计算机科学最新文献

{"title":"Adaptive orthogonal role-aware transformer for robust instruction alignment","authors":"Yeo-Chan Yoon,&nbsp;Sookyun Kim,&nbsp;Gyeongmin Kim","doi":"10.4218/etrij.2025-0449","DOIUrl":"https://doi.org/10.4218/etrij.2025-0449","url":null,"abstract":"<p>Large language models are vulnerable to prompt injection because transformer attention treats tokens too uniformly across system and user roles. We propose the <i>adaptive orthogonal role-aware transformer</i> (AORT), a lightweight retrofit that improves role-sensitive decoding. AORT preserves the base token embeddings and adds at each transformer layer a learned orthonormal re-encoding and a role-gated additive attention bias to maintain separation between instruction and user representations during context mixing. We evaluate AORT on direct and indirect prompt-injection benchmarks, including the <i>BIPIA-text</i> benchmark. AORT achieves the lowest attack success rate (ASR) in most settings and successfully resists all eight evaluated attacks on LLaMA 3.1–8B. AORT reduces the ASR from 13.6% to 2.1% on LLaMA 3.1–8B evaluated with the BIPIA-text benchmark. These findings indicate that explicit role-priority structure in the forward pass can substantially improve prompt-injection robustness with minimal architectural overhead.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"924-935"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2025-0449","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Explainable AI for clove quality grading: Benchmarking post hoc XAI and compositional interpretability under domain shift","authors":"Innocent Nyalala,&nbsp;Patrick Vincent Ndowo","doi":"10.4218/etrij.2026-0225","DOIUrl":"https://doi.org/10.4218/etrij.2026-0225","url":null,"abstract":"<p>Reliable agricultural commodity grading requires accuracy and explanations that match regulatory criteria. Using 4603 expert-graded clove images across four Zanzibar State Trading Corporation (ZSTC) quality grades, we benchmark seven post hoc explainable artificial intelligence (XAI) methods (Grad-CAM, Grad-CAM++, ScoreCAM, LIME, GradientSHAP, CLS-Attention, and Chefer LRP) across eight CNN/ViT architectures and compare them with a compositional segmentation–classification pipeline that builds interpretability into the model. We propose the Explanation Energy Ratio (EER) to quantify foreground-aligned explanations using ground-truth masks. EER varies widely by architecture explainer, with different explainers yielding different results for the same classifier. Under a synthetic background-replacement domain shift, robustness differs despite similar in-domain performance. LIME performs well on transformers but is slow; SHAP is too costly for edge use. The compositional pipeline achieves 99.45% F1 with 100% EER by design and is preferred by certified ZSTC graders for regulatory auditing.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"911-923"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2026-0225","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222802","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Management of IT Infrastructure Library Processes Based on Cloud Computing: A Comprehensive Study, Current Applications, and Future Research Directions","authors":"Xi Zhao","doi":"10.1002/nem.70056","DOIUrl":"https://doi.org/10.1002/nem.70056","url":null,"abstract":"<div>\u0000 \u0000 <p>This paper addresses the critical intersection of IT Infrastructure Library (ITIL) processes and cloud computing adoption, a relationship requiring structured examination despite ITIL's established role in IT service management. Organizations migrating to the cloud face complexities in risk reduction and seamless transfer, making the synergy between ITIL and cloud strategies essential. Although ITIL is recognized for enhancing business operations, measuring performance, and reducing costs, its specific influence on cloud-centric ITIL service strategy execution needs further elucidation. Employing a systematic literature review (SLR) methodology, this study analyzes recent research to examine how cloud computing influences ITIL service strategy procedures and their sub-processes. The findings aim to provide a clear understanding of this evolving landscape for IT professionals and researchers. The systematic review reveals that integrating ITIL with cloud computing not only drives substantial cost savings and enhances user experience but also necessitates adapting ITIL processes to the dynamic nature of cloud environments, particularly in areas of security and configuration management.</p>\u0000 </div>","PeriodicalId":14154,"journal":{"name":"International Journal of Network Management","volume":"36 6","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"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":"Development of two-phase framework for driver behavior classification and representative motion prototyping for a reliable simulation-based driver monitoring system","authors":"Kezia Amanda Kurniadi,&nbsp;Woojin Kim,&nbsp;Daesub Yoon","doi":"10.4218/etrij.2026-0207","DOIUrl":"https://doi.org/10.4218/etrij.2026-0207","url":null,"abstract":"<p>Driver monitoring systems (DMS) are critical for safe autonomous driving, which requires accurate behavior recognition and reliable simulation environments for validation. This study proposes a robust two-phase framework that combines a classification model and a prototype selection procedure to automatically generate realistic motion baselines from the acquired 3D skeleton data. Phase 1 uses a spatial temporal graph convolutional network to classify driver behaviors, and Phase 2 leverages class-wise mean trajectories to interpret motion prototypes and select the closest real sequence, creating a verifiable motion trajectory library for simulation-based safety testing. Our approach addresses DMS challenges through prototype visualization, robustness via discriminative embeddings, and standardized simulation scenarios, making it suitable for human–AI interactions in automotive safety systems. This framework produces a motion library of a real recorded sequence that can be directly used for simulation scenario generation and demonstrated through Metahuman retargeting and visualization in Unreal Engine.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"884-895"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2026-0207","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnosing judgment flip risk of large language models in fact verification under conflicting evidence","authors":"Jihyeon Roh,&nbsp;Kyoungman Bae","doi":"10.4218/etrij.2026-0242","DOIUrl":"https://doi.org/10.4218/etrij.2026-0242","url":null,"abstract":"<p>Large language model (LLM)-based fact verifiers can encounter conflicting evidence in retrieval-augmented pipelines, but their robustness to evidence composition remains underexamined. We introduce a flip-rate diagnostic protocol that measures how often initially correct binary judgments become incorrect when the supporting and refuting evidence is perturbed. Using six models and two datasets (FEVER and SciFact), we vary the evidence ratio, quantity, and ordering and evaluate chain-of-thought (CoT) prompting and three lightweight mitigations. On FEVER, one refuting passage alongside three supporting passages flips correct judgments in 3.3%–22.2% of cases; under full conflict, flip rates exceed 60% for every model on FEVER and 90% on SciFact. For most models, placing refuting evidence later also increases flip rates. CoT reduces flips under full conflict but is unreliable under mixed conflict. Conflict-aware stance-grouped prompting is the most consistent mitigation across datasets. These findings indicate that baseline accuracy alone is insufficient for assessing LLM fact-verifier reliability and that evidence composition should be treated as a separate axis of robustness evaluation.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"838-851"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2026-0242","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lost in the comments: Social context as a single-pass jailbreak and defense on agentic platforms","authors":"Taehyeon Yun,&nbsp;Seonhee Jeon,&nbsp;Geonwoo Kim,&nbsp;Jeongsu Park,&nbsp;Moohong Min","doi":"10.4218/etrij.2026-0190","DOIUrl":"https://doi.org/10.4218/etrij.2026-0190","url":null,"abstract":"<p>As LLM-based agents move onto multiagent platforms, they process direct prompts and surrounding social context—posts, comments, and upvotes—an attack surface that single-turn alignment never addressed. We built a simulation of Moltbook, a social network for AI agents, and a pipeline turning 100 JailBreakBench goals into platform-native posts with bystander comments of three valences: aggressive, ethical, and measured. Platform-context reformatting alone raises the attack success rate (ASR) of GPT-4o-mini from 7% to 71%, beating a twenty-query optimization baseline in one pass. Valence then acts as a bidirectional safety modulator: Ethical comments sharply suppress ASR, aggressive endorsements paradoxically reduce it by exposing intent, and measured, intellectually toned comments prove most dangerous, sustaining or elevating ASR. An ablation identifies valence rather than volume as key—a 35-fold rise in upvotes leaves ASR unchanged—and every finding reproduces under an independent safety classifier with human adjudication. Ambiguous social context thus poses a greater threat than explicit pressure, while safety-valenced signals offer a deployable defense for agentic platforms.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"796-806"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2026-0190","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data poisoning in LLM multiagent societies: Social proof drives collective decision failure in financial deliberation","authors":"Jeongsu Park,&nbsp;Yuji Lim,&nbsp;Geonwoo Kim,&nbsp;Taehyeon Yun,&nbsp;Moohong Min","doi":"10.4218/etrij.2026-0186","DOIUrl":"https://doi.org/10.4218/etrij.2026-0186","url":null,"abstract":"<p>Large language model (LLM) multiagent systems are increasingly deployed for high-stakes financial deliberation, but individually secure LLMs may become vulnerable when embedded in a peer-to-peer agent society. We study data poisoning in such societies using Moltbook, a deliberation test bed modeled after the agent-only social platform. A single malicious agent representing just 14% of a seven-agent society drives collective accuracy from 87.5% to 52.5%—not through prompt injection or technical exploits, but through social proof: After building peer trust, it injects fabricated data, inducing either cognitive infection, where defenders adopt the incorrect decision, or paralysis, where they abandon their judgment. Ablations across topologies and model families show that the collapse is structural rather than model-specific. Our system-level defense publicly tags claims untraceable to the official record, severing unwarranted trust transfer and restoring accuracy to 85.8%. Securing LLM multiagent systems requires system-level verification, not individual prompt alignment.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"783-795"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2026-0186","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimized multi-scale dual-modality feature pyramid network for early eye disease classification using fundus and OCT imaging","authors":"R. Kiruthiga, A. Saran Kumar","doi":"10.1007/s40747-026-02507-w","DOIUrl":"https://doi.org/10.1007/s40747-026-02507-w","url":null,"abstract":"","PeriodicalId":10524,"journal":{"name":"Complex & Intelligent Systems","volume":"33 1","pages":""},"PeriodicalIF":5.8,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149221440","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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":"SF-CAMNet: A spatial–frequency fusion network for robust spatial-domain image steganalysis","authors":"Mehnaz Batool,&nbsp;Adil Bashir","doi":"10.4218/etrij.2026-0036","DOIUrl":"https://doi.org/10.4218/etrij.2026-0036","url":null,"abstract":"<p>Spatial domain image steganalysis has become increasingly important due to the widespread use of digital images for communication and data sharing over the internet and social media. While steganography enables secure communication and digital watermarking, it can also be exploited for malicious purposes, making reliable detection essential. This study proposes SF-CAMNet, a deep learning framework for spatial-domain image steganalysis that integrates spatial and frequency-domain features. The model combines Spatial Rich Model (SRM) filters, the Convolutional Block Attention Module (CBAM), Mish activation, and a Discrete Wavelet Transform (DWT) branch to enhance the detection of subtle embedding artifacts. Performance was evaluated on the BOSSBase v1.01 dataset using S-UNIWARD, WOW, HILL, HUGO, and MiPOD at payloads of 0.2 and 0.4 bpp. Cross-dataset evaluation on the BOWS2 dataset further demonstrates the robustness and generalization capability of the proposed model. Experimental results show that SF-CAMNet achieves competitive performance compared with state-of-the-art spatial-domain image steganalysis methods.</p>","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"820-837"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etrij.2026-0036","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Special issue on trustworthy and safe AI: Interpretability, robustness, and security","authors":"Sungwon Yi,&nbsp;Dae Hun Nyang,&nbsp;Jaehyung Kim,&nbsp;Jaegul Choo,&nbsp;Daeseon Choi,&nbsp;Soonmin Hwang","doi":"10.4218/etr2.70162","DOIUrl":"https://doi.org/10.4218/etr2.70162","url":null,"abstract":"&lt;p&gt;Recent advances in artificial intelligence (AI), particularly in large language models (LLMs) and physical robotics, have rapidly reshaped daily life and industrial applications. As these systems become more capable and autonomous, ensuring their safety, trustworthiness, robustness, and alignment with human values has become increasingly important. In response to these challenges, this special issue on “Trustworthy and Safe AI: Interpretability, Robustness, and Security” brings together recent research that advances the safety and trustworthiness of AI across diverse models, systems, and real-world applications.&lt;/p&gt;&lt;p&gt;This special issue received 50 submissions, reflecting the growing research interest in trustworthy and safe AI. Following a rigorous peer-review process, 11 papers were selected for publication. Together, these studies span several interconnected dimensions of the field, ranging from emerging security threats in LLMs and multi-agent systems to robustness, autonomous system safety, and explainability for practical deployment.&lt;/p&gt;&lt;p&gt;The following five papers focus on emerging security and safety threats across modern AI systems. These studies examine how vulnerabilities arise from adversarial attacks, multi-agent interactions, social context, multilingual settings, and hidden information while exploring the corresponding mechanisms for detection and defense.&lt;/p&gt;&lt;p&gt;In [&lt;span&gt;1&lt;/span&gt;], the authors investigate a more evasive mixed clean/dirty-label X-to-X backdoor attack and propose an ensemble defense that combines complementary detectors such as Neural Cleanse, PT-RED, CEPA, and MMBD. This work is significant because it addresses both robustness and interpretability, and the ensemble improves resilience across diverse backdoor mechanisms while also supporting trigger inversion to help explain the nature of the detected attack. By jointly considering stronger adversarial attacks, generalized defense, and interpretable backdoor analysis, this study aligns well with the themes of Trustworthy and Safe AI: Interpretability, Robustness, and Security.&lt;/p&gt;&lt;p&gt;In [&lt;span&gt;2&lt;/span&gt;], the authors study how a single malicious agent can compromise the collective decision-making of otherwise well-aligned LLM agents in a collaborative financial multi-agent environment and proposes a data-lineage-based systemic defense. Its main contribution is to show that the interaction dynamics themselves can become a security vulnerability, extending AI safety concerns beyond isolated models to multi-agent societies. By analyzing poisoning, social conformity, and system-level verification across multiple models and network settings, this study offers timely insights into the robustness, safety, and trustworthy coordination of emerging autonomous multi-agent AI systems.&lt;/p&gt;&lt;p&gt;The paper [&lt;span&gt;3&lt;/span&gt;] titled “Lost in the comments: Social context as a single-pass jailbreak and defense on agentic platforms” by Jeongsu Park and colleagues investigates how comments","PeriodicalId":11901,"journal":{"name":"ETRI Journal","volume":"48 5","pages":"763-767"},"PeriodicalIF":2.0,"publicationDate":"2026-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.4218/etr2.70162","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149222801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}