IEEE transactions on artificial intelligence最新文献_第9页

A Multimodal Feature Distillation With Mamba-Transformer Network for Brain Tumor Segmentation With Incomplete Modalities 基于mamba -变压器网络的多模态特征提取用于不完全模态脑肿瘤分割

IEEE transactions on artificial intelligence Pub Date : 2026-09-01 Epub Date: 2026-03-30 DOI: 10.1109/TAI.2026.3678550

Ming Kang;Fung Fung Ting;Shier Nee Saw;Raphaël C.-W. Phan;Zongyuan Ge;Chee-Ming Ting

{"title":"A Multimodal Feature Distillation With Mamba-Transformer Network for Brain Tumor Segmentation With Incomplete Modalities","authors":"Ming Kang;Fung Fung Ting;Shier Nee Saw;Raphaël C.-W. Phan;Zongyuan Ge;Chee-Ming Ting","doi":"10.1109/TAI.2026.3678550","DOIUrl":"https://doi.org/10.1109/TAI.2026.3678550","url":null,"abstract":"Existing brain tumor segmentation methods usually utilize multiple magnetic resonance imaging (MRI) modalities in brain tumor images for segmentation, which can achieve better segmentation performance. However, in clinical applications, some modalities are often missing due to resource constraints, resulting in significant performance degradation for methods that rely on complete modality segmentation. In this article, we propose a multimodal feature distillation with Mamba-Transformer hybrid network (MMTSeg) for accurate brain tumor segmentation with missing modalities. We first employ a multimodal feature distillation (MFD) module to distill feature-level multimodal knowledge into different unimodalities to extract complete modality information. We further develop an unimodal feature enhancement (UFE) module to model the semantic relationship between global and local information. Finally, we built a cross-modal fusion (CMF) module to explicitly align the global correlations across modalities, even when some modalities are missing. Complementary features within and across modalities are refined by the Mamba-Transformer hybrid architectures in both the UFE and CMF modules, dynamically capturing long-range dependencies and global semantic information for complex spatial contexts. A boundary-wise loss function is employed as the segmentation loss of the proposed MMTSeg to minimize boundary discrepancies for a distance-based metric. Our ablation study demonstrates the importance of the proposed feature enhancement and fusion modules in the proposed network and the Transformer with Mamba block for improving the performance of brain tumor segmentation with missing modalities. Extensive experiments on the BraTS 2018 and BraTS 2020 datasets demonstrate that the proposed MMTSeg framework outperforms state-of-the-art methods when modalities are missing.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 9","pages":"5076-5090"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

No Time to Evade: Context-Aware Conditional Entropy Partitioning and Model Fusion for Robust Multiprocess Malware Detection 无时间逃避：上下文感知的条件熵划分和模型融合鲁棒多进程恶意软件检测

IEEE transactions on artificial intelligence Pub Date : 2026-09-01 Epub Date: 2026-03-30 DOI: 10.1109/TAI.2026.3678598

Lara Mauri;Ernesto Damiani

{"title":"No Time to Evade: Context-Aware Conditional Entropy Partitioning and Model Fusion for Robust Multiprocess Malware Detection","authors":"Lara Mauri;Ernesto Damiani","doi":"10.1109/TAI.2026.3678598","DOIUrl":"https://doi.org/10.1109/TAI.2026.3678598","url":null,"abstract":"Malware detection systems are increasingly challenged by polymorphic adversarial tactics that exploit variations in the division of tasks between threads and processes. Traditional detection models trained on monolithic, single-process traces struggle to generalize across such tactics. We propose a defense framework that integrates two complementary strategies: robustness against evasion through behavioral decomposition and architectural reinforcement through model-fusion-based learning. Our method employs an entropy-based partitioning strategy to decompose system call logs while explicitly preserving temporal and semantic dependencies via a violation-aware penalty mechanism. Our partitioning scheme reflects realistic polymorphic behaviors, ensuring that functionality is preserved across partitions. The framework is further strengthened through ensemble fusion, combining multiple submodels trained under heterogeneous adversarial split configurations. We evaluate the proposed approach on the Australian Defence Force Academy Linux Dataset (ADFA-LD) benchmark dataset, its multiprocess extension, ADFA-LD-Multi-Process (ADFA-LD-MP), and on the BarongTrace dataset of real malware traces. The experimentation shows that our framework substantially improves detection performance, particularly in high-split regimes where conventional models degrade. Partition-aware training, coupled with model fusion, yields improved F1-scores and enhanced generalization across diverse adversarial conditions.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 9","pages":"5117-5132"},"PeriodicalIF":0.0,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11457734","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148859747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IEEE Transactions on Artificial Intelligence Publication Information IEEE人工智能学报

IEEE transactions on artificial intelligence Pub Date : 2026-09-01 Epub Date: 2026-08-31 DOI: 10.1109/TAI.2026.3717078

引用次数: 0

Bayesian Probabilistic Knowledge From Diameter Prior for Decision Fusion to Detect Lung Nodule Heterogeneity 基于直径先验的贝叶斯概率知识决策融合检测肺结节异质性

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-02-18 DOI: 10.1109/TAI.2026.3665484

Md Rabiul Islam;Md Kamrul Hasan;Hasan Kurban;Erchin Serpedin

{"title":"Bayesian Probabilistic Knowledge From Diameter Prior for Decision Fusion to Detect Lung Nodule Heterogeneity","authors":"Md Rabiul Islam;Md Kamrul Hasan;Hasan Kurban;Erchin Serpedin","doi":"10.1109/TAI.2026.3665484","DOIUrl":"https://doi.org/10.1109/TAI.2026.3665484","url":null,"abstract":"Recent advances in deep learning (DL) have shown promising results in the identification of malignant lung nodules from computed tomography (CT) scans. However, conventional DL models primarily rely on spatial features and often lack clinical interpretability and the ability to incorporate domain-specific priors. To address these limitations, we integrate Bayesian diameter posterior probabilistic knowledge, allowing the model to leverage the well-established correlation between nodule size and malignancy likelihood. To enhance interpretability, we introduce a new deep-texture-shape (DTS) scoring scheme that offers oncologists a transparent, component-wise justification for malignancy prediction. Furthermore, relying solely on spatial features may overlook critical textural patterns, which play a significant role in malignancy assessment. To overcome this, we utilize local binary patterns (LBP), histogram of oriented gradients (HOG), and gray level cooccurrence matrix (GLCM) to extract rich textural features, capturing the fine-grained patterns that contribute to malignancy characterization. This multifaceted approach not only improves prediction accuracy but also enhances the model’s robustness. The proposed method, probabilistic knowledge-based decision fusion (ProKDF), is evaluated on two public datasets: 1) the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI); and 2) the International Society for Optics and Photonics and American Association of Physicists in Medicine (SPIE-AAPM). It achieves an F1-score of 87.48% (95% CI: 85.93%–89.03%) and an area under the receiver operating characteristic curve (AUC) of 92.56% (95% CI: 91.70%–93.42%) on LIDC-IDRI. Our findings suggest that integrating shape-based probabilistic knowledge and texture information enhances model performance and interpretability in detecting nodule heterogeneity.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 8","pages":"4518-4532"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Neural Network Intelligent Adaptive Optimal Output Feedback Attitude Tracking Control for Rigid Spacecraft Using Actor-Critic Reinforcement Learning 基于Actor-Critic强化学习的刚性航天器神经网络智能自适应最优输出反馈姿态跟踪控制

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-02-18 DOI: 10.1109/TAI.2026.3665496

Jinfan Zhang;Lin Zhao

引用次数: 0

Adaptive Reinforcement Learning-Driven Digital Twin for Multiobjective and Multimachine Process Optimization 多目标多机过程优化的自适应强化学习驱动数字孪生

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-02-09 DOI: 10.1109/TAI.2026.3655629

Chin-Yi Lin;Tzu-Liang Tseng;Tsung-Han Tsai

{"title":"Adaptive Reinforcement Learning-Driven Digital Twin for Multiobjective and Multimachine Process Optimization","authors":"Chin-Yi Lin;Tzu-Liang Tseng;Tsung-Han Tsai","doi":"10.1109/TAI.2026.3655629","DOIUrl":"https://doi.org/10.1109/TAI.2026.3655629","url":null,"abstract":"Achieving efficient and scalable multiobjective process optimization in the semiconductor industry poses significant challenges due to the complexity of multimachine and multiproduct scenarios, high-dimensional parameter spaces, and limited availability of real manufacturing data—especially in cutting-edge applications involving wide-bandgap (WBG) materials such as SiC and GaN. To address these issues, we introduce Adaptive Deep fEature-based Policy opTimization (ADEPT), an integrated digital twin (DT) framework that fuses multistage learning, reinforcement learning (RL), Gaussian processes (GPs), and dynamic deep feature extraction. Validated on real-world epitaxial (Epi) process data, ADEPT demonstrates robust performance under strict data constraints by leveraging GP-based uncertainty to guide RL exploration. Furthermore, by incorporating advanced meta-learning and transfer-learning (MLTL) strategies, ADEPT efficiently adapts to newly introduced machines and WBG-oriented product lines, expediting deployment and minimizing data requirements. We also provide rigorous mathematical convergence analysis, proving a high-probability guarantee of near-optimal solutions under appropriate assumptions, thus paving the way for future advancements in smart manufacturing.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 8","pages":"4291-4310"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626594","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Learning From Easy to Hard: Fingerprint Spoof Detection With Hard Sample Mining 从简单到困难的学习：指纹欺骗检测与硬样本挖掘

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-02-10 DOI: 10.1109/TAI.2026.3662649

Wenny Ramadha Putri;Farchan Raswa Hakim;Bach-Tung Pham;Shang-Kuan Chen;Chung-I Huang;Kuo-Chen Li;Shih-Lun Chen;Jia-Ching Wang

{"title":"Learning From Easy to Hard: Fingerprint Spoof Detection With Hard Sample Mining","authors":"Wenny Ramadha Putri;Farchan Raswa Hakim;Bach-Tung Pham;Shang-Kuan Chen;Chung-I Huang;Kuo-Chen Li;Shih-Lun Chen;Jia-Ching Wang","doi":"10.1109/TAI.2026.3662649","DOIUrl":"https://doi.org/10.1109/TAI.2026.3662649","url":null,"abstract":"To counter unauthorized access posed by presentation attacks in fingerprint authentication systems, deep learning-based fingerprint spoofing detection methods have been introduced. However, different artificial materials produce nonuniform spoof image representations, leading to a bias toward easy-to-recognize samples while ignoring hard-to-recognize ones. Hard sample mining is beneficial for accurately modeling a discriminative function. In this article, we propose a novel fingerprint spoof detection method with hard sample mining called EasyHard-FSD. Specifically, EasyHard-FSD introduces a self-training (teacher/student model), where a teacher model is learned using all original samples, and it is used to identify possible hard samples. EasyHard-FSD employs a hard mining module (HMM) to mask out hard samples for the student model. This strategy allows the student model to learn a better discriminative function for recognizing live and spoof fingerprint samples accurately. Experimental results on LiveDet2013 and LiveDet2015 demonstrate that EasyHard-FSD outperforms state-of-the-art (SOTA) methods in fingerprint spoof detection.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 8","pages":"4367-4375"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148626848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Prompt-Based Jailbreaking of Leading LLM Chatbots: A Survey of Attacks and Defenses 领先的LLM聊天机器人的快速越狱：攻击和防御的调查

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-02-17 DOI: 10.1109/TAI.2026.3665656

Brynn Knowlton;Jovani Campa;David Solis Gallo;Khalil Dajani;Nabeel Alzahrani

{"title":"Prompt-Based Jailbreaking of Leading LLM Chatbots: A Survey of Attacks and Defenses","authors":"Brynn Knowlton;Jovani Campa;David Solis Gallo;Khalil Dajani;Nabeel Alzahrani","doi":"10.1109/TAI.2026.3665656","DOIUrl":"https://doi.org/10.1109/TAI.2026.3665656","url":null,"abstract":"Generative artificial intelligence (AI) systems—particularly large language models (LLMs)—remain vulnerable to jailbreak attacks: adversarial prompts that bypass safeguards and elicit unsafe or restricted outputs. This survey synthesizes jailbreak research from 2023–2025, covering attack methods, defense strategies, and evaluation frameworks. Jailbreak techniques are grouped into five main categories: prompt-based injections, role-play conditioning, multiturn dialogue, multilingual or multimodal exploits, and optimization-driven pipelines. We also examine discovery methods such as human red teaming and automated attacker systems (e.g., AutoJailbreak and LLMStinger), and review defense mechanisms including supervised fine-tuning (SFT), reinforcement learning from human feedback (RLHF), adversarial fine-tuning, and output or pipeline filtering. Standardized benchmarks—such as PromptBench, JailbreakBench, and multimodal evaluation suites—are analyzed in terms of reproducibility, coverage, and robustness testing. Persistent issues of generalization, reproducibility, and scalability are discussed, along with trade-offs between safety alignment and model utility. This article provides an up-to-date reference clarifying the evolving landscape of jailbreak attacks and defenses in LLMs.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 8","pages":"4237-4251"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148627103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Emerging Paradigm of Quantum Machine Learning: Knowledge Insights and Future Prospects 量子机器学习的新兴范式：知识洞察和未来展望

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-03-03 DOI: 10.1109/TAI.2026.3669545

Keshav Singh Rawat;Tarun Sharma

{"title":"Emerging Paradigm of Quantum Machine Learning: Knowledge Insights and Future Prospects","authors":"Keshav Singh Rawat;Tarun Sharma","doi":"10.1109/TAI.2026.3669545","DOIUrl":"https://doi.org/10.1109/TAI.2026.3669545","url":null,"abstract":"The machines can handle tasks that require thinking and decisive capacity by using artificial intelligence. However, these systems require higher resources for their operations. As such, quantum machine learning (QML) has been explored as a parallel solution for such problems. In our study, we comprehensively analyzed the research evolution within the QML domain. We integrated the literature from Scopus and Web of Science to investigate underlying relationships through bibliometric information. The publication and citation trends, country collaboration, keyword cooccurrence, and thematic relationships helped to introspect the growth of QML. Our results show that quantum kernels, quantum natural language processing, contrastive learning, quantum reservoir computing, and explainable QML (XQML) are gaining attention within academia. It has been expanding across a diverse set of problem spaces such as management of resources, intrusion detection, process control, cloud computing, molecular and materials modeling, simulation of integrated circuits, and nuclear physics. The findings also highlight the existing challenges that are limiting the practical implementation of solutions within the noisy intermediate-scale quantum (NISQ) era. In response, we also identified actionable directives that can be followed for advancing the current state of research. These insights will provide future researchers with a single-point reference for guiding their exploration of QML.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 8","pages":"4252-4266"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148628180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

MSFFNet: Multistage Feature Fusion With Metadata Injection for Vehicle Reidentification 基于元数据注入的多阶段特征融合车辆再识别

IEEE transactions on artificial intelligence Pub Date : 2026-08-01 Epub Date: 2026-02-16 DOI: 10.1109/TAI.2026.3665179

Haotian Shi;Yuhui Deng;Jiande Huang;Qifen Yang;Peng Zhou;Ke Wang;Weiping Ding

{"title":"MSFFNet: Multistage Feature Fusion With Metadata Injection for Vehicle Reidentification","authors":"Haotian Shi;Yuhui Deng;Jiande Huang;Qifen Yang;Peng Zhou;Ke Wang;Weiping Ding","doi":"10.1109/TAI.2026.3665179","DOIUrl":"https://doi.org/10.1109/TAI.2026.3665179","url":null,"abstract":"Vehicle reidentification (Re-ID) is a crucial task in intelligent transportation systems and urban management, aiming to retrieve images of the same vehicle from different camera viewpoints. Prevailing solutions typically rely on computationally expensive modules or costly annotations, hindering their practical deployment. To overcome these limitations, we propose the novel multistage feature fusion network (MSFFNet), a lightweight and robust feature alignment framework. First, we design the multistage fusion head (MSFH) that generates robust feature representations by fusing multistage features based on the inherent hierarchical feature pyramid of CNNs. The resulting fusion features capture both the abstract semantics of high-level features and the fine-grained local details of low-level features. To alleviate the ineffective fusion caused by the varying distance distributions between features at different stages, we introduce the graph convolution-based fusion block (GCFB) to study the relationships between these features. Second, to mitigate the influence of irrelevant information during fusion, we introduce the metadata auxiliary module (MAM) that injects easily available metadata, such as camera IDs and viewpoints, into fused features. This allows the model to focus on discriminative vehicle information. Our model produces compact feature representations, occupies less storage, and achieves faster retrieval speeds, with a computing time of 1.74 ms per image. Extensive experiments demonstrate that our method achieves state-of-the-art performance on three widely used vehicle ReID benchmarks, establishing an optimal balance between speed and accuracy.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 8","pages":"4491-4503"},"PeriodicalIF":0.0,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148628242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0