研究方向热度分析

未来发展趋势预测

作者关系与合作网络

graph TD
    subgraph AI_Agent["AI Agent系统"]
        YZ1[Yusheng Zheng]
        JF[Jiakun Fan]
        QF[Quanzhi Fu]
        CW[Cong Wang]
        JZ[Jing Zhang]
        CS[Cosmo Santoni]
        HK[Hao Kang]
        
        YZ1 --> JF
        YZ1 --> QF
        YZ1 --> CW
    end
    
    subgraph Memory["内存管理"]
        KZ[Kaiyang Zhao]
        NG[Neha Gholkar]
        HM[Hasan Maruf]
        FS[Felix Schuhknecht]
        NR[Nick Rassau]
        
        KZ --> NG
        KZ --> HM
        FS --> NR
    end
    
    subgraph State_Evo["状态演化"]
        RJM[R. Jay Martin]
    end
    
    subgraph Quantum["量子计算"]
        JZY[John Zhuoyang Ye]
        JW[Jiyuan Wang]
        YQ[Yifan Qiao]
        JP[Jens Palsberg]
        
        JZY --> JW
        JZY --> YQ
        JZY --> JP
    end
    
    subgraph Realtime["实时系统"]
        RF[Rong Fu]
        YM[Yibo Meng]
        YZ2[Yuanhai Zhang]
        
        RF --> YM
    end
    
    style AI_Agent fill:#3b82f6,stroke:#1d4ed8,color:#fff
    style Memory fill:#8b5cf6,stroke:#6d28d9,color:#fff
    style State_Evo fill:#10b981,stroke:#059669,color:#fff
    style Quantum fill:#f59e0b,stroke:#d97706,color:#fff
    style Realtime fill:#ef4444,stroke:#dc2626,color:#fff
                        

核心研究机构分析

关键技术创新

方法论创新

论文详细列表

• Equilibria: Fair Multi-Tenant CXL Memory Tiering At Scale
  Kaiyang Zhao, Neha Gholkar, Hasan Maruf, Abhishek Dhanotia, Johannes Weiner, Gregory Price, Ning Sun, Bhavya Dwivedi, Stuart Clark, Dimitrios Skarlatos
  数据中心内存扩展的CXL解决方案，解决多租户环境下的内存层级化公平调度问题。
  CXL内存管理多租户 展开摘要
  Memory dominates datacenter system cost and power. Memory expansion via Compute Express Link (CXL) is an effective way to provide additional memory at lower cost and power, but its effective use requires software-level tiering for hyperscaler workloads. Existing tiering solutions, including current Linux support, face fundamental limitations in production deployments. First, they lack multi-tenancy support, failing to handle stacked homogeneous or heterogeneous workloads...
• Taking the Leap: Efficient and Reliable Fine-Grained NUMA Migration in User-space
  Felix Schuhknecht, Nick Rassau
  用户态NUMA内存迁移机制，优化跨NUMA区域的并行查询性能。
  NUMA内存迁移用户态 展开摘要
  Modern multi-socket architectures offer a single virtual address space, but physically divide main-memory across multiple regions, where each region is attached to a CPU and its cores. While this simplifies the usage, developers must be aware of non-uniform memory access (NUMA), where an access by a thread running on a core-local NUMA region is significantly cheaper than an access from a core-remote region...
• AgentCgroup: Understanding and Controlling OS Resources of AI Agents
  Yusheng Zheng, Jiakun Fan, Quanzhi Fu, Yiwei Yang, Wei Zhang, Andi Quinn
  首次系统性分析AI Agent在沙箱容器中的OS级资源动态，基于144个任务的实证研究。
  AI Agent资源控制cgroup 展开摘要
  AI agents are increasingly deployed in multi-tenant cloud environments, where they execute diverse tool calls within sandboxed containers, each call with distinct resource demands and rapid fluctuations. We present a systematic characterization of OS-level resource dynamics in sandboxed AI coding agents, analyzing 144 software engineering tasks from the SWE-rebench benchmark across two LLM models...
• Contextual Memory Virtualisation: DAG-Based State Management for LLM Agents
  Cosmo Santoni
  借鉴虚拟内存思想，将LLM累积理解作为版本控制的状态进行管理。
  LLM虚拟内存状态管理 展开摘要
  As large language models engage in extended reasoning tasks, they accumulate significant state -- architectural mappings, trade-off decisions, codebase conventions -- within the context window. This understanding is lost when sessions reach context limits and undergo lossy compaction. We propose Contextual Memory Virtualisation (CMV), a system that treats accumulated LLM understanding as version-controlled state...
• HALO: A Fine-Grained Resource Sharing Quantum Operating System
  John Zhuoyang Ye, Jiyuan Wang, Yifan Qiao, Jens Palsberg
  首个支持量子处理器细粒度资源共享的操作系统，解决量子云平台的资源利用率问题。
  量子计算操作系统资源共享 展开摘要
  As quantum computing enters the cloud era, thousands of users must share access to a small number of quantum processors. Users need to wait minutes to days to start their jobs, which only takes a few seconds for execution. Current quantum cloud platforms employ a fair-share scheduler, as there is no way to multiplex a quantum computer among multiple programs at the same time...
• Fork, Explore, Commit: OS Primitives for Agentic Exploration
  Cong Wang, Yusheng Zheng
  为Agent探索引入branch context抽象，提供copy-on-write状态隔离和原子提交语义。
  AI Agent进程隔离原子操作 展开摘要
  AI agents increasingly perform agentic exploration: pursuing multiple solution paths in parallel and committing only the successful one. Because each exploration path may modify files and spawn processes, agents require isolated environments with atomic commit and rollback semantics for both filesystem state and process state. We introduce the branch context, a new OS abstraction...
• Right to History: A Sovereignty Kernel for Verifiable AI Agent Execution
  Jing Zhang
  提出主权内核概念，为AI Agent执行提供防篡改、可验证的历史记录。
  AI Agent安全审计 展开摘要
  AI agents increasingly act on behalf of humans, yet no existing system provides a tamper-evident, independently verifiable record of what they did. As regulations such as the EU AI Act begin mandating automatic logging for high-risk AI systems, this gap carries concrete consequences -- especially for agents running on personal hardware, where no centralized provider controls the log...
• Bounded Local Generator Classes for Deterministic State Evolution
  R. Jay Martin
  形式化定义有界局部生成器类，证明增量更新成本与系统维度无关。
  形式化方法状态演化确定性系统 展开摘要
  We formalize a constructive subclass of locality-preserving deterministic operators acting on graph-indexed state systems. We define the class of Bounded Local Generator Classes (BLGC), consisting of finite-range generators operating on bounded state spaces under deterministic composition. Within this class, incremental update cost is independent of total system dimension...
• ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System
  Hao Kang, Ziyang Li, Xinyu Yang, Weili Xu, Yinfang Chen, Junxiong Wang, Beidi Chen, Tushar Krishna, Chenfeng Xu, Simran Arora
  端到端感知的Agent推理系统，优化KV缓存管理和工具编排。
  AI Agent推理系统KV缓存 展开摘要
  Large language models(LLMs) are now used to power complex multi-turn agentic workflows. Existing systems run agentic inference by loosely assembling isolated components: an LLM inference engine (e.g., vLLM) and a tool orchestrator (e.g., Kubernetes). Although agentic workflows involve multiple LLM and tool requests, these systems schedule and allocate resources separately on a per-request basis...
• Exploiting Dependency and Parallelism: Real-Time Scheduling and Analysis for GPU Tasks
  Yuanhai Zhang, Songyang He, Ruizhe Gou, Mingyue Cui, Boyang Li, Shuai Zhao, Kai Huang
  针对安全关键应用的GPU任务DAG调度分析方法。
  GPU调度实时系统DAG 展开摘要
  With the rapid advancement of Artificial Intelligence, the Graphics Processing Unit (GPU) has become increasingly essential across a growing number of safety-critical application domains. Applying a GPU is indispensable for parallel computing; however, the complex data dependencies and resource contention across kernels within a GPU task may unpredictably delay its execution time...
• Horizon-LM: A RAM-Centric Architecture for LLM Training
  Zhengqing Yuan, Lichao Sun, Yanfang Ye
  以RAM为中心的LLM训练架构，解耦模型规模与GPU内存容量。
  LLM训练内存架构分布式系统 展开摘要
  The rapid growth of large language models (LLMs) has outpaced the evolution of single-GPU hardware, making model scale increasingly constrained by memory capacity rather than computation. While modern training systems extend GPU memory through distributed parallelism and offloading across CPU and storage tiers, they fundamentally retain a GPU-centric execution paradigm...
• The Compute ICE-AGE: Invariant Compute Envelope under Addressable Graph Evolution
  Raymond Jay Martin
  BLGC理论的生产级实现，验证确定性语义状态基底的可行性。
  确定性计算图引擎AI架构 展开摘要
  This paper presents empirical results from a production-grade C++ implementation of a deterministic semantic state substrate derived from prior formal work on Bounded Local Generator Classes (Martin, 2026). The system was mathematically specified prior to implementation and realized as a CPU-resident graph engine operating under bounded local state evolution...
• Hardening the OSv Unikernel with Efficient Address Randomization
  Alex Wollman, John Hastings
  为OSv Unikernel引入ASLR风格的安全加固，减少内存布局可预测性。
  UnikernelASLR安全 展开摘要
  Unikernels are single-purpose library operating systems that run the kernel and application in one address space, but often omit security mitigations such as address space layout randomization (ASLR). In OSv, boot, program loading, and thread creation select largely deterministic addresses, leading to near-identical layouts across instances and more repeatable exploitation...
• TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler
  Rong Fu, Yibo Meng, Guangzhen Yao, Jiaxuan Lu, Zeyu Zhang, Zhaolu Kang, Ziming Guo, Jia Yee Tan, Xiaojing Du, Simon James Fong
  基于强化学习的实时调度器，使用Transformer和紧迫性分词器。
  强化学习实时调度Transformer 展开摘要
  Real-time schedulers must reason about tight deadlines under strict compute budgets. We present TempoNet, a reinforcement learning scheduler that pairs a permutation-invariant Transformer with a deep Q-approximation. An Urgency Tokenizer discretizes temporal slack into learnable embeddings, stabilizing value learning and capturing deadline proximity...
• Mitigating Timing-Based Attacks in Real-Time Cyber-Physical Systems
  Arkaprava Sain, Sunandan Adhikary, Soumyajit Dey
  在保证实时性的同时防护时序侧信道攻击，兼顾闭环控制行为。
  实时系统安全侧信道攻击 展开摘要
  Real-time cyber-physical systems depend on deterministic task execution to guarantee safety and correctness. Unfortunately, this determinism can unintentionally expose timing information that enables adversaries to infer task execution patterns and carry out timing-based attacks targeting safety-critical control tasks...
• Why iCloud Fails: The Category Mistake of Cloud Synchronization
  Paul Borrill
  从哲学角度分析云同步的"范畴错误"，揭示FITO假设的根本局限。
  分布式系统云同步理论 展开摘要
  iCloud Drive presents a filesystem interface but implements cloud synchronization semantics that diverge from POSIX in fundamental ways. This divergence is not an implementation bug; it is a Category Mistake -- the same one that pervades distributed computing wherever Forward-In-Time-Only (FITO) assumptions are embedded into protocol design...

技术关键词分布