Emerging Technologies

• Cross-lists
• Replacements

See recent articles

Showing new listings for Friday, 26 September 2025

Cross submissions (showing 12 of 12 entries )

[1] arXiv:2509.20364 (cross-list from cs.AI) [cn-pdf, pdf, html, other]

Title: An Approach to Checking Correctness for Agentic Systems Show Chinese title

Title: 一种检查代理系统正确性的方法

Thomas J Sheffler

Comments: 15 pages, 5 figures

Subjects: Artificial Intelligence (cs.AI) ; Emerging Technologies (cs.ET)

This paper presents a temporal expression language for monitoring AI agent behavior, enabling systematic error-detection of LLM-based agentic systems that exhibit variable outputs due to stochastic generation processes. Drawing from temporal logic techniques used in hardware verification, this approach monitors execution traces of agent tool calls and state transitions to detect deviations from expected behavioral patterns. Current error-detection approaches rely primarily on text matching of inputs and outputs, which proves fragile due to the natural language variability inherent in LLM responses. The proposed method instead focuses on the sequence of agent actions -- such as tool invocations and inter-agent communications -- allowing verification of system behavior independent of specific textual outputs. The temporal expression language provides assertions that capture correct behavioral patterns across multiple execution scenarios. These assertions serve dual purposes: validating prompt engineering and guardrail effectiveness during development, and providing regression testing when agents are updated with new LLMs or modified logic. The approach is demonstrated using a three-agent system, where agents coordinate to solve multi-step reasoning tasks. When powered by large, capable models, all temporal assertions were satisfied across many test runs. However, when smaller models were substituted in two of the three agents, executions violated behavioral assertions, primarily due to improper tool sequencing and failed coordination handoffs. The temporal expressions successfully flagged these anomalies, demonstrating the method's effectiveness for detecting behavioral regressions in production agentic systems. This approach provides a foundation for systematic monitoring of AI agent reliability as these systems become increasingly deployed in critical applications. Show Chinese abstract

本文提出了一种时间表达语言，用于监控人工智能代理的行为，使能够系统地检测基于大语言模型的代理系统的错误，这些系统由于随机生成过程而表现出可变的输出。 从硬件验证中使用的时间逻辑技术中汲取灵感，这种方法监控代理工具调用和状态转换的执行轨迹，以检测与预期行为模式的偏差。 当前的错误检测方法主要依赖于输入和输出的文本匹配，由于大语言模型响应中的自然语言变化性，这种方法显得脆弱。 所提出的方法则专注于代理动作的序列——如工具调用和代理间通信——从而能够在不依赖特定文本输出的情况下验证系统行为。 时间表达语言提供了断言，这些断言在多个执行场景中捕捉正确的行为模式。 这些断言具有双重用途：在开发过程中验证提示工程和安全机制的有效性，并在代理使用新的大语言模型或修改逻辑时提供回归测试。 该方法通过一个三代理系统进行了演示，其中代理协作解决多步骤推理任务。 当由大型、功能强大的模型驱动时，所有时间断言在多次测试运行中均得到满足。 然而，当在三个代理中的两个中替换为较小的模型时，执行违反了行为断言，主要是由于工具顺序不当和协调交接失败。 时间表达式成功地标记了这些异常，证明了该方法在检测生产环境中代理系统行为退化方面的有效性。 该方法为系统地监控人工智能代理的可靠性提供了基础，随着这些系统在关键应用中越来越广泛地部署，这一点尤为重要。

[2] arXiv:2509.20418 (cross-list from cs.CR) [cn-pdf, pdf, other]

Title: A Taxonomy of Data Risks in AI and Quantum Computing (QAI) - A Systematic Review Show Chinese title

Title: AI与量子计算（QAI）中的数据风险分类——系统综述

Grace Billiris, Asif Gill, Madhushi Bandara

Comments: 11 pages, 2 figures, 2 tables

Subjects: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI) ; Emerging Technologies (cs.ET)

Quantum Artificial Intelligence (QAI), the integration of Artificial Intelligence (AI) and Quantum Computing (QC), promises transformative advances, including AI-enabled quantum cryptography and quantum-resistant encryption protocols. However, QAI inherits data risks from both AI and QC, creating complex privacy and security vulnerabilities that are not systematically studied. These risks affect the trustworthiness and reliability of AI and QAI systems, making their understanding critical. This study systematically reviews 67 privacy- and security-related studies to expand understanding of QAI data risks. We propose a taxonomy of 22 key data risks, organised into five categories: governance, risk assessment, control implementation, user considerations, and continuous monitoring. Our findings reveal vulnerabilities unique to QAI and identify gaps in holistic risk assessment. This work contributes to trustworthy AI and QAI research and provides a foundation for developing future risk assessment tools. Show Chinese abstract

量子人工智能（QAI），即人工智能（AI）和量子计算（QC）的结合，有望带来变革性的进展，包括基于AI的量子密码学和抗量子加密协议。 然而，QAI继承了AI和QC的数据风险，造成了复杂隐私和安全漏洞，这些漏洞尚未被系统研究。 这些风险影响AI和QAI系统的可信度和可靠性，因此其理解至关重要。 本研究系统地回顾了67篇与隐私和安全相关的研究，以扩展对QAI数据风险的理解。 我们提出了一种22个关键数据风险的分类法，分为五个类别：治理、风险评估、控制实施、用户考虑因素和持续监控。 我们的研究结果揭示了QAI特有的漏洞，并指出了整体风险评估中的不足。 这项工作为可信AI和QAI的研究做出了贡献，并为开发未来的风险评估工具提供了基础。

[3] arXiv:2509.20547 (cross-list from quant-ph) [cn-pdf, pdf, html, other]

Title: Realization of Graphene Quantum Dots for Innovative Biosensor Development and Diverse Applications Show Chinese title

Title: 石墨烯量子点在创新生物传感器开发及多种应用中的实现

Kumar Gautam, Kumar Shubham, Hitesh Sharma, Divya Punia, Ajay K Sharma, Namisha Gupta, Varun Rathor, Vishakha Singh

Subjects: Quantum Physics (quant-ph) ; Emerging Technologies (cs.ET) ; Atomic and Molecular Clusters (physics.atm-clus)

This paper investigates quantum dots (QDs), which are miniature semiconductor structures with remarkable optical and electrical properties due to quantum confinement processes. Traditional QDs, such as CdTe, have been extensively investigated; however, they frequently exhibit toxicity and stability issues. Graphene quantum dots (GQDs) are emerging as a safer and more stable alternative to traditional QDs. GQDs are honeycomb-lattice carbon atoms with unique electronic and optical properties that make them promising candidates for biomedical, electronic, and energy storage applications. GQD synthesis methods (top-down and bottom-up) and their advantages over standard QDs include better photostability, biocompatibility, and configurable band gaps. GQDs are perfect for real-world uses like sensitive biosensing, real-time food safety monitoring, and smart packaging because of their low toxicity, high sensitivity, and affordability. These uses are all essential for cutting down on food grain waste. This emphasizes the growing significance of GQDs in advancing nanotechnology and their potential integration with quantum technologies, paving the door for creative solutions in biosensing, food safety, environmental monitoring, and future quantum electronics. Show Chinese abstract

本文研究了量子点（QDs），由于量子限制过程，这些微型半导体结构具有显著的光学和电学特性。 传统的QDs，如CdTe，已被广泛研究；然而，它们经常表现出毒性和稳定性问题。 石墨烯量子点（GQDs）正作为一种更安全、更稳定的传统QDs替代品而出现。 GQDs是由蜂窝状晶格的碳原子构成，具有独特的电子和光学特性，使它们成为生物医学、电子和能量存储应用的有前途的候选材料。 GQD的合成方法（自上而下和自下而上）及其优于标准QDs的优势包括更好的光稳定性、生物相容性和可调节的带隙。 由于低毒性、高灵敏度和成本低廉，GQDs非常适合实际应用，如灵敏的生物传感、实时食品安全监测和智能包装。 这些应用对于减少粮食浪费都至关重要。 这强调了GQDs在推进纳米技术和与量子技术潜在整合中的日益重要性，为生物传感、食品安全、环境监测和未来量子电子领域的创新解决方案铺平了道路。

[4] arXiv:2509.20603 (cross-list from cs.DC) [cn-pdf, pdf, html, other]

Title: Experience Deploying Containerized GenAI Services at an HPC Center Show Chinese title

Title: 在HPC中心部署容器化GenAI服务的经验

Angel M. Beltre, Jeff Ogden, Kevin Pedretti

Comments: 10 pages, 12 figures

Subjects: Distributed, Parallel, and Cluster Computing (cs.DC) ; Artificial Intelligence (cs.AI) ; Hardware Architecture (cs.AR) ; Emerging Technologies (cs.ET) ; Machine Learning (cs.LG)

Generative Artificial Intelligence (GenAI) applications are built from specialized components -- inference servers, object storage, vector and graph databases, and user interfaces -- interconnected via web-based APIs. While these components are often containerized and deployed in cloud environments, such capabilities are still emerging at High-Performance Computing (HPC) centers. In this paper, we share our experience deploying GenAI workloads within an established HPC center, discussing the integration of HPC and cloud computing environments. We describe our converged computing architecture that integrates HPC and Kubernetes platforms running containerized GenAI workloads, helping with reproducibility. A case study illustrates the deployment of the Llama Large Language Model (LLM) using a containerized inference server (vLLM) across both Kubernetes and HPC platforms using multiple container runtimes. Our experience highlights practical considerations and opportunities for the HPC container community, guiding future research and tool development. Show Chinese abstract

生成式人工智能（GenAI）应用程序由专用组件构建——推理服务器、对象存储、向量和图数据库以及用户界面——通过基于网络的API相互连接。 尽管这些组件通常被容器化并在云环境中部署，但这些能力在高性能计算（HPC）中心仍处于发展初期。 在本文中，我们分享了在已建立的HPC中心部署GenAI工作负载的经验，讨论了HPC和云计算环境的集成。 我们描述了整合HPC和Kubernetes平台的融合计算架构，该架构运行容器化的GenAI工作负载，有助于可重复性。 一个案例研究说明了使用容器化推理服务器（vLLM）在Kubernetes和HPC平台上跨多个容器运行时部署Llama大型语言模型（LLM）。 我们的经验突出了HPC容器社区的实际考虑因素和机遇，为未来的研究和工具开发提供了指导。

[5] arXiv:2509.20663 (cross-list from quant-ph) [cn-pdf, pdf, other]

Title: A Review on Quantum Circuit Optimization using ZX-Calculus Show Chinese title

Title: 基于ZX-演算的量子电路优化综述

Tobias Fischbach, Pierre Talbot, Pascal Bourvry

Subjects: Quantum Physics (quant-ph) ; Emerging Technologies (cs.ET)

Quantum computing promises significant speed-ups for certain algorithms but the practical use of current noisy intermediate-scale quantum (NISQ) era computers remains limited by resources constraints (e.g., noise, qubits, gates, and circuit depth). Quantum circuit optimization is a key mitigation strategy. In this context, ZX-calculus has emerged as an alternative framework that allows for semantics-preserving quantum circuit optimization. We review ZX-based optimization of quantum circuits, categorizing them by optimization techniques, target metrics and intended quantum computing architecture. In addition, we outline critical challenges and future research directions, such as multi-objective optimization, scalable algorithms, and enhanced circuit extraction methods. This survey is valuable for researchers in both combinatorial optimization and quantum computing. For researchers in combinatorial optimization, we provide the background to understand a new challenging combinatorial problem: ZX-based quantum circuit optimization. For researchers in quantum computing, we classify and explain existing circuit optimization techniques. Show Chinese abstract

量子计算对于某些算法有望实现显著的速度提升，但当前噪声中等规模量子（NISQ）时代计算机的实际应用仍受限于资源约束（例如，噪声、量子位、门和电路深度）。量子电路优化是一种关键的缓解策略。在此背景下，ZX演算作为一种替代框架出现，允许进行语义保持的量子电路优化。我们回顾了基于ZX的量子电路优化，根据优化技术、目标指标和预期的量子计算架构对其进行分类。此外，我们概述了关键挑战和未来研究方向，如多目标优化、可扩展算法和增强的电路提取方法。这项综述对组合优化和量子计算领域的研究人员都有价值。对于组合优化领域的研究人员，我们提供了理解一个新的具有挑战性的组合问题：基于ZX的量子电路优化的背景。对于量子计算领域的研究人员，我们对现有的电路优化技术进行了分类和解释。

[6] arXiv:2509.20741 (cross-list from eess.AS) [cn-pdf, pdf, html, other]

Title: Real-Time System for Audio-Visual Target Speech Enhancement Show Chinese title

Title: 用于音视频目标语音增强的实时系统

T. Aleksandra Ma, Sile Yin, Li-Chia Yang, Shuo Zhang

Comments: Accepted into WASPAA 2025 demo session

Subjects: Audio and Speech Processing (eess.AS) ; Emerging Technologies (cs.ET) ; Machine Learning (cs.LG)

We present a live demonstration for RAVEN, a real-time audio-visual speech enhancement system designed to run entirely on a CPU. In single-channel, audio-only settings, speech enhancement is traditionally approached as the task of extracting clean speech from environmental noise. More recent work has explored the use of visual cues, such as lip movements, to improve robustness, particularly in the presence of interfering speakers. However, to our knowledge, no prior work has demonstrated an interactive system for real-time audio-visual speech enhancement operating on CPU hardware. RAVEN fills this gap by using pretrained visual embeddings from an audio-visual speech recognition model to encode lip movement information. The system generalizes across environmental noise, interfering speakers, transient sounds, and even singing voices. In this demonstration, attendees will be able to experience live audio-visual target speech enhancement using a microphone and webcam setup, with clean speech playback through headphones. Show Chinese abstract

我们展示了一个RAVEN的实时演示，这是一个设计为完全在CPU上运行的实时音视频语音增强系统。 在单通道、仅音频的设置中，语音增强传统上被当作从环境噪声中提取干净语音的任务。 最近的工作探索了使用视觉提示（如嘴唇运动）来提高鲁棒性，特别是在有干扰说话者的情况下。 然而，据我们所知，之前没有工作展示了在CPU硬件上运行的实时音视频语音增强交互系统。 RAVEN通过使用来自音视频语音识别模型的预训练视觉嵌入来编码嘴唇运动信息，填补了这一空白。 该系统能够跨环境噪声、干扰说话者、瞬态声音甚至歌唱声音进行泛化。 在本次演示中，与会者将能够通过麦克风和网络摄像头设置体验实时音视频目标语音增强，并通过耳机播放干净语音。

[7] arXiv:2509.20933 (cross-list from cs.LO) [cn-pdf, pdf, other]

Title: A Coalgebraic Model of Quantum Bisimulation Show Chinese title

Title: 一种量子对等性的代数模型

Lorenzo Ceragioli (IMT School for Advanced Studies, Lucca, Italy), Elena Di Lavore (University of Pisa, Italy), Giuseppe Lomurno (University of Pisa, Italy), Gabriele Tedeschi (University of Pisa, Italy)

Comments: In Proceedings ACT 2024, arXiv:2509.18357

Journal-ref: EPTCS 429, 2025, pp. 249-269

Subjects: Logic in Computer Science (cs.LO) ; Emerging Technologies (cs.ET)

Recent works have shown that defining a behavioural equivalence that matches the observational properties of a quantum-capable, concurrent, non-deterministic system is a surprisingly difficult task. We explore coalgebras over distributions taking weights from a generic effect algebra, which subsumes probabilities and quantum effects, a physical formalism that represents the probabilistic behaviour of an open quantum system. To abide by the properties of quantum theory, we introduce monads graded on a partial commutative monoid, intuitively allowing composition of two processes only if they use different quantum resources, as prescribed by the no-cloning theorem. We investigate the relation between an open quantum system and its probabilistic counterparts obtained when instantiating the input with a specific quantum state. We consider Aczel-Mendler and kernel bisimilarities, advocating for the latter as it characterizes quantum systems that exhibit the same probabilistic behaviour for all input states. Finally, we propose operators on quantum effect labelled transition systems, paving the way for a process calculi semantics that is parametric over the quantum input. Show Chinese abstract

最近的研究表明，定义一个行为等价性，使其匹配具有量子能力的并发非确定性系统的观察特性，是一个令人惊讶的困难任务。 我们探讨了在分布上使用来自通用效应代数的权重的余代数，这涵盖了概率和量子效应，这是一种表示开放量子系统概率行为的物理形式。 为了遵守量子理论的性质，我们引入了一个在部分交换独异点上分级的单子，直观上允许仅当两个过程使用不同的量子资源时才能进行组合，这由不可克隆定理规定。 我们研究了开放量子系统与其在将输入实例化为特定量子状态时获得的概率对应物之间的关系。 我们考虑了Aczel-Mendler和核双模拟性，主张后者，因为它可以表征对于所有输入状态都表现出相同概率行为的量子系统。 最后，我们提出了量子效应标记转移系统的运算符，为一种对量子输入参数化的进程演算语义铺平了道路。

[8] arXiv:2509.20943 (cross-list from cs.CR) [cn-pdf, pdf, html, other]

Title: CTI Dataset Construction from Telegram Show Chinese title

Title: 从电报构建CTI数据集

Dincy R. Arikkat, Sneha B. T., Serena Nicolazzo, Antonino Nocera, Vinod P., Rafidha Rehiman K. A., Karthika R

Subjects: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI) ; Emerging Technologies (cs.ET)

Cyber Threat Intelligence (CTI) enables organizations to anticipate, detect, and mitigate evolving cyber threats. Its effectiveness depends on high-quality datasets, which support model development, training, evaluation, and benchmarking. Building such datasets is crucial, as attack vectors and adversary tactics continually evolve. Recently, Telegram has gained prominence as a valuable CTI source, offering timely and diverse threat-related information that can help address these challenges. In this work, we address these challenges by presenting an end-to-end automated pipeline that systematically collects and filters threat-related content from Telegram. The pipeline identifies relevant Telegram channels and scrapes 145,349 messages from 12 curated channels out of 150 identified sources. To accurately filter threat intelligence messages from generic content, we employ a BERT-based classifier, achieving an accuracy of 96.64%. From the filtered messages, we compile a dataset of 86,509 malicious Indicators of Compromise, including domains, IPs, URLs, hashes, and CVEs. This approach not only produces a large-scale, high-fidelity CTI dataset but also establishes a foundation for future research and operational applications in cyber threat detection. Show Chinese abstract

网络威胁情报（CTI）使组织能够预测、检测和缓解不断演变的网络威胁。 其有效性取决于高质量的数据集，这些数据集支持模型开发、训练、评估和基准测试。 构建这样的数据集至关重要，因为攻击向量和对手战术在不断演变。 最近，Telegram作为有价值的CTI来源而受到关注，提供了及时且多样化的威胁相关信息，有助于解决这些挑战。 在本工作中，我们通过提出一个端到端的自动化流程来解决这些挑战，该流程系统地从Telegram中收集和过滤与威胁相关的内容。 该流程识别相关的Telegram频道，并从150个已识别源中的12个精选频道中抓取145,349条消息。 为了准确地从通用内容中过滤威胁情报消息，我们采用了一个基于BERT的分类器，达到了96.64%的准确率。 从过滤后的消息中，我们整理了一个包含86,509个恶意指标的数据集，包括域名、IP地址、URL、哈希值和CVE。 这种方法不仅生成了一个大规模、高保真的CTI数据集，还为未来在网络威胁检测领域的研究和实际应用奠定了基础。

[9] arXiv:2509.21026 (cross-list from cs.NI) [cn-pdf, pdf, html, other]

Title: A Novel Integrated Architecture for Intent Based Approach and Zero Touch Networks Show Chinese title

Title: 一种基于意图的方法和零接触网络的新型集成架构

Neelam Gupta, Dibakar Das, Tamizhelakkiya K, Uma Maheswari Natarajan, Sharvari Ravindran, Komal Sharma, Jyotsna Bapat, Debabrata Das

Subjects: Networking and Internet Architecture (cs.NI) ; Emerging Technologies (cs.ET)

The transition to Sixth Generation (6G) networks presents challenges in managing quality of service (QoS) of diverse applications and achieving Service Level Agreements (SLAs) under varying network conditions. Hence, network management must be automated with the help of Machine Learning (ML) and Artificial Intelligence (AI) to achieve real-time requirements. Zero touch network (ZTN) is one of the frameworks to automate network management with mechanisms such as closed loop control to ensure that the goals are met perpetually. Intent- Based Networking (IBN) specifies the user intents with diverse network requirements or goals which are then translated into specific network configurations and actions. This paper presents a novel architecture for integrating IBN and ZTN to serve the intent goals. Users provides the intent in the form of natural language, e.g., English, which is then translated using natural language processing (NLP) techniques (e.g., retrieval augmented generation (RAG)) into Network Intent LanguagE (Nile). The Nile intent is then passed on to the BiLSTM and Q-learning based ZTN closed loop framework as a goal which maintains the intent under varying network conditions. Thus, the proposed architecture can work autonomously to ensure the network performance goal is met by just specifying the user intent in English. The integrated architecture is also implemented on a testbed using OpenAirInterface (OAI). Additionally, to evaluate the architecture, an optimization problem is formulated which evaluated with Monte Carlo simulations. Results demonstrate how ZTN can help achieve the bandwidth goals autonomously set by user intent. The simulation and the testbed results are compared and they show similar trend. Mean Opinion Score (MOS) for Quality of Experience (QoE) is also measured to indicate the user satisfaction of the intent. Show Chinese abstract

第六代（6G）网络的过渡在管理多样化应用的服务质量（QoS）和在不同网络条件下实现服务等级协议（SLAs）方面带来了挑战。因此，必须借助机器学习（ML）和人工智能（AI）来自动化网络管理，以满足实时需求。零接触网络（ZTN）是一种通过闭环控制等机制自动化网络管理的框架，以确保目标持续达成。基于意图的网络（IBN）用不同的网络需求或目标来指定用户意图，然后将其转换为具体的网络配置和操作。本文提出了一种集成IBN和ZTN的新架构，以实现意图目标。用户以自然语言（例如英语）形式提供意图，然后使用自然语言处理（NLP）技术（例如检索增强生成（RAG））将其翻译成网络意图语言（Nile）。然后将Nile意图作为目标传递给基于BiLSTM和Q-learning的ZTN闭环框架，该框架在不同网络条件下维持意图。因此，所提出的架构可以自主运行，只需指定用户的英语意图即可确保网络性能目标的实现。该集成架构还使用OpenAirInterface（OAI）在测试平台上进行了实现。此外，为了评估该架构，制定了一项优化问题，并通过蒙特卡洛模拟进行评估。结果展示了ZTN如何帮助自主实现由用户意图设定的带宽目标。仿真和测试平台结果进行了比较，显示出相似的趋势。还测量了用户体验质量（QoE）的平均意见分数（MOS），以表明用户对意图的满意度。

[10] arXiv:2509.21039 (cross-list from cs.DC) [cn-pdf, pdf, html, other]

Title: Mojo: MLIR-Based Performance-Portable HPC Science Kernels on GPUs for the Python Ecosystem Show Chinese title

Title: Mojo：基于MLIR的高性能计算科学内核，适用于Python生态系统的GPU性能可移植性

William F. Godoy, Tatiana Melnichenko, Pedro Valero-Lara, Wael Elwasif, Philip Fackler, Rafael Ferreira Da Silva, Keita Teranishi, Jeffrey S. Vetter

Comments: Accepted at the IEEE/ACM SC25 Conference WACCPD Workshop. The International Conference for High Performance Computing, Networking, Storage, and Analysis, St. Louis, MO, Nov 16-21, 2025. 15 pages, 7 figures. WFG and TM contributed equally

Subjects: Distributed, Parallel, and Cluster Computing (cs.DC) ; Computational Engineering, Finance, and Science (cs.CE) ; Emerging Technologies (cs.ET) ; Programming Languages (cs.PL)

We explore the performance and portability of the novel Mojo language for scientific computing workloads on GPUs. As the first language based on the LLVM's Multi-Level Intermediate Representation (MLIR) compiler infrastructure, Mojo aims to close performance and productivity gaps by combining Python's interoperability and CUDA-like syntax for compile-time portable GPU programming. We target four scientific workloads: a seven-point stencil (memory-bound), BabelStream (memory-bound), miniBUDE (compute-bound), and Hartree-Fock (compute-bound with atomic operations); and compare their performance against vendor baselines on NVIDIA H100 and AMD MI300A GPUs. We show that Mojo's performance is competitive with CUDA and HIP for memory-bound kernels, whereas gaps exist on AMD GPUs for atomic operations and for fast-math compute-bound kernels on both AMD and NVIDIA GPUs. Although the learning curve and programming requirements are still fairly low-level, Mojo can close significant gaps in the fragmented Python ecosystem in the convergence of scientific computing and AI. Show Chinese abstract

我们探索了新型Mojo语言在GPU上的科学计算工作负载的性能和可移植性。 作为第一个基于LLVM的多级中间表示（MLIR）编译器基础设施的语言，Mojo旨在通过结合Python的互操作性和类似CUDA的语法，实现编译时可移植的GPU编程，从而缩小性能和生产率的差距。 我们针对四个科学工作负载进行测试：七点模板（内存受限）、BabelStream（内存受限）、miniBUDE（计算受限）和Hartree-Fock（带有原子操作的计算受限）；并在NVIDIA H100和AMD MI300A GPU上与供应商基准进行比较。 我们表明，Mojo在内存受限内核上的性能与CUDA和HIP相当，但在AMD GPU上的原子操作以及在AMD和NVIDIA GPU上的快速数学计算受限内核上存在差距。 尽管学习曲线和编程要求仍然较为底层，Mojo可以在科学计算和AI的融合中缩小Python生态系统中的重大差距。

[11] arXiv:2509.21137 (cross-list from cs.DC) [cn-pdf, pdf, html, other]

Title: From GPUs to RRAMs: Distributed In-Memory Primal-Dual Hybrid Gradient Method for Solving Large-Scale Linear Optimization Problem Show Chinese title

Title: 从GPU到RRAM：用于解决大规模线性优化问题的分布式内存内原始对偶混合梯度方法

Huynh Q. N. Vo, Md Tawsif Rahman Chowdhury, Paritosh Ramanan, Gozde Tutuncuoglu, Junchi Yang, Feng Qiu, Murat Yildirim

Comments: Main Article (12 Pages, 3 Figures), Appendix (4 Pages)

Subjects: Distributed, Parallel, and Cluster Computing (cs.DC) ; Hardware Architecture (cs.AR) ; Emerging Technologies (cs.ET)

The exponential growth of computational workloads is surpassing the capabilities of conventional architectures, which are constrained by fundamental limits. In-memory computing (IMC) with RRAM provides a promising alternative by providing analog computations with significant gains in latency and energy use. However, existing algorithms developed for conventional architectures do not translate to IMC, particularly for constrained optimization problems where frequent matrix reprogramming remains cost-prohibitive for IMC applications. Here we present a distributed in-memory primal-dual hybrid gradient (PDHG) method, specifically co-designed for arrays of RRAM devices. Our approach minimizes costly write cycles, incorporates robustness against device non-idealities, and leverages a symmetric block-matrix formulation to unify operations across distributed crossbars. We integrate a physics-based simulation framework called MELISO+ to evaluate performance under realistic device conditions. Benchmarking against GPU-accelerated solvers on large-scale linear programs demonstrates that our RRAM-based solver achieves comparable accuracy with up to three orders of magnitude reductions in energy consumption and latency. These results demonstrate the first PDHG-based LP solver implemented on RRAMs, showcasing the transformative potential of algorithm-hardware co-design for solving large-scale optimization through distributed in-memory computing. Show Chinese abstract

计算工作负载的指数增长正在超越传统架构的能力，这些架构受到基本限制的约束。 基于RRAM的存内计算（IMC）通过提供模拟计算，显著提高了延迟和能耗的效率，提供了有希望的替代方案。 然而，为传统架构开发的现有算法并不适用于IMC，特别是对于约束优化问题，频繁的矩阵重新编程对IMC应用来说成本过高。 在这里，我们提出了一种分布式存内原始对偶混合梯度（PDHG）方法，专门针对RRAM器件阵列进行协同设计。 我们的方法减少了昂贵的写入周期，增强了对器件非理想性的鲁棒性，并利用对称块矩阵公式在分布式交叉条中统一操作。 我们集成了一种基于物理的仿真框架MELISO+，以在现实设备条件下评估性能。 与大规模线性规划的GPU加速求解器进行基准测试表明，我们的RRAM求解器在能耗和延迟方面实现了高达三个数量级的减少，同时保持了相当的准确性。 这些结果展示了第一个基于PDHG的在线性规划求解器在RRAM上的实现，展示了算法-硬件协同设计在通过分布式存内计算解决大规模优化问题方面的变革潜力。

[12] arXiv:2509.21147 (cross-list from cs.CR) [cn-pdf, pdf, html, other]

Title: Emerging Paradigms for Securing Federated Learning Systems Show Chinese title

Title: 新兴的用于保护联邦学习系统的范式

Amr Akmal Abouelmagd, Amr Hilal

Subjects: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI) ; Emerging Technologies (cs.ET) ; Machine Learning (cs.LG)

Federated Learning (FL) facilitates collaborative model training while keeping raw data decentralized, making it a conduit for leveraging the power of IoT devices while maintaining privacy of the locally collected data. However, existing privacy- preserving techniques present notable hurdles. Methods such as Multi-Party Computation (MPC), Homomorphic Encryption (HE), and Differential Privacy (DP) often incur high compu- tational costs and suffer from limited scalability. This survey examines emerging approaches that hold promise for enhancing both privacy and efficiency in FL, including Trusted Execution Environments (TEEs), Physical Unclonable Functions (PUFs), Quantum Computing (QC), Chaos-Based Encryption (CBE), Neuromorphic Computing (NC), and Swarm Intelligence (SI). For each paradigm, we assess its relevance to the FL pipeline, outlining its strengths, limitations, and practical considerations. We conclude by highlighting open challenges and prospective research avenues, offering a detailed roadmap for advancing secure and scalable FL systems. Show Chinese abstract

联邦学习（FL）在保持原始数据去中心化的同时促进了协作模型训练，使其成为利用物联网设备力量同时保持本地收集数据隐私的途径。 然而，现有的隐私保护技术存在显著障碍。 诸如多方计算（MPC）、同态加密（HE）和差分隐私（DP）的方法通常会带来高昂的计算成本，并且可扩展性有限。 本综述考察了有望提升FL中隐私性和效率的新兴方法，包括可信执行环境（TEEs）、物理不可克隆函数（PUFs）、量子计算（QC）、基于混沌的加密（CBE）、神经形态计算（NC）和群体智能（SI）。 对于每种范式，我们评估其与FL流程的相关性，概述其优势、局限性和实际考虑因素。 最后，我们强调了开放性的挑战和未来的研究方向，为推进安全且可扩展的FL系统提供了详细路线图。

Replacement submissions (showing 2 of 2 entries )

[13] arXiv:2505.02003 (replaced) [cn-pdf, pdf, other]

Title: Closed-loop control of seizure activity via real-time seizure forecasting by reservoir neuromorphic computing Show Chinese title

Title: 通过基于储备神经形态计算的实时癫痫发作预测实现癫痫活动的闭环控制

Maryam Sadeghi, Darío Fernández Khatiboun, Yasser Rezaeiyan, Saima Rizwan, Alessandro Barcellona, Andrea Merello, Marco Crepaldi, Gabriella Panuccio, Farshad Moradi

Subjects: Artificial Intelligence (cs.AI) ; Emerging Technologies (cs.ET) ; Human-Computer Interaction (cs.HC)

Closed-loop brain stimulation holds potential as personalized treatment for drug-resistant epilepsy (DRE) but still suffers from limitations that result in highly variable efficacy. First, stimulation is typically delivered upon detection of the seizure to abort rather than prevent it; second, the stimulation parameters are established by trial and error, requiring lengthy rounds of fine-tuning, which delay steady-state therapeutic efficacy. Here, we address these limitations by leveraging the potential of neuromorphic computing. We present a neuromorphic reservoir computing hardware system capable of driving real-time personalized free-run stimulations based on seizure forecasting, wherein each forecast triggers an electrical pulse rather than an arbitrarily predefined fixed-frequency stimulus train. The system achieves 83.33% accuracy in forecasting seizure occurrences during the training phase. We validate the system using hippocampal spheroids coupled to 3D microelectrode array as a simplified testbed, achieving seizure reduction >97% during the real-time processing while primarily using instantaneous stimulation frequencies within 20 Hz, well below what typically used in clinical practice. Our work demonstrates the potential of neuromorphic systems as a next-generation neuromodulation strategy for personalized DRE treatment, leveraging their sparse and event-driven processing for real-time applications. Show Chinese abstract

闭环脑刺激作为药物难治性癫痫（DRE）的个性化治疗具有潜力，但仍存在导致疗效高度变异的局限性。首先，刺激通常在检测到癫痫发作时进行以终止而非预防；其次，刺激参数是通过试错法确定的，需要多次精细调整，这延迟了稳定状态的治疗效果。在这里，我们通过利用类神经计算的潜力来解决这些限制。我们提出了一种类神经储备计算硬件系统，能够根据癫痫预测驱动实时个性化的自由运行刺激，其中每次预测都会触发一个电脉冲，而不是任意预定义的固定频率刺激序列。该系统在训练阶段对癫痫发作的预测准确率达到83.33%。我们使用与三维微电极阵列耦合的海马体球状体作为简化测试平台验证了该系统，在实时处理期间实现了超过97%的癫痫减少率，主要使用低于20赫兹的瞬时刺激频率，远低于临床实践中通常使用的频率。我们的工作展示了类神经系统作为下一代神经调节策略在个性化DRE治疗中的潜力，利用其稀疏和事件驱动的处理方式用于实时应用。

[14] arXiv:2509.19395 (replaced) [cn-pdf, pdf, html, other]

Title: HARLI CQUINN: Higher Adjusted Randomness with Linear In Complexity QUantum INspired Networks for K-Means Show Chinese title

Title: HARLI CQUINN：用于K均值的线性复杂度量子启发网络的更高调整随机性

Jiten Oswal, Saumya Biswas

Subjects: Quantum Physics (quant-ph) ; Emerging Technologies (cs.ET)

We contrast a minimalistic implementation of quantum k-means algorithm to classical k-means algorithm. With classical simulation results, we demonstrate a quantum performance, on and above par, with the classical k-means algorithm. We present benchmarks of its accuracy for test cases of both well-known and experimental datasets. Despite extensive research into quantum k-means algorithms, our approach reveals previously unexplored methodological improvements. The encoding step can be minimalistic with classical data imported into quantum states more directly than existing approaches. The proposed quantum-inspired algorithm performs better in terms of accuracy and Adjusted Rand Index (ARI) with respect to the bare classical k-means algorithm. By investigating multiple encoding strategies, we provide nuanced insights into quantum computational clustering techniques. Show Chinese abstract

我们对比了量子k均值算法的最小实现与经典k均值算法。 通过经典模拟结果，我们展示了量子性能，其表现与经典k均值算法相当甚至更优。 我们提供了对已知和实验数据集测试用例的准确性基准。 尽管对量子k均值算法进行了大量研究，但我们的方法揭示了之前未探索的方法改进。 编码步骤可以是最小化的，与现有方法相比，经典数据可以更直接地导入量子状态。 提出的量子启发算法在准确性和调整后的兰德指数（ARI）方面优于原始的经典k均值算法。 通过研究多种编码策略，我们提供了对量子计算聚类技术的细致见解。