Computer Science Review最新文献

{"title":"State-of-the-art image and video-based steganalysis techniques: A comprehensive review, challenges and future recommendations for digital forensic experts","authors":"Payal Mittal ,&nbsp;Ravinder Kaur ,&nbsp;Mukesh Dalal","doi":"10.1016/j.cosrev.2025.100852","DOIUrl":"10.1016/j.cosrev.2025.100852","url":null,"abstract":"<div><div>In this age of rapid technological progress, digital communication through multimedia such as images and videos has become one of the most prominent ways of sharing and exchanging data. However, digital communication is now often used to obscure confidential or malicious information via steganography, posing significant risks to national security and digital safety. This challenge highlights the crucial role of steganalysis for digital forensic experts, which aims to detect and retrieve concealed data encoded in multimedia files. This paper presented a comprehensive review of the state-of-the-art steganalysis techniques based on images and videos, by incorporating classical Machine Learning (ML) and advanced Deep Learning (DL) approaches, unlike previous reviews that focuses on specific domains or algorithms. The paper thoroughly examines the current detection frameworks, existing image and video benchmark datasets and steganalysis tools available for digital forensic experts. The paper provides a comparative analysis of the existing techniques by highlighting their advantages and disadvantages. Experimental evaluation is done by using the widely adopted BOSSBase dataset, on which two existing steganalysis techniques are evaluated for the demonstration of practical insights. The findings emphasize that the advanced deep learning architectures surpass conventional machine learning approaches, while also recognizing persistent issues such as dataset imbalance and generalization issues. The novelty of the presented work lies in its unified coverage of image and video steganalysis techniques from a forensic expert’s viewpoint. The paper also offers future research recommendations to enhance multimedia security and aid forensic professionals in developing next-generation detection techniques.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100852"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145509611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The rise and resilience of multivariate cryptography: Advances, pitfalls, and promising pathways","authors":"Dheerendra Mishra,&nbsp;Kuldeep Namdeo,&nbsp;Namita Srivastava","doi":"10.1016/j.cosrev.2025.100860","DOIUrl":"10.1016/j.cosrev.2025.100860","url":null,"abstract":"<div><div>Multivariate Public Key Cryptography (MPKC) offers a robust framework for secure communications in the post-quantum era by harnessing the complexity of solving multivariate quadratic polynomial equations over finite fields. This inherent difficulty not only delivers strong quantum resistance but also supports high-security applications and additional functionalities. However, these advantages come with notable trade-offs, including larger key sizes and sophisticated security evaluations. In this paper, we present a comprehensive survey of MPKC, beginning with its mathematical foundations illustrated through accessible examples and an introduction to the essential terminology and core algorithms, such as Grobner basis methods, the Berlekamp algorithm, and extended-linearization methods, etc., that underpin effective implementations.</div><div>Our work critically examines state-of-the-art encryption and signature schemes, covering approaches inspired by Oil–Vinegar, Rainbow, and HFE-based constructions, as well as some essential designs like EFLASH, HFEv, Simple Matrix Encryption, and MAYO, etc. We review contemporary attack strategies including differential, rank, direct attacks, and physical attacks, and discuss how recent cryptanalytic breakthroughs (e.g., vulnerabilities revealed in Rainbow and UOV) underscore the need for more robust design choices. Furthermore, we highlight emerging research directions by exploring innovative techniques that integrate linear codes, neural network frameworks, employing other special polynomials, and novel algebraic constructions such as Sidon spaces and the Triple Unlocking Mechanism, alongside advanced signature variants like blind, ring, and proxy signatures. By synthesising theoretical insights with practical cryptanalysis, this survey not only identifies the current limitations, such as decryption failures, excessive key sizes, and susceptibilities to MinRank, reconciliation attacks, and other attacks, but also outlines future research challenges and requirements.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100860"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145609227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Understanding data spaces: A Systematic Mapping Study of foundations, technical building blocks, and sectoral adoption","authors":"Anhelina Kovach ,&nbsp;Leticia Montalvillo ,&nbsp;Jorge Lanza ,&nbsp;Pablo Sotres ,&nbsp;Aitor Urbieta","doi":"10.1016/j.cosrev.2025.100819","DOIUrl":"10.1016/j.cosrev.2025.100819","url":null,"abstract":"<div><div>Data spaces are emerging as a key paradigm for enabling sovereign, secure, and interoperable data sharing across sectors. Beyond data governance, they represent a transformation in communication architectures—where communication is no longer merely about establishing connections, but about <em>who is allowed to share what, under which conditions, and for what purpose</em>. Despite growing attention, the research landscape remains fragmented and under-synthesized. This paper presents a Systematic Mapping Study (SMS) of 149 peer-reviewed publications, analyzing the conceptual foundations, technical building blocks, and sectoral adoption of data spaces. Following established SMS methodologies, we classify the literature across key technical themes defined by the Data Spaces Support Centre (DSSC) and assess methodological maturity, technical novelty, and application domains. Our findings show that 46.3% of studies address data value creation enablers, 30.8% focus on data interoperability, and 22.9% explore data sovereignty. The study provides a structured synthesis of current research and offers guidance for advancing federated, trust-aware communication infrastructures.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100819"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145093958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Parameterized inapproximability: From Clique to PIH","authors":"Yijia Chen ,&nbsp;Bingkai Lin","doi":"10.1016/j.cosrev.2025.100834","DOIUrl":"10.1016/j.cosrev.2025.100834","url":null,"abstract":"<div><div>Parameterized approximation, first proposed by Mike Fellows, approaches NP-hard problems by allowing the running time of an approximation algorithm to be superpolynomial in the parameter of an problem instance yet still polynomial in the size of the instance itself. One of the main open questions in the area is whether we can approximate the parameterized clique problem within some nontrivial ratio. It is also conjectured by Fellows that no such algorithms exist. In this article, we explain some recent progress on this question.</div><div>Similarly to the classical polynomial time inapproximability of the clique problem, the constraint satisfaction problem, i.e., <span>CSP</span>, plays a key role in most of the known inapproximability results of the parameterized clique problem. As a matter of fact, the parameterized inapproximability hypothesis, i.e., PIH, concerning the binary <span>CSP</span> has been long believed as a viable path towards the inapproximability of the parameterized clique problem. Although it turns out that those recent results do not rely on PIH, the method discovered for the parameterized clique problem leads to a proof of a version of PIH under the exponential time hypothesis, which we will also explain in this article.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100834"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145221411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Gender Diversity Interventions in Software Engineering: A Comprehensive Review of Existing Practices","authors":"Claudia Maria Cutrupi ,&nbsp;Letizia Jaccheri ,&nbsp;Alexander Serebrenik","doi":"10.1016/j.cosrev.2025.100812","DOIUrl":"10.1016/j.cosrev.2025.100812","url":null,"abstract":"<div><h3>Context:</h3><div>Over the years, the software engineering (SE) research community has examined the role of gender within the field, leading to numerous studies on the challenges faced by women and minorities. Despite this, there is a significant research gap on the effectiveness of various interventions designed to address these persistent challenges.</div></div><div><h3>Objective:</h3><div>This review analyzed the current state of interventions that address gender disparities in the SE context. The goal of the review is to understand the various interventions implemented in educational and industrial settings and to identify their impact on women and other targeted minorities.</div></div><div><h3>Methods:</h3><div>We conducted a systematic literature review, analyzing 33 studies that reported on interventions addressing gender disparities in SE.</div></div><div><h3>Results:</h3><div>We identified the most common interventions implemented over the years, the challenges they addressed, the organizations and target users involved, the specific actions proposed, the underlying theories, the research methodologies used, and the impact of these interventions.</div></div><div><h3>Conclusion:</h3><div>The findings revealed a lack of interventions that aim at retaining women in the SE field, with few programs investing in employees’ satisfaction with the work environment. The findings also show a lack of outreach programs to create meaningful connections with companies and provide support in finding job opportunities. Moreover, only one intervention incorporates long-term activities, and very few interventions build on a theoretical background.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100812"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049628","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Techniques in parameterized approximation","authors":"Ariel Kulik ,&nbsp;Hadas Shachnai","doi":"10.1016/j.cosrev.2025.100833","DOIUrl":"10.1016/j.cosrev.2025.100833","url":null,"abstract":"<div><div>Approximation algorithms and parameterized complexity are two classic approaches for coping with NP-hard problems. The field of parameterized approximation which combines the two approaches has flourished in recent years, with a myriad of algorithmic results as well as lower bounds. In this survey we give an introduction to the field and highlight some of the main techniques developed for the design of parameterized approximation algorithms and for deriving hardness results.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100833"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145229017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sequential recommender systems: A methodological taxonomy and research frontiers","authors":"Yanbo Zhou ,&nbsp;Gang-Feng Ma ,&nbsp;Xilin Wen ,&nbsp;Xu-Hua Yang ,&nbsp;Yi-Cheng Zhang","doi":"10.1016/j.cosrev.2025.100818","DOIUrl":"10.1016/j.cosrev.2025.100818","url":null,"abstract":"<div><div>In the era of information overload, sequential recommender systems have emerged as pivotal tools for modeling user preferences through dynamic behavioral pattern mining. These systems transcend conventional recommendation paradigms by explicitly modeling temporal dependencies in user–item interactions, preference evolution, and contextual dynamics. This study presents a methodologically structured taxonomy of sequential recommender systems through four analytical dimensions: (1) Sequential Modeling, which includes methods ranging from statistical techniques to deep learning architectures to understand user behavior patterns; (2) Temporal Dynamics Modeling, which involves time-aware collaborative filtering and deep temporal modeling; (3) Network-Enhanced Modeling, which leverages graph neural networks, heterogeneous graphs, dynamic graphs, and hypergraphs to explore structural dependencies; and (4) Robust Representation Learning, which encompasses contrastive mechanisms and techniques driven by large language models (LLMs). These algorithms focus on different aspects of sequential recommendation, including but not limited to capturing dynamic interests, modeling long- and short-term preferences, and addressing issues such as data sparsity, noise, and bias, which affect the performance and user experience of recommender systems in practical applications. Furthermore, we summarize and discuss promising future research directions to provide theoretical and methodological insights. The constructed taxonomy not only organizes existing methodological innovations, but also reveals fundamental limitations in current evaluation protocols, providing a roadmap for advancing both theoretical foundations and practical applications in this domain.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100818"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From task-specific to foundation models: A paradigm shift in medical vision-language analysis","authors":"Muhammad Umair Ali ,&nbsp;Amad Zafar ,&nbsp;Seonghan Kim ,&nbsp;Kwang Su Kim ,&nbsp;Seung Won Lee","doi":"10.1016/j.cosrev.2025.100831","DOIUrl":"10.1016/j.cosrev.2025.100831","url":null,"abstract":"<div><div>Integrating vision-language models (VLMs) into medical imaging drives a paradigm shift from task-specific systems toward generalist foundation models (FMs) capable of zero-shot and few-shot reasoning across diverse clinical domains. This review presents a comprehensive model-centric taxonomy, categorizing over 135 studies into three key developmental stages: (1) task-specific VLMs, (2) modular/adapter-based/prompt-tuned VLMs, and (3) foundation models. We systematically assess each category regarding architectural innovations, learning paradigms, clinical applications, and evaluation metrics. Our analysis reveals that the recent advances in multimodal contrastive learning, prompt engineering, and scalable transformer-based architectures significantly enhance generalizability, data efficiency, and multimodal interpretability in medical AI. Furthermore, we synthesize bibliometric trends and delineate methodological transitions through a PRISMA-based systematic review. This review article concludes with a discussion on the challenges and provides a roadmap for developing clinically reliable, data-efficient, and versatile VLMs, highlighting their transformative potential for improving diagnostic accuracy, workflow automation, and decision support in healthcare.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100831"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corrigendum to “Dealing with high dimensional multi-view data: A comprehensive review of non-negative matrix factorization approaches in data mining and machine learning” [Computer Science Review 58 (2025) 100788]","authors":"Nafiseh Soleymani,&nbsp;Mohammad Hossein Moattar,&nbsp;Reza Sheibani","doi":"10.1016/j.cosrev.2025.100841","DOIUrl":"10.1016/j.cosrev.2025.100841","url":null,"abstract":"","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100841"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Graph diffusion models: A comprehensive survey of methods and applications","authors":"Yuntao Shou ,&nbsp;Wei Ai ,&nbsp;Tao Meng ,&nbsp;Keqin Li","doi":"10.1016/j.cosrev.2025.100854","DOIUrl":"10.1016/j.cosrev.2025.100854","url":null,"abstract":"<div><div>Diffusion models have rapidly emerged as a new paradigm in generative modeling. Therefore, we aim to provide a comprehensive review of graph diffusion models. We introduce various forms of diffusion models (i.e., DDPMs, SDEs, and SGMs), their working mechanisms, and how they can be extended to graph data. Specifically, graph diffusion models follow the modeling process of diffusion models, implement the diffusion process in graph data, and gradually denoise and generate new graph structures through reverse steps. The application of graph diffusion models is mainly focused on the application scenarios of generating molecules and proteins, but graph diffusion models also show potential in recommendation systems and other fields. We explore the performance and advantages of graph diffusion models in these specific applications, such as using them to discover new drugs and predict protein structures. Furthermore, we also discuss the problem of evaluating graph diffusion models and their existing challenges. Due to the complexity and diversity of graph data, the authenticity of generated samples is an important and challenging task. We analyze their limitations and propose potential improvement directions to better measure the effectiveness of graph diffusion models. The summary of existing methods mentioned is in our Github: <span><span>https://github.com/yuntaoshou/Graph-Diffusion-Models</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"59 ","pages":"Article 100854"},"PeriodicalIF":12.7,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145485543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}