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.

Existing linguistic steganography methods primarily rely on content transformations to conceal secret messages. However, they often cause subtle yet looking-innocent deviations between normal and stego texts, posing potential security risks in real-world applications. To address this challenge, we propose a content-preserving linguistic steganography paradigm for perfectly secure covert communication without modifying the cover text. Based on this paradigm, we introduce CLstega (Content-preserving Linguistic steganography), a novel method that embeds secret messages through controllable distribution transformation. CLstega first applies an augmented masking strategy to locate and mask embedding positions, where MLM (masked language model)-predicted probability distributions are easily adjustable for transformation. Subsequently, a dynamic distribution steganographic coding strategy is designed to encode secret messages by deriving target distributions from the original probability distributions. To achieve this transformation, CLstega elaborately selects target words for embedding positions as labels to construct a masked sentence dataset, which is used to fine-tune the original MLM, producing a target MLM capable of directly extracting secret messages from the cover text. This approach ensures perfect security of secret messages while fully preserving the integrity of the original cover text. Experimental results demonstrate that CLstega can achieve a 100% extraction success rate, and outperforms existing methods in security, effectively balancing embedding capacity and security.

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 ubiquity of social media platforms facilitates malicious linguistic steganography, posing significant security risks. However, detection is severely hampered by two fundamental issues during model training. Firstly, extreme class imbalance (less than 1% steganographic samples) induces a strong decision bias. Secondly, the invisibility of generative steganography means its features are nearly indistinguishable from benign text; this similarity, compounded by their extreme rarity, leads to severe feature marginalization, where faint steganographic signals are completely overwhelmed. To directly address these optimization-level challenges, we propose FADRW (Feature-Aware Modulated and Dynamically Reweighted Loss), a novel loss function framework engineered for few-shot steganalysis. FADRW employs Dynamic Reweighting to progressively counteract decision bias, and a Feature-Aware Modulation module to structurally reshape the feature space, preventing feature marginalization by enhancing the separability of these subtle features. Extensive experiments on datasets from three real-world social platforms demonstrate that FADRW significantly outperforms state-of-the-art methods, particularly in the challenging few-shot steganographic sample scenario.

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.

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.

Visible watermarks are generally embedded into digital images to claim their ownership for copyright protection. Unfortunately, the watermark removal models based on Deep Neural Networks (DNNs) are able to remove the watermarks from watermarked images, posing a great threat to image copyright protection. To prevent the watermark from being removed, watermark vaccines, i.e., adversarial perturbations, are usually added to the watermarked images to attack the target models, making them unable to remove the watermarks. However, the existing approaches indiscriminately add the watermark vaccine to the whole image region, and have not considered the vaccine failure caused by image noises, thereby still suffering from the issues of low imperceptibility and robustness. To address the above issues, we propose an Attention-Guided Watermark Vaccine (AGWV) scheme. Specifically, we propose pixel-feature attention (PFA) to identify the proper region for adding watermark vaccine, so as to achieve high imperceptibility for the added watermark vaccine. Then, we adopt image noises to perturb the vaccinated images and further optimize the watermark vaccine to correct the attention bias caused by image noise, thereby enhancing the robustness of watermark vaccines. Moreover, we design a vaccine evaluation model to intuitively evaluate the protective performances of watermark vaccines. Extensive experiments demonstrate that the proposed AGWV outperforms the state-of-the-arts in the aspects of both imperceptibility and robustness for defending against watermark removal models. Supplementary Material is available at https://github.com/YujiangLi0v0/ICME25.git

With the widespread application of large language models (LLMs) in the field of natural language processing (NLP), copyright protection issues are becoming increasingly important.As an effective means of copyright protection, watermarking technology can help developers and users prove the copyright ownership of the model.However, existing watermarking methods struggle to optimize both watermark effectiveness and model performance simultaneously.To overcome this challenge, in this paper, we propose a backdoor watermarking method named CRMark based on Chain-of-Thought (CoT) and reinforcement learning.This method embeds backdoor symbols into the prompt of the datasets and adds copyright information as the watermark into the model response.Reinforcement learning is further employed to alleviate the performance degradation of the watermark model in normal tasks.Experimental results show that the CRMark does not reduce the model's original task performance while effectively maintaining the effectiveness of backdoor watermarks, with the watermark success rate of up to 96.5%.

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.

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.

• A novel federated architecture for industrial fault inspection. • Alternating learning maintains specialised and collective knowledge. • Prototype calibration enhances the detection of rare but critical faults. • Variational consolidation mitigates catastrophic forgetting in adaptive systems. This paper introduces Federated Evolving Computing Architecture (FECA), a novel approach for fault inspection in textile manufacturing that addresses three critical challenges in distributed environments: statistical heterogeneity, class imbalance, and catastrophic forgetting. The architecture comprises three innovative components: Alternating Federated Evolving Learning (AFEL), which enables distributed inspection nodes to maintain specialised expertise while benefiting from collective knowledge; Prototypical Imbalance Calibration (PIC), which enhances the detection of rare but critical faults by calibrating prototype representations of minority classes; and Variational Consolidation for Adaptive Retention (VCAR), which mitigates catastrophic forgetting through a dual mechanism that maintains probabilistic representations of prototype distributions while selectively preserving critical parameters. The proposed architecture provides a robust framework for continuous improvement of industrial inspection systems without compromising production continuity, offering a promising solution for quality control in modern textile manufacturing.

Linguistic steganalysis faces the challenge of increasingly high-quality stego text, making it difficult to distinguish these text from cover text. The two main issues with current methods are: 1) deep learning models tend to overfit, which hurts their ability to apply to new situations, and 2) feature fusion models don't mix different types of features effectively, leading to poor results. In this letter, we propose aTextDualAttention linguistic steganalysis methodFusingStatistical andMulti-layerSemantic features (TDA-FSMS). TDA-FSMS firstly extracts multi-layer semantic features using different encoder layers from Enhanced Representation through knowledge integration (ERNIE), combining shallow and deep features to relieve potential overfitting. TDA-FSMS also designs a text dual attention network that simultaneously maps both multi-layer semantic and statistical features into a shared high-dimensional space to bring in a smooth feature fusion. Experimental results show that the text dual attention and the multi-layer semantic fusion enable TDA-FSMS to improve steganalysis performance than existing methods.

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.

With the rapid advancements in gene editing and DNA synthesis technologies, DNA has emerged as a next-generation physical steganographic medium due to its high information capacity, robustness, and superior concealment capabilities. While this steganography can be used to protect information security, it also poses a risk of being exploited for illicit transmission of harmful information. Consequently, it is imperative to discern steganographic DNA sequences from a vast array of DNA sequences. To address this challenge, this letter introduces a high-performance DNA steganalysis framework named DS-CLF. Specifically, the DS-CLF framework leverages a Transformer encoder to extract and learn DNA features within a supervised contrastive learning framework using ingeniously constructed DNA sequence triplets. Extensive experimental results demonstrate that the DS-CLF framework is highly effective in capturing DNA sequence features, and its detection capabilities for the latest DNA steganography techniques significantly outperform the best methods to date.

Linguistic steganography based on text generation is a hot topic in the field of text information hiding. Previous studies have managed to improve the syntactic quality of steganography texts using natural language processing techniques based on deep learning, but their steganography models still lack the ability to control the semantic and contextual characteristics in texts, which is caused by the shortage of relevant information they can obtain. This results in a great decline in the imperceptibility of steganographic texts. To address the problem, we propose a context-aware linguistic steganography method based on neural machine translation called NMT-Stega. The model generates translation containing secret messages based on the neural machine translation model with semantic fusion and language model reference units. In this way, the semantics and contexts of translation are controlled by the additional semantic and contextual features acquired from the text to be translated. Also, a new encoding that combines arithmetic coding with a waiting mechanism is proposed in our model. This method solves the low embedding capacity problem of waiting mechanism while ensuring the semantic and contextual characteristics of steganographic text are less modified. Experimental results show that our model outperforms the previous models and encoding methods in semantic correlation, embedding capacity and imperceptibility.

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.

Incorporating visual knowledge into text-only dialogue systems has become a potential direction to imitate the way humans think, imagine, and communicate.However, existing multimodal dialogue systems are either confined by the scale and quality of available datasets or the coarse concept of visual knowledge.To address these issues, we provide a new paradigm of constructing multimodal dialogues as well as two datasets extended from text-only dialogues under such paradigm (RESEE-WoW, RESEE-DD).We propose to explicitly split the visual knowledge into finer granularity ("turn-level" and "entity-level").To further boost the accuracy and diversity of augmented visual information, we retrieve them from the Internet or a large image dataset.To demonstrate the superiority and universality of the provided visual knowledge, we propose a simple but effective framework RESEE to add visual representation into vanilla dialogue models by modality concatenations.We also conduct extensive experiments and ablations w.r.t.different model configurations and visual knowledge settings.Empirically, encouraging results not only demonstrate the effectiveness of introducing visual knowledge at both entity and turn level but also verify the proposed model RESEE outperforms several state-of-the-art methods on automatic and human evaluations.By leveraging text and vision knowledge, RESEE can produce informative responses with real-world visual concepts.Our code is available at https: //github.com/ImKeTT/ReSee.

Linguistic steganalysis is the technology to distinguish whether looking-innocent texts hide covert (possibly hazardous) messages. Traditional methods, dominantly focusing on internal linguistic difference in texts, are seriously challenged by the recent linguistic steganography technology that can reduce the difference to near zero. However, even via the most advanced linguistic steganography methods, due to the random and uncontrollable message bits, steganographic texts may express content against common sense knowledge. To fully employ this defect of linguistic steganography, we propose LINK, a novel Linguistic steganalysis framework with the help of external Knowledge. We link texts to the external knowledge database, and employ Graph Neural Networks (GNNs) to translate linked knowledge into knowledge features, while linguistic features will be captured by the same modules from existing methods. Knowledge features and linguistic features will be combined to make final decisions. Extensive experimental results show that owing to additional external knowledge, the proposed framework can effectively compensate for the shortcomings of existing methods.1

Covert channels serve the construction of cyberspace security. By realizing the secure transmission of data, it is widely used in political and financial fields. Blockchain covert channels have higher reliability and concealment compared to traditional network-based covert channels. However, existing blockchain covert storage channels need to create a large number of transactions to transmit covert information. Creating transactions requires a transation fee, which means that the implementation of blockchain covert storage channels requires a high cost. Besides, created transactions remain on-chain permanently, leading to the threat of covert information being detected. To overcome these limitations, we propose a blockchain covert timing channel framework. Specifically, we utilize inv and getdata messages in the Bitcoin transaction broadcast as carriers and propose three modulation modes to achieve covert channels without cost and leaving no trace. We evaluate the concealment of our modes by K-S, KLD tests, and machine learning approaches. Experimental results show the indistinguishability between traffic carrying covert information and normal traffic. Our channels promise a capacity of 2.4 bit/s.

Steganography is an important technology for ensuring the security of cyberspace and the privacy of communications. In the last decade, emerging biotechnology has made it possible for DNA to be used as a promising steganographic carrier with high hidden capacity, high imperceptibility and high feasibility. However, severe statistical distortion might appear in steganographic carriers generated by existing DNA steganographies when they are compared with the natural ones. Therefore, efforts are being made to seek an advanced strategy to generate quasi-natural steganographic carriers with a strong anti-steganalysis capability. In this work, we first thoroughly analyze and model the numerous complicated statistical properties that exist in natural DNA chains, and then utilize the LSTM model to learn the serialized statistical properties. After obtaining an optimal sequence model that highly satisfies the statistical properties of natural DNA chains, we utilize the Adaptive Dynamic Grouping (ADG) algorithm to perform information hiding. In addition, we have carried out experimental analysis and verification from the perspectives of perceptual-imperceptibility, statistical-imperceptibility, and anti-steganalysis capability, all of which show that our proposed steganography method vastly outperforms previous DNA steganographic methods, taking a successful step towards achieving higher security DNA steganography.

Due to the widespread use of text in cyberspace, linguistic steganography, which hides secret information into normal texts, develops quickly in these years. While linguistic steganography protects users’ privacy, it also has the risk of being abused to endanger network security. Therefore, its corresponding detection technology, namely linguistic steganalysis, has attracted more and more researchers’ attention in the past several years. However, most of the current linguistic steganalysis methods rely heavily on a large number of labeled samples, which presents a significant gap from real-world scenarios where labeled steganographic samples are difficult to obtain. In this paper, we proposed the Pre-trained Language model with Self-training for Few-shot Linguistic Steganalysis (LSFLS) method which effectively copes with few-shot linguistic steganalysis through a small number of labeled samples and some auxiliary unlabeled samples. Numerous experiments have proved that the proposed method can achieve high detection accuracy of linguistic steganalysis when only a few labeled samples are provided (even less than 10), significantly improving the detection ability of existing methods in few-shot scenario. Furthermore, the experimental results demonstrate that the proposed method can maintain good detection capability in the case of data source mismatch and label unbalance. We believe that our work will greatly advance the practical application of linguistic steganalysis techniques.

With the rapid development of the internet and social media, linguistic steganography can be easily abused in social networks to make considerable damage to varied aspects like personal privacy, network virus and national defense. Currently, considerable linguistic steganalysis methods are proposed to detect harmful steganographic carriers. However, almost all the existing methods fail in real social networks, since they are only devoted to the linguistic features that are extreme insufficient owing to the extreme sparsity and extreme fragmentation challenges of real social networks. In this paper, we attempt to fill the long-standing gap that the datasets and effective methods are absent for hunting steganographic texts in social network scenarios. Concretely, we construct a dataset called Stego-Sandbox to simulate the real social network scenarios, which contains texts and their relation. And we propose an effective linguistic steganalysis framework integrating linguistic features contained in texts and context features represented by these connections. Extensive experimental results demonstrate owing to the captured context features, our proposed framework can effectively compensate for shortcomings of these existing methods and tremendously improve their detection ability in real social network scenarios.

Vertical federated learning (VFL) is a privacy-preserving machine learning paradigm that can learn models from features distributed on different platforms in a privacy-preserving way. Since in real-world applications the data may contain bias on fairness-sensitive features (e.g., gender), VFL models may inherit bias from training data and become unfair for some user groups. However, existing fair machine learning methods usually rely on the centralized storage of fairness-sensitive features to achieve model fairness, which are usually inapplicable in federated scenarios. In this paper, we propose a fair vertical federated learning framework (FairVFL), which can improve the fairness of VFL models. The core idea of FairVFL is to learn unified and fair representations of samples based on the decentralized feature fields in a privacy-preserving way. Specifically, each platform with fairness-insensitive features first learns local data representations from local features. Then, these local representations are uploaded to a server and aggregated into a unified representation for the target task. In order to learn a fair unified representation, we send it to each platform storing fairness-sensitive features and apply adversarial learning to remove bias from the unified representation inherited from the biased data. Moreover, for protecting user privacy, we further propose a contrastive adversarial learning method to remove private information from the unified representation in server before sending it to the platforms keeping fairness-sensitive features. Experiments on three real-world datasets validate that our method can effectively improve model fairness with user privacy well-protected.

With the rapid development of Natural Language Processing (NLP), more and more linguistic steganography methods have appeared in recent years, which may bring great challenges to the protection of cyberspace security. Due to the powerful feature extraction capabilities of Deep neural networks (DNN) to learn semantic features of large volumes of text, traditional steganalysis methods using manual features have gradually evolved into DNN-based methods. However, whether these DNN-based steganalysis methods can extract enough carrier features to achieve efficient steganalysis so that they can completely replace traditional methods based on handcrafted features remains an open question. To explore the answer, in this paper, we propose a new steganalysis method to integrate semantic and statistical features. We use BERT to extract semantic features and TF-IDF with AutoEncoder to obtain statistical features of the input text. Finally, we design a fusion mechanism to combine these two features. The experimental results show that due to the addition of statistical features, the proposed model can significantly improve the detection performance over current DNN-based linguistic steganalysis models.

With the rapid development of natural language processing technology and linguistic steganography, linguistic steganalysis gains considerable interest in recent years. Current advanced methods dominantly focus on statistical features in semantic view yet ignore syntax structure of text, which leads to limited performance to some newly statistically indistinguishable steganography algorithms. To fill this gap, in this paper, we propose a novel linguistic steganalysis framework named SeSy to integrate bothsemantic andsyntactic features. Specifically, we propose to employ transformer-architecture language model as semantics extractor and leverage a graph attention network to retain syntactic features. Extensive experimental results show that owing to additional syntactic information, the SeSy framework effectively brings about remarkable improvement to current advanced linguistic steganalysis methods.

Heterogeneous parallel steganography (HPS) has become a new trend of current streaming media voice steganography, which hides secret information in the frames of streaming media with multiple kinds of orthogonal steganography. Because of the complexity and imperceptibility of HPS, detecting its existence is a challenge for previous steganalysis methods, especially in the case of short sliding window length and low embedding rate. In order to improve the situation, we design a fast and efficient detection method named the key feature extraction and fusion network (KFEF) based on attention mechanism. The proposed model is able to effectively extract the key characteristic of the exceptions due to steganography and fuse the extracted features for different steganographic algorithms used in HPS. Experimental results show that the proposed method significantly improves the classification accuracy in detecting both low embedding rate samples and short segment samples. In addition, the detection time consumption is shorter than other methods and meets real-time requirements. Finally, with the help of attention we can predict the approximate locations of secret information which may bring new ideas to further steganalysis.

In recent years, linguistic generative steganography has been greatly developed. The previous works are mainly to optimize the perceptual-imperceptibility and statistical-imperceptibility of the generated steganographic text, and the latest developments show that they have been able to generate steganographic texts that look authentic enough. However, we noticed that these works generally cannot control the semantic expression of the generated steganographic text, and we believe this will bring potential security risks. We named this kind of security challenges as cognitive-imperceptibility. We think this is a new challenge that the generative steganography models must strive to overcome in the future. In this letter, we conduct some preliminary attempts to solve this challenge. Experimental results show that the proposed methods can further constrain the semantic expression of the generated steganographic text on the basis of ensuring certain perceptual-imperceptibility and statistical-imperceptibility, so as to enhance its cognitive-imperceptibility.

Previous works about linguistic steganography such as synonym substitution and sampling-based methods usually manipulate observed symbols explicitly to conceal secret information, which may give rise to security risks. In this letter, in order to preclude straightforward operation on observed symbols, we explored generation-based linguistic steganography in latent space by means of encoding secret messages in the selection of implicit attributes (semanteme) of natural language. We proposed a novel framework of linguistic semantic steganography based on rejection sampling strategy. Concretely, we utilized controllable text generation model for embedding and semantic classifier for extraction. In experiments, a model based on CTRL and BERT is implemented for further quantitative assessment. Results reveal that our approach is able to achieve satisfactory efficiency as well as nearly perfect imperceptibility. Our code is available at https://github.com/YangzlTHU/Linguistic-Steganography-and-Steganalysis/tree/master/Steganography/Linguistic-Semantic-Steganography.

In recent years, linguistic steganography based on text auto-generation technology has been greatly developed, which is considered to be a very promising but also a very challenging research topic. Previous works mainly focus on optimizing the language model and conditional probability coding methods, aiming at generating steganographic sentences with better quality. In this paper, we first report some of our latest experimental findings, which seem to indicate that the quality of the generated steganographic text cannot fully guarantee its steganographic security, and even has a prominent perceptual-imperceptibility and statistical-imperceptibility conflict effect (Psic Effect). To further improve the imperceptibility and security of generated steganographic texts, in this paper, we propose a new linguistic steganography based on Variational Auto-Encoder (VAE), which can be called VAE-Stega. We use the encoder in VAE-Stega to learn the overall statistical distribution characteristics of a large number of normal texts, and then use the decoder in VAE-Stega to generate steganographic sentences which conform to both of the statistical language model as well as the overall statistical distribution of normal sentences, so as to guarantee both the perceptual-imperceptibility and statistical-imperceptibility of the generated steganographic texts at the same time. We design several experiments to test the proposed method. Experimental results show that the proposed model can greatly improve the imperceptibility of the generated steganographic sentences and thus achieves the state of the art performance.

Extracting correlation features between codes-words with high computational efficiency is crucial to steganalysis of Voice over IP (VoIP) streams. In this paper, we utilized attention mechanisms, which have recently attracted enormous interests due to their highly parallelizable computation and flexibility in modeling correlation in sequence, to tackle steganalysis problem of Quantization Index Modulation (QIM) based steganography in compressed VoIP stream. We design a light-weight neural network named Fast Correlation Extract Model (FCEM) only based on a variant of attention called multi-head attention to extract correlation features from VoIP frames. Despite its simple form, FCEM outperforms complicated Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs) models on both prediction accuracy and time efficiency. It significantly improves the best result in detecting both low embedded rates and short samples recently. Besides, the proposed model accelerates the detection speed as twice as before when the sample length is as short as 0.1s, making it a excellent method for online services.

With the growing attention on multimedia security and rapid development of natural language processing technologies, various linguistic steganographic algorithms based on automatic text generation technology have been proposed increasingly, which brings great challenges in maintaining security of cyberspace. The prevailing linguistic steganalysis methods based on neural networks only conduct linguistic steganalysis with feature vectors from last layer of neural network, which may be insufficient for neural linguistic steganalysis. In this paper, we propose a neural linguistic steganalysis scheme based on densely connected Long short-term memory networks (LSTM) with feature pyramids which can incorporate more low level features to detect generative text steganographic algorithms. In the proposed framework, words in text are firstly mapped into semantic space with a hidden representation for better exploitation of the semantic features. Then, stacked bidirectional Long short-term memory networks are ultilized to extract different levels of semantic features. In order to incorporate more low level features from neural networks, we introduced two components: dense connections and feature pyramids to enhance the low level features in feature vectors. Finally, the semantic features from all levels are fused and we use a sigmoid layer to categorize the input text as cover or stego. Experiments showed that the proposed scheme can achieve the state-of-the-art results in detecting recently proposed linguistic steganographic algorithms.

Training of the convolution neural network (CNN) is a problem of global optimisation. This study proposed a hybrid modified particle swarm optimisation (MPSO) and conjugate gradient (CG) algorithm for efficient training of CNN. The training involves MPSO–CG to avoid trapping in local minima. Particularly, improvements in the MPSO by introducing a novel approach for control parameters, improved parameters updating criteria, a novel parameter in the velocity update equation, and fusion of the CG allows handling the issues in training CNN. In this study, the authors validate the proposed MPSO algorithm on three benchmark mathematical test functions and also compared with three different variants of the baseline particle swarm optimisation algorithm. Furthermore, the performance of the proposed MPSO–CG is also compared with other training algorithms focusing on the analysis of computational cost, convergence, and accuracy based on a standard problem specific to classification applications on CIFAR‐10 dataset and face and skin detection dataset.

in this letter, we present a novel and extremely fast steganalysis method for voice over ip (voip) streams, driven by the need for a quick and accurate detection of possible steganography in VoIP streams. We firstly analyzed the correlations in carriers. To better exploit the correlations in code-words, we mapped vector quantization code-words into a semantic space. In order to achieve high detection efficiency, only one hidden layer was utilized to extract the correlations between these code-words. Finally, based on the extracted correlation features, we used the softmax classifier to categorize the input stream carriers. To boost the performance of this proposed model, we incorporate a simple knowledge distillation framework into the training process. Experimental results show that the proposed method achieves state-of-the-art performance both in detection accuracy and efficiency. In particular, the processing time of this method on average is only about 0.05% when sample length is as short as 0.1 s, attaching strong practical value to online serving of steganography monitor.

Steganalysis of the Quantization Index Modulation (QIM) steganography in VoIP (Voice-over IP) stream is conducted in this research. VoIP is a popular media streaming and communication service on the Internet. QIM steganography makes it possible to hide secret information in VoIP streams. Detecting short and low embedding rates of QIM steganography samples remains an unsolved challenge. Recently, neural network models have been demonstrated to be capable of achieving remarkable performances and be successfully applied to many different tasks. The mainstream architectures of neural network include Convolution Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), which adopt totally different ways to understand various signals. In this paper, we first indicate a proper way to combine the strengths of these two architectures and then construct a novel and unified model called CNN-LSTM network to detect QIM-based steganography. In our model, Bidirectional Long Short-Term Memory Recurrent Neural Network (Bi-LSTM) is utilized to capture long time contextual information in carriers and CNN was used subsequently to capture both local features and global ones as well as temporal carrier features. Experiments showed that our model can achieve the state-of-art result in detecting QIM-based steganography in VoIP streams.

With the rapid development of natural language processing technology in the past few years, the steganography by text synthesis has been greatly developed. These methods can analyze the statistical feature distribution of a large number of training samples, and then generate steganographic texts that conform to such statistical distribution. For these steganography methods, previous steganalysis methods show unsatisfactory detection performance, which remains an unsolved problem and poses a great threat to the security of cyberspace. In this letter, we proposed a fast and efficient text steganalysis method to solve this problem. We first analyzed the correlations between words in these generated steganographic texts. Then, we map each word to a semantic space and used a hidden layer to extract the correlations between these words. Finally, based on the extracted correlation features, we used the softmax classifier to classify the input text. Experimental results show that the proposed model can achieve a high detection accuracy, which shows a state-of-the-art performance.

With the rapid development of natural language processing technologies, more and more text steganographic methods based on automatic text generation technology have appeared in recent years. These models use the powerful self-learning and feature extraction ability of the neural networks to learn the feature expression of massive normal texts. Then, they can automatically generate dense steganographic texts which conform to such statistical distribution based on the learned statistical patterns. In this letter, we observe that the conditional probability distribution of each word in the automatically generated steganographic texts will be distorted after embedded with hidden information. We use recurrent neural networks to extract these feature distribution differences and then classify those features into cover text and stego text categories. Experimental results show that the proposed model can achieve high detection accuracy. Besides, the proposed model can even make use of the subtle differences of the feature distribution of texts to estimate the amount of hidden information embedded in the generated steganographic text.

Linguistic steganography based on text carrier auto-generation technology is a current topic with great promise and challenges. Limited by the text automatic generation technology or the corresponding text coding methods, the quality of the steganographic text generated by previous methods is inferior, which makes its imperceptibility unsatisfactory. In this paper, we propose a linguistic steganography based on recurrent neural networks, which can automatically generate high-quality text covers on the basis of a secret bitstream that needs to be hidden. We trained our model with a large number of artificially generated samples and obtained a good estimate of the statistical language model. In the text generation process, we propose fixed-length coding and variable-length coding to encode words based on their conditional probability distribution. We designed several experiments to test the proposed model from the perspectives of information hiding efficiency, information imperceptibility, and information hidden capacity. The experimental results show that the proposed model outperforms all the previous related methods and achieves the state-of-the-art performance.

In this paper we describe the creation and the presentation of interactive fractal paintings with paint systems based on VF(Video Feedback). VF occurs when a camera points to its own display. VF is able to generate evolving fractal patterns. But it is difficult to create artworks under specified subjects based on VF. To solve this difficulty, we implemented video feedback paint system by minimally modifying the bare-bones VF system. The system enables the user to draw shapes directly on a display engaged in VF. These shapes geometrically confine the VF dynamics and compose a designed 2D composition which can include specific subject matter. Artworks created using the system can generate evolving patterns with the viewer manipulating anything possible between the camera and the display yet always conform to designed 2D compositions.

A product image survey and retrieval model is constructed by the method of Kansei engineering. And a product image survey and retrieval system (PISR system) based on Kansei engineering is built according to this model, which is consisted of the product image survey subsystem and the product image retrieval subsystem. The online survey is carried out on Internet and then the product Kansei image based on consumers is extracted through various data analysis methods, according to the result of data analysis, a picture of 2D or 3D image scale is produced and it is convenient to obtain the information of product Kansei image for the designers, the product Kansei image databases are built and the designers can search the matching products by inputting image words as a reference, which makes designers and the consumers reach an agreement on product Kansei image. The application of the PISR system is universal to all kinds of products. It is successfully verified that this system can help the designers gain product orientation, conduct the process of product design and improve the overall efficiency of product development.