AI Agent

Workflow Design Analysis for High Resolution Satellite Image Analysis

Ioannis Paraskevakos, Matteo Turilli, Bento Collares Gonçalves, Heather J. Lynch, Shantenu Jha

专题命中 工作流自动化 ：workflow(title)

Comments 10 pages

Network and Agent Dynamics with Evolving Protection against Systemic Risk

Chulwook Park

专题命中 工作流自动化 ：agent(title)

RNASeqR: an R package for automated two-group RNA-Seq analysis workflow

Kuan-Hao Chao, Yi-Wen Hsiao, Yi-Fang Lee, Chien-Yueh Lee, Liang-Chuan Lai, Mong-Hsun Tsai, Tzu-Pin Lu, Eric Y. Chuang

专题命中 工作流自动化 ：workflow(title)

Comments 10 pages, 5 figures

A new method to determine multi-angular reflectance factor from lightweight multispectral cameras with sky sensor in a target-less workflow applicable to UAV

Klaus Schneider-Zapp, Manuel Cubero-Castan, Dai Shi, Christoph Strecha

专题命中 工作流自动化 ：workflow(title)

Journal ref Remote Sensing of Environment 229, 60-68, 2019

A Scientific Workflow System for Satellite Data Processing with Real-Time Monitoring

Minh Duc Nguyen

专题命中 工作流自动化 ：workflow(title)

Comments 4 pages, 1 figure, Mathematical Modeling and Computational Physics 2017 (MMCP 2017)

Journal ref EPJ Web of Conferences 173, 05012 (2018)

Workflow description to dynamically model β-arrestin signaling networks

Romain Yvinec, Mohammed Akli Ayoub, Francesco De Pascali, Pascale Crépieux, Eric Reiter, Anne Poupon

专题命中 工作流自动化 ：workflow(title)

Comments in press

Journal ref Methods in Molecular Biology (2018)

Agent-Based Implementation of Particle Hopping Traffic Model With Stochastic and Queuing Elements

Camilla Champion, Cody Champion

专题命中 工作流自动化 ：agent(title)

Comments Submission to the 2014 COMAP Mathematical Contest in Modeling, minor changes made to improve clarity

Redox agent enhanced chemical mechanical polishing of thin film diamond

Soumen Mandal, Evan L. H. Thomas, Laia Gines, David Morgan, Joshua Green, Emmanuel B. Brousseau, Oliver A. Williams

专题命中 工作流自动化 ：agent(title)

Vertebral body segmentation with GrowCut: Initial experience, workflow and practical application

Jan Egger, Christopher Nimsky, Xiaojun Chen

专题命中 工作流自动化 ：workflow(title)

Comments 10 pages

Journal ref SAGE Open Medicine, Volume 5, pp. 1-10, Nov. 2017

revisit: a Workflow Tool for Data Science

Norman Matloff, Reed Davis, Laurel Beckett, Paul Thompson

专题命中 工作流自动化 ：workflow(title)

Streaming supercomputing needs workflow-enabled programming-in-the-large

Justin M Wozniak, Jonathan Ozik, Daniel S. Katz, Michael Wilde

专题命中 工作流自动化 ：workflow(title)

Finding AND-OR Hierarchies in Workflow Nets

Jacek Sroka, Jan Hidders

专题命中 工作流自动化 ：workflow(title)

Bidding policies for market-based HPC workflow scheduling

Andrew Burkimsher, Leandro Soares Indrusiak

专题命中 工作流自动化 ：workflow(title)

Comments 2nd International Workshop on Dynamic Resource Allocation and Management in Embedded, High Performance and Cloud Computing DREAMCloud 2016 (arXiv:cs/1601.04675), DREAMCloud/2016/06

Stochastic simulators based optimization by Gaussian process metamodels - Application to maintenance investments planning issues

Thomas Browne, Bertrand Iooss, Loïc Le Gratiet, Jérome Lonchampt

专题命中 工作流自动化 ：planning(title)

Comments ENBIS 2015, Sep 2015, Prague, Czech Republic. 2015

Deployment of an Innovative Resource Choice Method for Process Planning

Alexandre Candlot, Nicolas Perry, Alain Bernard, Samar Ammar-Khodja

专题命中 工作流自动化 ：planning(title)

Journal ref CIRP Journal of Manufacturing Systems 35, 5 (2006) 487-506

Agalma: an automated phylogenomics workflow

Casey W. Dunn, Mark Howison, Felipe Zapata

专题命中 工作流自动化 ：workflow(title)

Comments 17 pages, 4 figures

Journal ref BMC Bioinformatics 14 (2013) 330

BW - Eye Ophthalmologic decision support system based on clinical workflow and data mining techniques-image registration algorithm

Ricardo Martins

专题命中 工作流自动化 ：workflow(title)

Network Decontamination with a Single Agent

Yessine Daadaa, Asif Jamshed, Mudassir Shabbir

专题命中 工作流自动化 ：agent(title)

Demonstrating that chlorine dioxide is a size-selective antimicrobial agent and high purity ClO2 can be used as a local antiseptic

Zoltán Noszticzius, Maria Wittmann, Kristóf Kály-Kullai, Zoltán Beregvári, István Kiss, László Rosivall, János Szegedi

专题命中 工作流自动化 ：agent(title)

Comments 32 pages incl. 12 pages of Supplementary information; 6+1 figures, 3 tables

Three Steps to Heaven: Semantic Publishing in a Real World Workflow

Phillip Lord, Simon Cockell, Robert Stevens

专题命中 工作流自动化 ：workflow(title)

Comments Published as part of SePublica 2012

Simulation Factory: Taming Application Configuration and Workflow on High-End Resources

Michael W. Thomas, Erik Schnetter

专题命中 工作流自动化 ：workflow(title)

Comments 10 pages, accepted by CBHPC 2010

Hate: no choice. Agent simulations

Krzysztof Kulakowski, Malgorzata J. Krawczyk, Przemyslaw Gawronski

专题命中 工作流自动化 ：agent(title)

Comments 17 pages, 5 figures

Optimizing Latency and Reliability of Pipeline Workflow Applications

Anne Benoit, Veronika Rehn-Sonigo, Yves Robert

专题命中 工作流自动化 ：workflow(title)

AgentSpec: Speculative Decoding for Batch Inference of LLM Agents

AgentSpec：面向大语言模型智能体批量推理的推测解码算法

Xin Wang, Ziming Miao, Yi Zhu, Hui Shen, Zhongwei Wan, Fan Yang, Mi Zhang

机构 * The Ohio State University（俄亥俄州立大学） ； Microsoft Research（微软研究院） ； University of Michigan（密歇根大学） 纠错

专题命中 工作流自动化 ：agent(abstract)；workflow(abstract)；分类 cs.CL

AI总结 针对LLM智能体批量推理响应慢的问题，本研究提出AgentSpec推测解码算法，通过结构隔离的draft机制和感知冗余的预算分配提升效率，在多工作负载与模型上验证其性能优于现有最优方法。

Comments EMNLP 2026

Markets, Not Planners: Decentralized Orchestration of LLM Agents with Private Information

市场而非规划者：利用私有信息对大语言模型智能体进行去中心化协调

Xiao Liu, Haoyang Li, Songwei Li, Hongbo Fang, Fengli Xu, Feng Shi, James Evans

专题命中 工作流自动化 ：agent(abstract)；agentic(abstract)；分类 cs.CL

AI总结 针对中心化LLM智能体协调的瓶颈与易操控问题，提出去中心化的AgentLance重复劳动力市场机制，经多类任务验证，其匹配专长、转向廉价智能体的表现优于基线方法，还可通过修正市场失灵进一步提升效率。

Comments Working paper

How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks

智能体在自动研究（AutoResearch）中如何失败？针对100项真实前沿研究任务的端到端诊断评估

Yanlin Fei, Nazhou Liu, Xinmiao Yu, Shaolong Chen, Lei Li, Rahul Thapa, Madalina Ciobanu, Navan Preet Singh, Qingqing Mao, Ritankar Das

专题命中 工作流自动化 ：agent(abstract)；agentic(abstract)；分类 cs.CL

AI总结 本研究推出AutoResearchEval评估平台，构建含45种失效模式的ARFT分类体系，发现当前智能体缺乏元认知循环是核心缺陷，公开发布相关资源以推动自主科学发现研究。

Comments *Equal Contribution (alphabetical order by last name)

FinixDoc: Rethinking Financial Document Parsing Beyond Saturated Benchmarks

FinixDoc：重新思考超出饱和基准的金融文档解析

Hang Wang, Jin Zhang, Guoliang Xu, Pengyue Lu, Yao Li, Zijiao Zhang, Tianyu Huang, Weiqi Xiong, Yulong Wang, Chuqiao Lu, Wenkang Huang, Kai Yang, Yadong Li, Hui Li, Xingzhong Xu, Xiao Xu

机构 * Ant Group（蚂蚁集团） 纠错

专题命中 工作流自动化 ：workflow(abstract)；agentic(abstract)；分类 cs.AI

AI总结 本研究针对金融文档解析基准与实际需求的差距，提出端到端智能体解析系统FinixDoc，核心为4B规模视觉语言模型FinixDoc-VL，在自建评估套件FinixDocBench上实现最优性能。

Spike-Killer: Evidence-Gated LLM Assistance for Safe Performance Diagnosis on a Real Windows Workstation

Spike-Killer：面向真实Windows工作站安全性能诊断的证据门控大语言模型辅助方案

Baocheng Zeng, Jinhao Yang

专题命中 工作流自动化 ：agent(abstract)；workflow(abstract)；分类 cs.SE

AI总结 该研究提出Spike-Killer，一种人在环的证据门控工作流，用于安全诊断Windows工作站的帧时间问题，以CS2为目标应用验证了其可审计性，为LLM辅助智能体提供了可信操作模式。

Comments 6 pages. Experience report; no causal CS2 frame-time improvement or autonomous-safety claim is made

Write, Execute, Refine: From Skill Followers to Skill Optimizers via Reinforcement Learning from Execution Feedback

编写、执行、优化：通过执行反馈强化学习从技能跟随者到技能优化器

Kang Peng, Zhiwei Zhang, Yichen Zhang, Zezhong Wang, Yiming Du, Geng Tu, Baojun Wang, Bin Liang, Ruifeng Xu, Kam-Fai Wong

机构 * Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳）） ； The Chinese University of Hong Kong（香港中文大学） ； Huawei Technologies Co., Ltd.（华为技术有限公司） ； Harbin Institute of Technology（哈尔滨工业大学） 纠错

专题命中 工作流自动化 ：agent(abstract)；workflow(abstract)；分类 cs.CL

AI总结 本研究提出WER多阶段框架，通过冻结执行器、程序验证器打分及混合轨迹优化训练技能优化器，在BFCL v4和tau2-bench上显著提升智能体工具使用性能，4B优化器表现优于通用模型。

ML-AutoResearch: Training Machine Learning Research Agents with Automatically Generated Environments

通过合成任务扩展实现AI科学家

Ziyang Cai, Amir Saeidi, Harkirat Behl

机构 * Princeton University（普林斯顿大学） ； Microsoft Research（微软研究院） 纠错

专题命中 工作流自动化 ：agent(abstract_cn)；AI agent(abstract)；分类 cs.AI

AI总结 本文提出一种合成环境生成管道，用于训练能从实践中学习的机器学习代理。通过真实数据集验证和自调试循环，生成高质量合成任务，提升Qwen3-4B和Qwen3-8B在MLGym上的性能，AUP指标分别提升9%和12%。