, detection risks, and high training resource demands. To address these, this paper introduces SCF-Stega, a controllable method based on Semantic Communications Framework. By using a knowledge graph to guide secret encoding and dynamically adjusting large language model outputs, SCF-Stega enhances text imperceptibility and semantic coherence. Experiments show improved text quality and strong resistance to steganalysis, without needing additional training data.

Steganalysis aims to detect the presence of concealed information within seemingly normal carriers in network transmissions, playing a crucial role in maintaining cybersecurity. With the rapid development of social networks, steganalysis techniques targeting social network texts have attracted significant interest from researchers in recent years. However, existing steganalysis techniques for social texts generally focus on analyzing the statistical features of the text itself, neglecting the relational features between texts, thereby limiting their detection capabilities. In this paper, we propose a novel text steganalysis feature enhancement method—TGCA. This method considers the relational features between texts by introducing GNN, while utilizing Transformers to expand the receptive field of GNNs and incorporating a cross-attention mechanism to reduce the aggregation of noise, thus mitigating the inherent limitations of GNNs. As a result, TGCA more effectively extracts and integrates textual and topological features, enhancing the model's performance in detecting steganographic texts. Experimental results demonstrate that TGCA outperforms existing methods by better leveraging graph and textual features to identify malicious steganographic texts. Our code is available at https://github.com/PandaaKai/TGCA.

Efficient detection of steganographic text in public networks is critical for maintaining cyberspace security. Current text steganalysis algorithms focus on improving feature extraction models but face challenges with fragmented network texts in real-world environments, limiting their practical use. To address this, we propose a novel linguistic steganalysis method called KIKE, which integrates Knowledge Infusion and Knowledge Encoding. KIKE utilizes knowledge graphs to enhance semantic feature extraction and employs graph neural networks for cognitive verification. Experimental results show that KIKE significantly improves detection performance, offering practical value in information security.

The current watermark-removal neural networks (WRNNs) can effectively remove the watermarks from watermarked images without damaging their host images, which poses a significant threat to image copyright protection. As one of the most effective technologies of preventing watermarks from being removed, the watermark vaccine generally attacks the WRNNs by generating and adding the adversarial perturbations to watermarked images. However, the existing watermark vaccine schemes perturb all the pixels of watermarked images, which makes it difficult to find a good trade-off between attack efficiency and imperceptibility. To address the above issues, we propose a Dual-Population Watermark Vaccine (DPWV) scheme. In this scheme, we formulate the task of adding adversarial perturbation as a bi-objective optimization problem, and address it by decoupling the space of adversarial perturbation addition to the Intensity Population (IP)-based and Position Population (PP)-based subspaces to search for the optimal solution. The extensive experiments demonstrate that the proposed scheme significantly outperforms the state-of-the-arts in the aspects of attack efficiency and imperceptibility, simultaneously, with the improvements of 45%-55% attack efficiency and 30%-40% attack imperceptibility.

With the rapid advancement of face forgery technologies, the quality of manipulated images has significantly improved, posing a severe threat to information security. In response, deepfake detection has emerged as an effective countermeasure against the misuse of these technologies. Sequential deepfake detection,as a specialized extension, targets face images with multi-step manipulation. However, a key challenge in this task is defending against unknown image degradation that occurs during transformation, which is not widely addressed in previous research. This paper introduces a robust detection framework named RSFDF, aimed at enhancing detection capabilities when images are subjected to degradation operations. RSFDF incorporates two critical modules:ATEM and ESCM. ATEM assists the network in focusing on important features while suppressing irrelevant information; ESCM refines the attention mechanism to increase the model’s focus on edge contours, aiding in the judgment of sequential forgeries. Experiments show that RSFDF exhibits significant improvements in robustness against unknown image degradations.

Corporate credit rating serves as a crucial intermediary service in the market economy, playing a key role in maintaining economic order. Existing credit rating models rely on financial metrics and deep learning. However, they often overlook insights from non-financial data, such as corporate annual reports. To address this, this paper introduces a corporate credit rating framework that integrates financial data with features extracted from annual reports using FinBERT, aiming to fully leverage the potential value of unstructured text data. In addition, we have developed a large-scale dataset, the Comprehensive Corporate Rating Dataset (CCRD), which combines both traditional financial data and textual data from annual reports. The experimental results show that the proposed method improves the accuracy of the rating predictions by 8–12%, significantly improving the effectiveness and reliability of corporate credit ratings.

Corporate credit rating is a problem of classifying high-dimensional feature vectors. Traditional corporate credit rating methods primarily rely on statistical models, machine learning models, or neural network models, all of which face limitations in handling complex inter-company relationships and nonlinear features. Particularly in cases of insufficient sample sizes, the performance of traditional models is often compromised. This paper proposes an innovative corporate credit rating method, CoReGraph-CR, which integrates Transformer, Graph Neural Network, and contrastive learning techniques to address these limitations. The contrastive learning method effectively addresses the issue of limited labeled data, improving the model’s performance with small sample sizes. Experimental results demonstrate that CoReGraph-CR outperforms existing GNN-based corporate credit rating methods in terms of classification accuracy, recall, and F1 scores, with significant improvements across various credit rating levels. Additionally, interpretability experiments enhance the model’s transparency, making it more trustworthy for practical applications.

The rapid advancement of large language model (LLM) technology has facilitated its integration into various domains of professional and daily life. However, the persistent challenge of LLM hallucination has emerged as a critical limitation, significantly compromising the reliability and trustworthiness of AI-generated content. This challenge has garnered significant attention within the scientific community, prompting extensive research efforts in hallucination detection and mitigation strategies. Current methodological frameworks reveal a critical limitation: traditional uncertainty quantification approaches demonstrate effectiveness primarily within conventional question-answering paradigms, yet exhibit notable deficiencies when confronted with non-canonical or adversarial questioning strategies. This performance gap raises substantial concerns regarding the dependability of LLM responses in real-world applications requiring robust critical thinking capabilities. This study aims to fill this gap by proposing an uncertainty quantification scenario in the task of generating with multiple facts. We have meticulously constructed a set of trap questions contained with fake names. Based on this scenario, we innovatively propose a novel and robust uncertainty quantification method(RU). A series of experiments have been conducted to verify its effectiveness. The results show that the constructed set of trap questions performs excellently. Moreover, when compared with the baseline methods on four different models, our proposed uncertainty quantification method has demonstrated great performance, with an average increase of 0.1-0.2 in ROCAUC values compared to the best performing baseline method, providing new sights and methods for addressing the hallucination issue of LLMs.

Fine-grained visual classification aims to distinguish visually similar categories and plays a critical role in practical applications. However, it faces challenges such as difficulty in localizing key regions and interference from background noise. Most methods fail to exploit semantic information effectively. This work propose a fine-grained classification method based on Semantically Guided Key Region Localization(SG-KRL). We first use deep global semantic information to guide feature fusion, enabling precise localization and adaptive cropping of the primary object region. Then, we apply semantic weighting in both channel and spatial dimensions to enhance object features, followed by multi-scale window selection to identify key parts. Finally, a graph-based representation is constructed using the semantic relationships of window positions, ensuring reliable classification. Experiments on three public benchmarks show our method balances accuracy and efficiency, outperforming existing methods. Additionally, we construct a luxury goods dataset containing genuine and counterfeit samples. Experiments on this dataset show high precision, demonstrating the method’s effectiveness in identifying subtle craftsmanship differences. Therefore, it can also contribute to intellectual property protection and combating illicit trade.

In natural language processing (NLP), the focus has shifted from encoder-only tiny language models like BERT to decoder-only large language models(LLMs) such as GPT-3. However, LLMs' practical application in the financial sector has revealed three limitations: (1) LLMs often perform worse than fine-tuned BERT on discriminative tasks despite costing much higher computational resources, such as market sentiment analysis in financial reports; (2) Application on generative tasks heavily relies on retrieval augmented generation (RAG) methods to provide current and specialized information, with general retrievers showing suboptimal performance on domain-specific retrieval tasks; (3) There are additional inadequacies in other feature-based scenarios, such as topic modeling. We introduce FinBERT2, a specialized bidirectional encoder pretrained on a high-quality, financial-specific corpus of 32b tokens. This represents the largest known Chinese financial pretraining corpus for models of this parameter size. As a better backbone, FinBERT2 can bridge the gap in the financial-specific deployment of LLMs through the following achievements: (1) Discriminative fine-tuned models (Fin-Labelers) outperform other (Fin)BERT variants by 0.4%-3.3% and leading LLMs by 9.7%-12.3% on average across five financial classification tasks. (2) Contrastive fine-tuned models (Fin-Retrievers) outperform both open-source (e.g., +6.8% avg improvement over BGE-base-zh) and proprietary (e.g., +4.2% avg improvement over OpenAI's text-embedding-3-large) embedders across five financial retrieval tasks; (3) Building on FinBERT2 variants, we construct the Fin-TopicModel, which enables superior clustering and topic representation for financial titles. Our work revisits financial BERT models through comparative analysis with contemporary LLMs and offers practical insights for effectively utilizing FinBERT in the LLMs era.

Current steganographic techniques predominantly focus on single-domain security designs, while neglecting the fact that cross-domain conversions between spatial and frequency domains may compromise embedded features, introducing detectable noise and artifacts that render stego images vulnerable to steganalyzers. This letter proposes a high-performance image steganography by dual-domain adversarial training to enhance both the security and image quality in the spatial and JPEG domains. The proposed method employs a dual-domain adversarial training strategy, integrating spatial and JPEG-domain steganalyzers to guide the generator toward producing compression-resilient stego images. In addition, a dual-objective loss function is introduced, consisting of a spatial fidelity loss to ensure visual imperceptibility and a frequency-domain consistency loss to mitigate compression-induced distortions. This design enables the model to effectively learn domain-aware embedding strategies, thereby achieving enhanced cross-domain robustness and security. Extensive experiments demonstrate that the proposed method outperforms other advanced image steganographic methods in terms of security and robustness.

Light-field (LF) images contain rich visual information and have broader application scenarios than traditional images. However, their complex structure also makes their copyright protection more challenging. Currently, there are few watermarking schemes suitable for LF images, and most of them fail to restore the original image after embedding the watermark. In addition, geometric attacks remain a difficult problem in the field of LF image watermarking. In this study, we propose a multiple reversible robust LF image watermarking scheme based on code division multiplexing (CDM) and quaternion polar harmonic Fourier moments (QPHFMs). This scheme embeds multiple identical watermarks into the LF macro-pixel image, and the compensation information for information loss caused by watermark embedding is reversibly embedded into the LF sub-aperture images. The watermark can be extracted and the original LF image can be fully recovered if the image has not been attacked. The watermark can be extracted to verify the copyright ownership of the LF image even when the image has been attacked. Experimental results demonstrate that the proposed watermarking scheme is resistant to various attacks and exhibits strong robustness.

In recent years, there has been an increasing number of information hiding techniques based on network streaming media, focusing on how to covertly and efficiently embed secret information into real-time transmitted network media signals to achieve concealed communication. The misuse of these techniques can lead to significant security risks, such as the spread of malicious code, commands, and viruses. Current steganalysis methods for network voice streams face two major challenges: efficient detection under low embedding rates and short duration conditions. These challenges arise because, with low embedding rates (e.g., as low as 10%) and short transmission durations (e.g., only 0.1s), detection models struggle to acquire sufficiently rich sample features, making effective steganalysis difficult. To address these challenges, this paper introduces a Dual-View VoIP Steganalysis Framework (DVSF). The framework first randomly obfuscates parts of the native steganographic descriptors in VoIP stream segments, making the steganographic features of hard-to-detect samples more pronounced and easier to learn. It then captures fine-grained local features related to steganography, building on the global features of VoIP. Specially constructed VoIP segment triplets further adjust the feature distances within the model. Ultimately, this method effectively address the detection difficulty in VoIP. Extensive experiments demonstrate that our method significantly improves the accuracy of streaming voice steganalysis in these challenging detection scenarios, surpassing existing state-of-the-art methods and offering superior near-real-time performance.

The popularity of Deepfake technology has raised the challenge of recognizing real and fake faces. While detection methods already exist, most of them are passive forensics and face challenges of generalizability and migration. Currently, some research attempts to protect the original image by priorly inserting invisible information. However, there are still shortcomings in terms of image quality and information robustness due to information embedding, i.e., watermarking. Therefore, we employ the robustness of perceptual hash coding and combine it with information hiding techniques to propose a proactive Deepfake detection solution, referred to as PWPH in this paper. Our approach is simple and efficient: first, the image containing a face is divided into two parts: FA (face area), and NFA (non-face area). A perceptual hash code is generated from the non-face area (NFA). Then, the hash codes are embedded as watermarks into the FA. At the extraction stage, we use the same method as the encoder to retrieve the embedded watermark from FA. The watermark is then compared with the hash code generated from the NFA of the detected image. The extracted watermark is sensitive to distortion and may vanish during Deepfake processing. Experimental results validate that our method, requiring just one encoder and decoder, enables active detection and source tracking. Furthermore, its efficacy in typical Deepfake scenarios such as face swapping and expression reconstruction is confirmed through comparison with prior arts.

In this article, based upon reinforcement learning (RL) and reduced control techniques, an${H}_{\infty }$output tracking control method is represented for nonlinear two-time-scale industrial systems with external disturbances and unknown dynamics. First, the original${H}_{\infty }$output tracking problem is transformed into a reduced problem of the augmented error system. Based on zero-sum game idea, the Nash equilibrium solution is given and the tracking Hamilton–Jacobi–Isaacs (HJI) equation is established. Then, to handle the issue of unmeasurable states of the virtual reduced system, full-order system state data are collected to reconstruct the reduced system states, and the model-free RL algorithm is proposed to solve the tracking HJI equation. Next, the algorithm implementation is given under the actor–critic–disturbance framework. It is proved that the control policy obtained from reconstructed state data can make the augmented error system asymptotically stable and satisfy theL$_{\mathbf{2}}$gain condition. Finally, the effectiveness of the proposed method is illustrated by the permanent-magnet synchronous motor experiment.

The grid connection of renewable energy poses challenges to the coordinated control of coal-fired power generation systems. Model uncertainty makes model-driven methods less effective due to the lack of adaptive capability. Large inertia of thermal process leads to local aggregation of state information, and the direct grafting reinforcement learning methods will affect the learning efficiency due to insufficient data utilization. To this end, this article proposes dual-prioritized experience replay value distribution deep deterministic policy gradient (DPER-VDP3G) algorithm. Value distribution is introduced to reflect the influence of model uncertainty on the evaluation of coordinated control policy, thus improving the accuracy of prediction cost function. The DPER is designed to reduce the nonuniform sampling bias and remove redundant data to enhance sample diversity. Comparative experiments demonstrate the advantages of the proposed method for improving network training efficiency, ameliorating load tracking accuracy and speed, and reducing energy consumption.

Variational autoencoder (VAE) is widely used in tasks of unsupervised text generation due to its potential of deriving meaningful latent spaces, which, however, often assumes that the distribution of texts follows a common yet poor-expressed isotropic Gaussian. In real-life scenarios, sentences with different semantics may not follow simple isotropic Gaussian. Instead, they are very likely to follow a more intricate and diverse distribution due to the inconformity of different topics in texts. Considering this, we propose a flow-enhanced VAE for topic-guided language modeling (FET-LM). The proposed FET-LM models topic and sequence latent separately, and it adopts a normalized flow composed of householder transformations for sequence posterior modeling, which can better approximate complex text distributions. FET-LM further leverages a neural latent topic component by considering learned sequence knowledge, which not only eases the burden of learning topic without supervision but also guides the sequence component to coalesce topic information during training. To make the generated texts more correlative to topics, we additionally assign the topic encoder to play the role of a discriminator. Encouraging results on abundant automatic metrics and three generation tasks demonstrate that the FET-LM not only learns interpretable sequence and topic representations but also is fully capable of generating high-quality paragraphs that are semantically consistent.

Generalizing deepfake detection has posed a great challenge to digital media forensics, as inferior performance is obtained when training sets and testing sets are domain-mismatched. In this paper, we show that a CNN-based detection model can significantly improve performance by fixing domain bias. Specifically, we propose a novel Fixing Domain Bias network (FDBN). FDBN does not rely on manual features, but is based on three core designs. Firstly, a domain-invariant network based on randomly stylized normalization is devised to constrain the domain discrepancy in the feature space. Then, through adversarial learning, a generalizing representation in the stylized distribution is learned to enhance the shared feature bias among manipulation methods in the domain-specific network. Finally, to encourage equality of biases among different domains, we utilize the bias extrapolation penalty strategy by suppressing the expected bias on the extremely-performing domains. Extensive experiments demonstrate that our framework achieves effectiveness and generalization towards unseen face forgeries.

In recent years, generating talking faces has become a popular research area due to their applications in various fields. However, most current models require high computational demands, which limits their practicality. To address this issue, some researchers have developed phoneme-face indexes to generate talking videos quickly and efficiently. But when the training video is too short, it is not possible to create mappings for all phonemes. To overcome this limitation, we introduced a large-scale phoneme-face dictionary to complete the feature mapping, designed a novel method for fast phoneme-face indexes search and trained a generative adversarial network (GAN) to generate video from phoneme-face sequences. Our proposed method is capable of completing the phoneme-face mapping using less than 10 seconds training video of the target person based on the large-scale dictionary and fast search algorithm and reducing the preprocessing and training time for talking videos generation.

In the field of reversible data hiding (RDH), designing a high-precision predictor to reduce the embedding distortion and developing an effective embedding strategy to minimize the distortion caused by embedding information are the two most critical aspects. In this paper, we propose a new RDH method, including a predictor based on a transformer and a novel embedding strategy with multiple embedding rules. In the predictor part, we first design a transformer-based predictor. Then, we propose an image division method to divide the image into four parts, which can use more pixels as context. Compared with other predictors, the transformer-based predictor can extend the range of pixels for prediction from neighboring pixels to global ones, making it more accurate in reducing the embedding distortion. In the embedding strategy part, we first propose a complexity measurement with pixels in the target blocks. Then, we develop an improved prediction error ordering rule. Finally, we provide an embedding strategy including multiple embedding rules for the first time. The proposed RDH method can effectively reduce the distortion and provide satisfactory results in improving the visual quality of data-hidden images, and experimental results show that the performance of our RDH method is leading the field.

This article investigates the tracking control problem for a class of nonlinear multi-input-multi-output (MIMO) uncertain singularly perturbed systems (SPSs) with full-state constraints. The underlying issues become more challenging because two-time-scale characteristics and full state constraints are involved. To this end, first, the adaptive neural network (NN) control method is designed to handle system uncertainties in the design process. Second, the nonlinear state-dependent coordinate transformation functions are employed to avoid the violation of full-state constraints and feasibility conditions for intermediate controllers. Furthermore, by introducing an appropriate ε -dependent Lyapunov function, the potential ill-conditioned numerical problems in the design process of SPSs are avoided, and the stability of the nonlinear SPSs is proven. Finally, two examples are presented to illustrate the validity of the proposed adaptive NN control scheme.

Face forgery instances involving celebrities are on the rise, owing to the ease with which their large quantity of videos may be accessible on the Internet, world leaders particularly. While current face manipulation detectors have achieved impressive results on several open datasets, which incorporate persons with various identities, they show performance degradation on these high-quality ones targeting at celebrities. What is more, these online videos usually undergo compression processing, marking the detection task harder. Besides, more face manipulation techniques arise for celebrities other than face-swap, such as lip-synchronize and image-animation, with which most works have not been concerned. This paper proposes a dual stream learning facial and speaking patterns method to protect celebrities against deepfakes. We design an action unit module based on facial action coding system along with anAction Unit Transformer(AUT) to exploit facial expressions embeddings. Besides, our method's dual stream architecture utilizes aTemporal Convolutional Network(TCN) to extract lip motion pattern and learns the relatedness between facial and speaking patterns. Our method could protect the person of interest (POI) against deepfakes in an end-to-end manner. Extensive experiments show that our method achieves better performance and has a higher resistance to video compression than state-of-the-art detection models.

Convolutional Neural Networks (CNN) based detectors perform well in face manipulation detection, but are still limited by redundant information. Some methods focus on blending boundary to localize manipulation regions, discarding a part of useless information like background of image. But these methods still contain deceptive information such as facial regions without texture, which occupies resources and affects detection accuracy. Besides, these methods left out some features useful for identification. Therefore, this paper proposes a module by conducting semantic masks to guide detectors focus on face. The semantic segmentation masks focus on the facial features such as hair, eyes and other important areas, which can offer effective face identification high level semantic features. Our method uses masks as an attention-based data augmentation module and is simple for many DeepFake detection models to integrate. Experiments on multiple detectors with and without our module show our module's effectiveness. Without modifying their structural design, our approach enables CNN-based detectors to perform better. Especially, our method is well-suited for protecting the person of interest against face forgery.

In order to deal with optimal setpoint tracking (OST) problems, a discounted cost function has been introduced in the existing work. However, the optimal tracking controllers developed according to the discounted cost function may not ensure asymptotic tracking and the stability of the closed-loop systems. To overcome these limitations, in this article, we propose a novel adaptive optimal control method to minimize a cost function without a discount factor. The proposed method starts from a reformulation of the infinite-horizon OST problem for linear discrete-time systems with external disturbances. We derive an algebraic Riccati equation for solving the OST problem, whose solution is uniquely determined under mild conditions. It is proved that the obtained controller accommodates the disturbance and realizes the output tracking with zero steady-state error. In the framework of reinforcement learning, a$Q$-learning algorithm is devised to learn the suboptimal control policy by using measured data. The present learning algorithm does not require that the disturbance is measurable and can be implemented completely model-free. Finally, two examples on dc motor system and F-16 aircraft plant are provided to corroborate our design methodology.

Tag deduplication is an emerging technique to eliminate redundancy in cloud storage, which works by signing integrity tags with a content-associated key instead of user-associated secret key. To achieve public auditability in this scenario, the linkage between cloud users and their integrity tags is firstly re-established in current solutions, which provides a potential side channel to malicious third-party auditor to steal the existence privacy of a certain target file. Such kind of attack, which is also possible among classic public auditing schemes, still cannot be well resisted and is now becoming a big obstacle in using this technique. In this paper, we propose a secure aggregation-based tag deduplication scheme (ATDS), which takes the lead to consider resistance against side channel attack during the process of public verification. To deal with this problem, we define a user-associated integrity tag based on the defined content-associated polynomial and devise a Lagrangian interpolation-based aggregation strategy to achieve tag deduplication. With the help of this technique, content-associated public key is able to be utilized instead of a user-associated one to achieve auditing. Once the verification is passed, the TPA is just only able to make sure that the verified data are correctly corresponding to at least a group of users in cloud storage, rather than determining specific owners. The security analysis and experiment results show that the proposed scheme is able to resist side channel attack and is more efficient compared with the state of the art.

Rhombus predictor is an effective technique to achieve prediction error expansion based reversible data hiding. Considering the correlation of adjacent pixels, it achieves high performance prediction of the central pixel with the help of its surrounding four pixels in a rhombus cell. However, for cells with large fluctuation, such correlation is rather weak, leading to poor accuracy of prediction. In this paper, we propose a reversible data hiding scheme based on improved rhombus predictor, which takes the lead to consider consistencies along horizontal, vertical and diagonal directions of the rhombus cell simultaneously so that pixels with higher consistency are employed together to make up the predictor. To reduce the prediction error once watermark bits are not fully embedded, we further present a corresponding fluctuation based sorting strategy. The experimental results show that, with the same amount of watermark bits embedded, the proposed scheme is able to achieve better performance comparing with the classic scheme and the state-of-the art.

Reversible data hiding (RDH) into encrypted images is of increasing attention to researchers as the original content can be perfectly reconstructed after the embedded data are extracted while the content owner's privacy remains protected. The existing RDH techniques are designed for grayscale images and, therefore, cannot be directly applied to palette images. Since the pixel values in a palette image are not the actual color values, but rather the color indexes, RDH in encrypted palette images is more challenging than that designed for normal image formats. To the best knowledge of the authors, there is no suitable RDH scheme designed for encrypted palette images that has been reported, while palette images have been widely utilized. This has motivated us to design a reliable RDH scheme for encrypted palette images. The proposed method adopts a color partitioning method to use the palette colors to construct a certain number of embeddable color triples, whose indexes are self-embedded into the encrypted image so that a data hider can collect the usable color triples to embed the secret data. For a receiver, the embedded color triples can be determined by verifying a self-embedded check code that enables the receiver to retrieve the embedded data only with the data hiding key. Using the encryption key, the receiver can roughly reconstruct the image content. Experiments have shown that our proposed method has the property that the presented data extraction and image recovery are separable and reversible. Compared with the state-of-the-art works, our proposed method can provide a relatively high data-embedding payload, maintain high peak signal-to-noise ratio values of the decrypted and marked images, and have a low computational complexity.

This paper investigates the problem of fuzzy controller design for a class of Takagi-Sugeno (T-S) fuzzy singularly perturbed switched systems. By using the average dwell time approach together with the piecewise Lyapunov function technique, a set of well-conditioned sufficient conditions for the existence of controller is proposed, under which the overall switched closed-loop system is asymptotically stable. A state feedback controller depending on the singular perturbation parameter ε, which is shown to work well for all ε ∈ (0, ε0), where ε0is the stability bound of singularly perturbed systems, is developed. In addition, when ε is sufficiently small, the ε-dependent controller can be reduced to an ε-independent one. Then, an ε-independent state feedback stabilization controller design method is proposed in terms of linear matrix inequalities. Furthermore, under the controller, the stability bound estimation problem of the overall switched closed-loop system is solved. Finally, an inverted pendulum system is used to show the feasibility and effectiveness of the obtained results.

Due to its compact binary codes and efficient search scheme, image hashing method is suitable for large-scale image retrieval. In image hashing methods, Hamming distance is used to measure similarity between two points. For K-bit binary codes, the Hamming distance is an into and bounded by K. Therefore, there are many returned images share the same Hamming distances with the query. In this paper, we propose an efficient image ranking method based on bit importance of binary code. Compared with the returned images of Hamming distance, important bits of query image are detected. Then, large weights are assigned to important bits and small weights are assigned to minor bits. The advantage of this proposed method is calculation efficiency. Evaluations on two large-scale image data sets demonstrate the efficacy of our binary code ranking method based on bit importance.