IEEE Transactions on Services Computing最新文献_第4页

AirLoc: A QoS-Aware Mobile Location Service Framework via UAV-Assisted LoRa Networks AirLoc：基于无人机辅助LoRa网络的qos感知移动定位服务框架

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-08-20 DOI: 10.1109/TSC.2026.3725677

Ciyuan Chen;Zhuqing Xu;Feng Shan;Runqun Xiong;Weizheng Wang;Zhimeng Yin;Ye Wei;Junzhou Luo;Xiaohua Jia

{"title":"AirLoc: A QoS-Aware Mobile Location Service Framework via UAV-Assisted LoRa Networks","authors":"Ciyuan Chen;Zhuqing Xu;Feng Shan;Runqun Xiong;Weizheng Wang;Zhimeng Yin;Ye Wei;Junzhou Luo;Xiaohua Jia","doi":"10.1109/TSC.2026.3725677","DOIUrl":"10.1109/TSC.2026.3725677","url":null,"abstract":"As the demand for Location-Based Services (LBS) grows in remote or resource-limited regions, providing high Quality of Service (QoS) location for wide mobile applications has become increasingly crucial. Existing approaches typically rely on fixed gateways and static nodes, which is unsuitable or expensive in demanding environments such as wild fields or temporary sites. In this paper, we propose AirLoc, a QoS-aware framework that ensures long-range, accurate, and low-cost location services for wide mobile scenarios. Considering that LoRa (Long Range) is suitable for localization in wide areas due to its long-range and low-power nature. AirLoc uses only one single legacy Uncrewed Aerial Vehicle (UAV)-assisted LoRa gateway to track mobile LoRa targets without relying on dense infrastructure. AirLoc is mainly based on the key observation that LoRa’s RSSI is insensitive at long ranges and unstable at short ranges. To address these issues, we propose three key service modules for AirLoc framework. We first propose a LoRa spreading factor-based long-range localization module to determine the coarse region of the targets. Additionally, we design an RSSI reconstruction–based module to achieve accurate tracking in short ranges. Finally, a QoS-driven interaction protocol module is used to operate the architecture with minimal overhead. We implement AirLoc on both a real-world testbed that includes a UAV-assisted LoRa gateway and a large-scale simulation platform. Extensive real-world experiments show that AirLoc accurately tracks mobile targets in a <inline-formula><tex-math>$550 {rm m} times 400 {rm m}$</tex-math></inline-formula> area at low power, making it a QoS-aware, modular service architecture that easily integrates into existing LBS systems.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3347-3360"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148768153","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}

引用次数: 0

RefactorCoderQA: Benchmarking LLMs for Multi-Domain Coding Question Solutions in Cloud and Edge Deployment RefactorCoderQA：在云和边缘部署中对多域编码问题解决方案进行基准测试的llm

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-07-27 DOI: 10.1109/TSC.2026.3717248

Shadikur Rahman;Aroosa Hameed;Gautam Srivastava;Syed Muhammad Danish

{"title":"RefactorCoderQA: Benchmarking LLMs for Multi-Domain Coding Question Solutions in Cloud and Edge Deployment","authors":"Shadikur Rahman;Aroosa Hameed;Gautam Srivastava;Syed Muhammad Danish","doi":"10.1109/TSC.2026.3717248","DOIUrl":"10.1109/TSC.2026.3717248","url":null,"abstract":"To optimize the reasoning and problem-solving capabilities of Large Language Models (LLMs), we propose a structured multi-agent prompting framework inspired by a cloud–edge collaborative architecture. The framework consists of three specialized agents: GuideLLM, which provides methodological guidance; SolverLLM, which generates code solutions; and JudgeLLM, which automatically evaluates solution correctness and quality. To evaluate and demonstrate the effectiveness of this architecture in realistic settings, we introduce RefactorCoderQA, a comprehensive benchmark designed to evaluate and enhance the performance of LLMs across multi-domain coding tasks. Motivated by the limitations of existing benchmarks, RefactorCoderQA systematically covers various technical domains, including Software Engineering, Data Science, Machine Learning, and Natural Language Processing, using authentic coding challenges from Stack Overflow. We propose <italic>RefactorCoder-MoE, a fine-tuned mixture-of-experts (MoE) code language model based on DeepSeek-Coder-7B-Instruct, adapted to the RefactorCoderQA benchmark using QLoRA for domain-specific coding question answering. Extensive experiments reveal that RefactorCoder-MoE achieves strong and competitive performance, significantly outperforming all evaluated open-source and commercial baselines, with an overall accuracy of 76.84%.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3594-3606"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148595595","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}

引用次数: 0

Multi-Dimensional Autoscaling of Stream Processing Services on Edge Devices 边缘设备上流处理服务的多维自缩放

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-07-08 DOI: 10.1109/TSC.2026.3711142

Boris Sedlak;Philipp Raith;Andrea Morichetta;Víctor Casamayor Pujol;Schahram Dustdar

{"title":"Multi-Dimensional Autoscaling of Stream Processing Services on Edge Devices","authors":"Boris Sedlak;Philipp Raith;Andrea Morichetta;Víctor Casamayor Pujol;Schahram Dustdar","doi":"10.1109/TSC.2026.3711142","DOIUrl":"10.1109/TSC.2026.3711142","url":null,"abstract":"Edge devices have limited resources, which inevitably leads to situations where local resources do not suffice to serve the changing request load. While existing autoscaling mechanisms focus entirely on resource scaling, Edge devices require alternative ways to sustain Service Level Objectives (SLOs) of deployed services. To address these issues, we introduce a Multi-dimensional Autoscaling Platform (MUDAP) that supports fine-grained vertical scaling across service- and resource-level dimensions. MUDAP supports service-specific scaling tailored to available parameters, e.g., scale data quality or model size for a particular service. To optimize service execution on Edge devices, where multiple services compete for one global resource pool, we present a scaling agent using Regression Analysis of Structural Knowledge (RASK). The RASK agent efficiently learns a continuous regression model of the processing environment for optimizing the global SLO fulfillment across the deployed services. We compared our approach with two autoscalers—the Kubernetes VPA and a reinforcement learning agent—for scaling up to 9 services on a single Edge device. Our results showed that RASK can infer an accurate regression model in merely 20 iterations (i.e., observe 200 s of processing). By increasingly adding elasticity dimensions, RASK sustained the highest request load with 28% fewer SLO violations, compared to baselines.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3510-3523"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11599677","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148414723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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}

引用次数: 0

ObliviGate: Toward Architecture-Oblivious Privacy-Preserving Inference for Malicious Security 遗忘：面向架构无关的恶意安全隐私保护推理

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-04-22 DOI: 10.1109/TSC.2026.3686705

Nianyun Song;Fuyi Wang;Rongfang Bie;Yu Guo;Leo Yu Zhang;Xiaohua Jia

{"title":"ObliviGate: Toward Architecture-Oblivious Privacy-Preserving Inference for Malicious Security","authors":"Nianyun Song;Fuyi Wang;Rongfang Bie;Yu Guo;Leo Yu Zhang;Xiaohua Jia","doi":"10.1109/TSC.2026.3686705","DOIUrl":"10.1109/TSC.2026.3686705","url":null,"abstract":"Neural network inference is increasingly outsourced to cloud platforms, offering scalable computation with minimal local overhead. To mitigate privacy risks, secure multi-party computation (MPC) is commonly used to protect model parameters and user inputs. However, most MPC-based approaches assume that model architectures are publicly known, threatening intellectual property and facilitating various privacy attacks. In this work, we present ObliviGate, a novel scheme for architecture-oblivious, privacy-preserving inference built on a fully outsourced two-party computation (2 PC) setting. ObliviGate introduces a principled two-level obfuscation mechanism: intra-layer obfuscation injects dummy neurons to conceal layer configurations, while inter-layer obfuscation inserts dummy edges with a secure gating mechanism to hide network topology. We further leverage differential privacy (DP) to guide dummy operation allocation, providing formal privacy guarantees while maintaining model functionality with negligible accuracy degradation. To ensure robustness against malicious cloud providers, we introduce an optimized malicious-secure inference protocol based on function secret sharing (FSS) with preprocessing, achieving strong verifiability with low online latency. Extensive theoretical analysis and experiments across multiple neural network architectures show that ObliviGate delivers <inline-formula><tex-math>$1.1times sim 14.8times$</tex-math></inline-formula> speedups over prior studies, making it a practical, scalable solution for real-world outsourced inference.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3156-3169"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147735582","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}

引用次数: 0

PMLSF: A Privacy-Preserving Machine Learning Service Framework for Efficient Task Allocation in Mobile Crowdsensing PMLSF：一种保护隐私的机器学习服务框架，用于移动众测中高效的任务分配

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-04-16 DOI: 10.1109/TSC.2026.3684881

Zeyuan Li;Xuechi Chen;Yuxin Liu;Neal N. Xiong;Houbing Herbert Song;Tian Wang

{"title":"PMLSF: A Privacy-Preserving Machine Learning Service Framework for Efficient Task Allocation in Mobile Crowdsensing","authors":"Zeyuan Li;Xuechi Chen;Yuxin Liu;Neal N. Xiong;Houbing Herbert Song;Tian Wang","doi":"10.1109/TSC.2026.3684881","DOIUrl":"10.1109/TSC.2026.3684881","url":null,"abstract":"The proliferation of machine learning services in mobile crowdsensing (MCS) brings both new opportunities and severe privacy challenges. Location leakage, identity linkage, and centralized trust bottlenecks remain critical risks, while conventional differential privacy (DP) techniques often degrade service responsiveness due to noise generation latency and increased training overhead. To address these issues, this paper proposes PMLSF, a Privacy-Preserving Machine Learning Service Framework that integrates differential privacy, deep reinforcement learning (DRL), and blockchain-based orchestration to enable secure, efficient, and scalable task allocation in MCS environments. Specifically, we design Optimized Geo-indistinguishability (OptGI), a lightweight DP mechanism that reduces noise generation latency while maintaining rigorous privacy guarantees. We further introduce TopK Double Deep Q-Network (TopK-DDQN), a task allocation service that prunes infeasible actions and supports latency-aware decision-making over privatized inputs. To eliminate single points of failure, PMLSF employs a permissioned blockchain orchestration layer, ensuring anonymity, accountability, and distributed trust without imposing prohibitive overhead. Extensive experiments demonstrate that PMLSF achieves low-latency privacy protection and significantly improves allocation utility and profit compared with state-of-the-art methods, positioning it as a scalable foundation for integrating DP methodologies into machine learning-driven MCS services.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3226-3240"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147695758","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}

引用次数: 0

ORACL: Optimized Reasoning for Autoscaling via Chain of Thought With LLMs for Microservices ORACL：通过思维链和llm为微服务优化自动伸缩推理

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-06-29 DOI: 10.1109/TSC.2026.3707982

Haoyu Bai;Muhammed Tawfiqul Islam;Minxian Xu;Rajkumar Buyya

{"title":"ORACL: Optimized Reasoning for Autoscaling via Chain of Thought With LLMs for Microservices","authors":"Haoyu Bai;Muhammed Tawfiqul Islam;Minxian Xu;Rajkumar Buyya","doi":"10.1109/TSC.2026.3707982","DOIUrl":"10.1109/TSC.2026.3707982","url":null,"abstract":"Modern microservice and serverless applications require autoscaling to balance resource utilization and Quality of Service (QoS). Since effective scaling decisions often depend on root-cause analysis (RCA), controllers must reason over a large and evolving decision space. Existing approaches either require costly per-deployment training or depend on brittle hand-crafted rules. We propose ORACL, <underline>Optimized <underline>Reasoning for <underline>Autoscaling via <underline>Chain of Thought with <underline>LLMs for Microservices, an LLM-guided framework for joint RCA and resource allocation. ORACL converts runtime telemetry into natural-language state sketches and uses an instruction-tuned, reinforcement-learning-regularized LLM to identify root causes, prune candidate actions, and recommend safe resource adjustments. We curate, profile, and annotate over 80 GB of traces and metrics drawn from open-source microservice workloads to train this model. Experiments show that ORACL improves RCA accuracy by 15% over state-of-the-art baselines, accelerates training by up to 24×, and improves QoS by 6% in short-term scenarios. These results demonstrate that LLM-guided reasoning provides generalizable, interpretable, and actionable resource management for cloud-native systems.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3540-3553"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148371842","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}

引用次数: 0

MicroInspect: A Progressive Framework for Metric-Level Root Cause Localization in Microservices MicroInspect：一个用于微服务中度量级根本原因定位的渐进框架

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-07-15 DOI: 10.1109/TSC.2026.3713967

Yanshang Yin;Tiantian Zhu;Tieming Chen;Mingqi Lv;Qijie Song;Baiyang Ji

{"title":"MicroInspect: A Progressive Framework for Metric-Level Root Cause Localization in Microservices","authors":"Yanshang Yin;Tiantian Zhu;Tieming Chen;Mingqi Lv;Qijie Song;Baiyang Ji","doi":"10.1109/TSC.2026.3713967","DOIUrl":"10.1109/TSC.2026.3713967","url":null,"abstract":"The inherent complexity of microservices makes them prone to cascading failures, yet locating the precise root cause remains a critical challenge. Existing multimodal methods often fuse logs, metrics, and traces through coarse feature concatenation, which can introduce semantic interference. Moreover, although some recent studies have advanced from service-level localization to metric-level localization, they usually rely on static correlation or causal models and thus struggle to capture dynamic fault propagation. We present MicroInspect, a progressive framework for metric-level root cause localization in microservices. MicroInspect first disentangles multimodal features to reduce semantic interference and dynamically fuses them with a Mixture-of-Experts gate adapted to fault propagation stages, supporting accurate anomaly detection and service-level localization. It then employs a Dynamic Neural Granger Causality Network (DNGCN), guided by anomaly detection signals, to infer time-varying metric dependencies and progressively refine root causes from faulty services to metric-level causes. Evaluation on four public benchmarks (TrainTicket, SocialNetwork, FAMOS-Mall, and GAIA) shows that MicroInspect achieves the best overall performance across anomaly detection, service-level localization, and metric-level root cause localization, with statistically significant improvements in F1-score and HR@1 under paired <inline-formula><tex-math>$t$</tex-math></inline-formula>-tests (<inline-formula><tex-math>$p< 0.05$</tex-math></inline-formula>).","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3496-3509"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148460814","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}

引用次数: 0

PriPTune: Privacy-Preserving LLM Fine-Tuning Service Framework via Blockchain and TEE-GPU Collaborative Computing PriPTune：基于区块链和TEE-GPU协同计算的隐私保护LLM微调服务框架

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-08-14 DOI: 10.1109/TSC.2026.3723953

Lexuan Lin;Ming Li;Zhihuan Yang;Jiasi Weng;Rui Wu;Jian Weng

{"title":"PriPTune: Privacy-Preserving LLM Fine-Tuning Service Framework via Blockchain and TEE-GPU Collaborative Computing","authors":"Lexuan Lin;Ming Li;Zhihuan Yang;Jiasi Weng;Rui Wu;Jian Weng","doi":"10.1109/TSC.2026.3723953","DOIUrl":"10.1109/TSC.2026.3723953","url":null,"abstract":"Fine-tuning large language models (LLMs) on private datasets is an effective approach for building domain-specific services, but outsourcing training may expose sensitive data, intermediate activations, gradients, and trainable parameters to untrusted infrastructure. Blockchain can provide transparent governance, while Trusted Execution Environments (TEEs) offer attestable execution; however, TEE CPU throughput alone cannot sustain LLM-scale matrix computation, and direct GPU offloading exposes sensitive tensors. We present PriPTune, a privacy-preserving LoRA fine-tuning framework that combines on-chain governance with TEE-GPU collaborative computing under explicit TEE and side-channel assumptions. Smart contracts manage initialization, authorization, evidence anchoring, and model release, while off-chain TEEs enforce confidential execution. PriPTune keeps raw activations, gradients, LoRA adapters, optimizer states, and mixing secrets inside the TEE, and offloads only frozen-weight GEMMs over block-mixed tensors to an untrusted GPU. A TEE-held permutation and block-mixing transform hide raw token-level activations and avoid the same-order TEE-side correction required by additive masking. PriPTune further supports probabilistic GPU-result auditing to detect manipulation with configurable overhead. Evaluations show that PriPTune recovers ROUGE-L and BERTScore within 1.6% of plaintext LoRA, reduces protected decoder-block latency from a CPU-only upper bound of 1091.8 ms to 30.4 ms, and provides practical blockchain confirmation and throughput for lifecycle governance.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3568-3580"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148728105","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}

引用次数: 0

Structure-Aware Serverless Orchestration via Graph Reinforcement Learning: Joint Optimization of Function Fusion and Placement 基于图强化学习的结构感知无服务器编排：功能融合和布局的联合优化

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-08-18 DOI: 10.1109/TSC.2026.3724938

Donghong Xu;Zhaoyan Ding

{"title":"Structure-Aware Serverless Orchestration via Graph Reinforcement Learning: Joint Optimization of Function Fusion and Placement","authors":"Donghong Xu;Zhaoyan Ding","doi":"10.1109/TSC.2026.3724938","DOIUrl":"10.1109/TSC.2026.3724938","url":null,"abstract":"Function-as-a-Service (FaaS) platforms are increasingly used for data-intensive applications that execute as Directed Acyclic Graphs (DAGs). In these workflows, function fusion and placement are coupled: fusing dependent functions can reduce data transfer and cold-start overhead, but it also changes the memory footprint and the set of feasible deployment nodes. We propose <bold>GRL-Orch, a graph-reinforcement-learning orchestration framework for this coupled control problem. GRL-Orch encodes workflow topology with a Graph Neural Network (GNN), uses a factorized joint policy for fusion and placement, and applies feasibility-aware repair and placement masks before residual Out-Of-Memory (OOM) penalties are considered. Trace-derived simulator results using Azure Functions 2019 profiles show that the learned GRL-Orch policy reaches a mean objective of 1.147 on 100 held-out workflows, below the strongest Critical-Path-Fusion baseline at 1.184. An extended 500-workflow audit preserves the advantage, with objectives 1.368 versus 1.449. A live Apache OpenWhisk replay on small DAGs further shows consistent fusion-policy ranking in a 15-setting core matrix, while a 24-setting stress extension preserves 20/24 best-policy agreement and exposes deviations on size-20 diamond workflows. Together, these results characterize GRL-Orch as a structure-aware control design for serverless workflow orchestration, combining trace-derived scale with OpenWhisk grounding for key fusion-policy trends.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3648-3660"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148755524","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}

引用次数: 0

Service-Centric Adolescent Mental Health Early Warning Solutions and Enabling Technologies: Survey and Framework 以服务为中心的青少年心理健康早期预警解决方案和使能技术：调查和框架

IF 6.2 2区计算机科学

IEEE Transactions on Services Computing Pub Date : 2026-09-01 Epub Date: 2026-07-28 DOI: 10.1109/TSC.2026.3717586

Jiawei Dang;Jiali Zhu;Junyang Chen;Haodi Zhang;Liang-Jie Zhang

{"title":"Service-Centric Adolescent Mental Health Early Warning Solutions and Enabling Technologies: Survey and Framework","authors":"Jiawei Dang;Jiali Zhu;Junyang Chen;Haodi Zhang;Liang-Jie Zhang","doi":"10.1109/TSC.2026.3717586","DOIUrl":"10.1109/TSC.2026.3717586","url":null,"abstract":"The increasing global prevalence of adolescent mental health challenges demands proactive early warning systems. Although such systems have attracted growing research interest, most prior work emphasizes algorithmic performance, while the surrounding service infrastructure remains underexplored. To address key limitations in current practice, including data silos, insufficient multimodal data integration, and inadequate privacy protection during continuous model evolution, this paper presents a service-oriented, end-to-end framework for adolescent mental health early warning, in which data governance, modeling, and feedback are organized as interoperable microservices. The framework adopts a lakehouse architecture as a governed data layer that unifies multimodal schemas, data lineage, access control, and reusable analytics to facilitate cross-institution collaboration. Built upon this layer, we develop multimodal feature representations that integrate textual, visual, auditory, and physiological signals, together with a modular early warning service engine. To support continual refinement without exposing sensitive data, we further propose Dynamic Federated Co-Evolution (DFCE), a paradigm that couples cross-institutional federated learning with an expert-in-the-loop feedback mechanism, allowing models to co-evolve with expert feedback across institutions. As a proof of concept for the DFCE component, experiments on multimodal sentiment benchmarks show that DFCE maintains competitive predictive performance while enabling privacy-preserving collaboration under continual updates. We position these results as evidence of algorithmic and systems-level feasibility for privacy-preserving, cross-institutional collaboration, without claiming clinical diagnostic validity.","PeriodicalId":13255,"journal":{"name":"IEEE Transactions on Services Computing","volume":"19 5","pages":"3678-3694"},"PeriodicalIF":6.2,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148596248","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}

引用次数: 0