2026-09论文中文摘要
Rubric-Calibrated Preferences：基于项目反应理论的大语言模型判断跨查询校准
Rubric-Calibrated Preferences: Cross-Query Calibration of LLM Judgments via Item Response Theory
arXiv 2609.35739 · 2026-09-28作为研究者的AI智能体集群：进展、挑战与开放问题
AI Agent Swarms as Researchers: Progress, Challenges, and Open Questions
arXiv 2609.35719 · 2026-09-28更高效的并行$(Δ+1)$-边着色
More Efficient Parallel $(Δ+1)$-Edge Coloring
arXiv 2609.35693 · 2026-09-28具有常数最大最小份额保证的期望真实机制
Truthful-in-Expectation Mechanism with Constant Maximin-Share Guarantee
arXiv 2609.35670 · 2026-09-28高代价碰撞下的动态唤醒
Dynamic Wakeup under Costly Collisions
arXiv 2609.35662 · 2026-09-28Tracekit：面向自主编码智能体的防篡改意图-推理-行动审计系统
Tracekit: Tamper-Evident Intent-Reasoning-Action Auditing for Autonomous Coding Agents
arXiv 2609.35659 · 2026-09-28应用于毫米波波束管理的网络数字孪生的能量驱动评估
Energy-Driven Evaluation of Network Digital Twinning Applied to mmWave Beam Management
arXiv 2609.35644 · 2026-09-28资源与响应性：大规模移动目标防御控制平面下SDN控制器运行时基准测试
Resource versus Responsiveness: Benchmarking SDN Controller Runtimes for a Moving-Target-Defense Control Plane at Scale
arXiv 2609.35585 · 2026-09-28软介电复合材料中嵌入梁网络的混合维机电耦合
Mixed-Dimensional Electromechanical Coupling of Embedded Beam Networks in Soft Dielectric Composites
arXiv 2609.35584 · 2026-09-28竞争随机顺序相关k聚类
Competitive Random-Order Correlation k-Clustering
arXiv 2609.35555 · 2026-09-28鱼群中的位置选择与鲁棒集群行为：生物混合实验与建模
Positional choice and robust collective behavior in fish schools: biohybrid experiments and modeling
arXiv 2609.35554 · 2026-09-28INTCC：一种交互式机密计算框架
INTCC: A Framework for Interactive Confidential Computing
arXiv 2609.35552 · 2026-09-28Argus：智能体驱动、参考校准、树引导的系统软件级瓶颈定位
Argus: Agentic, Reference-Calibrated, Tree-Guided, System-Software-Level Bottleneck Localization
arXiv 2609.35508 · 2026-09-28通过系统感知图边界学习检测智能电网中的虚假数据注入与不稳定运行
Detecting False Data Injection and Unstable Operation in Smart Grid via System-Aware Graph Boundary Learning
arXiv 2609.35506 · 2026-09-28超越标量探针：利用ReLU网络中的向量值输出进行特征提取
Beyond Scalar Probes: Exploiting Vector-Valued Outputs in ReLU Networks For Signature Extraction
arXiv 2609.35487 · 2026-09-28Hermite Brings a Laptop：分析Frieze-Jerrum舍入可改进聚类问题的近似算法
Hermite Brings a Laptop: Analyzing Frieze-Jerrum Rounding Yields Improved Approximations for Clustering Problems
arXiv 2609.35453 · 2026-09-28LLM辅助的密码协议自动安全性证明：我们还有多远？
LLM-Assisted Automatic Security Proofs for Cryptographic Protocols: How Far Are We?
arXiv 2609.35434 · 2026-09-28生成式检索器能在学习语义ID的同时不遗忘语言生成能力吗？
Can Generative Retrievers Learn Semantic IDs Without Forgetting How to Speak?
arXiv 2609.35430 · 2026-09-28基于AI的漏洞评估能力与网络攻击图分析
AI-Based Vulnerability Assessment Capability and Cyber Attack Graph Analysis
arXiv 2609.35414 · 2026-09-28重拾社交媒体中的社交属性
Reclaiming the social in social media
arXiv 2609.35413 · 2026-09-28硬阈值下的二元$k$-中心及其在委托投票中的应用
Binary $k$-Center under a Hard Threshold with Applications to Delegated Voting
arXiv 2609.35386 · 2026-09-28注意力拒绝服务（Attentional DoS）：X平台上的重复转发、集体注意力与信息扩散
Attentional DoS: Repeat Reposting, Collective Attention, and Information Diffusion on X
arXiv 2609.35385 · 2026-09-28可补充预算下的在线资源分配
Online Resource Allocation with Replenishable Budgets
arXiv 2609.35384 · 2026-09-28通过自动化可扩展云原生编排实现QKD-PQC混合网络仿真
Hybrid QKD-PQC Network Emulation through Automated and Scalable Cloud-Native Orchestration
arXiv 2609.35358 · 2026-09-28在线问题的纯尾约束
Pure Tail Constraints for Online Problems
arXiv 2609.35337 · 2026-09-28$\text{Almost-}\bigoplus\text{P} = \text{BP·}\bigoplus\text{P}$以及Toda定理的随机预言机证明
$\mathrm{Almost}\text{-}\oplus\mathrm{P} = \mathrm{BP}\cdot\oplus\mathrm{P}$ and a Random-Oracle Proof of Toda's Theorem
arXiv 2609.35326 · 2026-09-28NATS JetStream中配置引发的投递故障：检测与修复
Configuration-Induced Delivery Failures in NATS JetStream: Detection and Remediation
arXiv 2609.35310 · 2026-09-28信号还是噪声？多模态GraphRAG中的模态贡献与协作
Signal or Noise? Modality Contribution and Cooperation in Multimodal GraphRAG
arXiv 2609.35304 · 2026-09-28国际出口冷链中产热奶酪托盘的实验研究
Experimental Investigation of Heat-generating Cheese Pallets During an International Export Cold Chain
arXiv 2609.35301 · 2026-09-28作为安全资源的持续参与：有界参与通道
Sustained Participation as a Security Resource: The Bounded Participation Channel
arXiv 2609.35300 · 2026-09-28大语言模型时代下的模型还有未来吗？
Is there a future for models in the LLM era?
arXiv 2609.35281 · 2026-09-28针对杂务的期望可信MMS分配
Truthful-in-Expectation MMS Allocations for Chores
arXiv 2609.35280 · 2026-09-28Weaver：一种面向AI-RAN计算共享且支持基础模型训练的系统
Weaver: A System for AI-RAN Compute Sharing with Foundation Model Training
arXiv 2609.35276 · 2026-09-28面向软件交付决策关卡的智能体安全审计器持续保证
Continuous Assurance of Agentic Security Auditors for Software Delivery Decision Gates
arXiv 2609.35266 · 2026-09-28使用商用硬件与开源软件实现数据二极管（Data Diodes）
Implementing Data Diodes Using Commodity Hardware and Open Source Software
arXiv 2609.35256 · 2026-09-28MEGATRON：一款采用28nm工艺的模拟PCM CiM/数字片上系统，用于边缘生成式AI，能效达57.5 TOPS/W，参数密度达1.52 Mparam/mm²
MEGATRON: a 28nm Analog PCM CiM/Digital System-on-Chip for Edge GenAI at 57.5 TOPS/W and 1.52 Mparam/mm${}^2$
arXiv 2609.35254 · 2026-09-28带离群点的加权超度量嵌入的5/2近似算法
A $5/2$-Approximation for Weighted Ultrametric Embedding with Outliers
arXiv 2609.35235 · 2026-09-28Poster: 迈向ProofWeave：面向持续智能体安全保障的隐私最小化、完整性锚定证据平面
Poster: Towards ProofWeave: A Privacy-Minimised, Integrity-Anchored Evidence Plane for Continuous Agentic Assurance
arXiv 2609.35234 · 2026-09-28面向高效且高性价比LLM社会调查模拟的自适应资源分配
Adaptive Resource Allocation for Effective and Efficient LLM Social Survey Simulation
arXiv 2609.35216 · 2026-09-28OT-PCA：基于离线模板的HQC密钥恢复明文检查预言机侧信道攻击新方法
OT-PCA: New Key-Recovery Plaintext-Checking Oracle Based Side-Channel Attacks on HQC with Offline Templates
arXiv 2609.35205 · 2026-09-28LLM编码智能体中的轨迹级安全债务
Trajectory-Level Security Debt in LLM Coding Agents
arXiv 2609.35199 · 2026-09-28随机分配的无嫉妒分解：解决四主体情形及后续拓展
Envy-Free Decompositions of Random Assignments: Settling Four Agents, and What Lies Beyond
arXiv 2609.35192 · 2026-09-28Solana 区块与交易分析
Analyzing Solana's Blocks and Transactions
arXiv 2609.35171 · 2026-09-28生成式AI时代可视化仪表板的未来
The Future of Visualization Dashboards in the Age of Generative AI
arXiv 2609.35170 · 2026-09-28LENS：检索增强生成中集合级投毒的整体效果弱于各部分之和
LENS: The Sum Is Worse Than the Parts for Set-Level Poisoning in Retrieval-Augmented Generation
arXiv 2609.35155 · 2026-09-28AReaL-TIK：通过优化中间表示（IR）实现统一强化学习内核的有状态智能体优化
AReaL-TIK: Stateful Agentic Optimization of Unified RL Kernels through an Optimization IR
arXiv 2609.35140 · 2026-09-28通过对Dijkstra算法的平滑分析求解低拉伸生成树
Low-Stretch Spanning Trees via Smoothed Analysis of Dijkstra's Algorithm
arXiv 2609.35136 · 2026-09-28RenderRank：利用压缩视觉令牌学习文本重排序
RenderRank: Learning to Rerank Text with Compressed Visual Tokens
arXiv 2609.35069 · 2026-09-28高效的基于TCitH的SLH-DSA替代方案：Mirath的跨层ASIC设计
Efficient TCitH-Based Alternatives to SLH-DSA: Cross-Layer ASIC Design of Mirath
arXiv 2609.35053 · 2026-09-28基于全局列表式学习与渐进双加权的推荐系统流行度偏差缓解方法
Mitigating Popularity Bias in Recommendation with Global Listwise Learning and Progressive Bi-Weighting
arXiv 2609.35041 · 2026-09-28
