2026-10论文中文摘要
$(sn,n)$-Dyck 路径的面积渐近性
The Area Asymptotics of $(sn,n)$-Dyck Paths
arXiv 2610.07671 · 2026-10-07基于粗粒度标签的雷达智能检测
Radar Intelligent Detection with Coarse-Grained Labels
arXiv 2610.07670 · 2026-10-07评估葡萄栽培田间研究中的人机协作工作流
Evaluating human-AI workflows for field research in viticulture
arXiv 2610.07669 · 2026-10-07CACHEFORGE：面向性能与硬件效率的LLM引导端到端生成式缓存替换策略
CACHEFORGE: LLM-Guided End-to-End Generative Cache Replacement Policy for Performance and Hardware Efficiency
arXiv 2610.07668 · 2026-10-07硼中子俘获治疗中粗粒化效应预测的不可行框架：结构可辨识性极限与微观覆盖
A No-Go Framework for Coarse-Grained Effect Prediction in Boron Neutron Capture Therapy: Structural Identifiability Limits and Microscopic Coverage
arXiv 2610.07667 · 2026-10-07SENSE：状态感知的情感导航故事引擎
SENSE: State-aware Emotion Navigation Storytelling Engine
arXiv 2610.07666 · 2026-10-07机械化弱隔离级别的事务性异常模式
Mechanizing Transactional Anomalous Patterns for Weak Isolation Levels
arXiv 2610.07665 · 2026-10-07面向连续可压缩流的矢量离散统一气体动理学格式
Vectorial discrete unified gas kinetic scheme for continuum compressible flows
arXiv 2610.07664 · 2026-10-07用户行为模拟的联合工作流与提示优化
Joint Workflow and Prompt Optimization for User Behavior Simulation
arXiv 2610.07663 · 2026-10-07MS-ECG-FM：利用多源对比学习实现更通用的心电图基础模型以用于健康监测
MS-ECG-FM: Towards a More Universal Electrocardiogram Foundation Model for Health Monitoring using Multi-source Contrastive Learning
arXiv 2610.07662 · 2026-10-07大规模激活门控通道在大语言模型中的研究
Massive Activation Gating Channel in Large Language Models
arXiv 2610.07661 · 2026-10-07DLoop: 循环推测解码
DLoop: Looped Speculative Decoding
arXiv 2610.07659 · 2026-10-07基于压力的耦合广义合成迭代格式用于稀薄气体流动模拟
A pressure-based coupled general synthetic iterative scheme for rarefied gas flow simulation
arXiv 2610.07658 · 2026-10-07规则止步之处，法官登场：多智能体系统安全中的判断边界测量
Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security
arXiv 2610.07657 · 2026-10-07单层镍酸盐La1.4Sr0.6NiO4中的Luttinger液体行为
Luttinger Liquid Behavior in a Single-Layer Nickelate La1.4Sr0.6NiO4
arXiv 2610.07656 · 2026-10-07均匀离散扩散模型在估计具有小有效支持大小的分布时是极小极大最优的
Uniform Discrete Diffusion Models are Minimax Optimal for Estimating Distributions with Small Effective Support Size
arXiv 2610.07655 · 2026-10-07用于安全的在线策略蒸馏是否存在后门风险？
Does On-Policy Distillation for Safety Pose Backdoor Risks?
arXiv 2610.07654 · 2026-10-07辐射传输方程的低秩SI-DSA与GMRES-DSA方法及自适应精度控制
Low-rank SI-DSA and GMRES-DSA for Radiative Transfer Equation with Adaptive Accuracy Control
arXiv 2610.07653 · 2026-10-07SMART：通过大规模合成预训练实现关节物体的零样本仿真到现实操作
SMART: Zero-Shot Sim-to-Real Articulated Object Manipulation via Large-Scale Synthetic Pretraining
arXiv 2610.07652 · 2026-10-07表面等离激元波激发原子链中的亚辐射态
Excitation of subradiant states in atom chains by surface plasmon waves
arXiv 2610.07651 · 2026-10-07硅语言：面向异构机器人的机器人原生知识交换框架
Silicon Language: A Robot-Native Knowledge Exchange Framework for Heterogeneous Robots
arXiv 2610.07650 · 2026-10-07OntoPlan：一种基于本体论的场景表示与智能体框架，用于可扩展的机器人任务规划
OntoPlan: An Ontology-Grounded Scene Representation and Agentic Framework for Scalable Robot Task Planning
arXiv 2610.07649 · 2026-10-07响亮而清晰：动态激活引导提升嘈杂环境下的语音可懂度
Loud and Clear: Dynamic Activation Steering for Improving Speech Intelligibility in Noisy Environments
arXiv 2610.07647 · 2026-10-07匹配对象还是关系？追踪VLM内部的抽象推理
Matching Object or Relation? Tracing Abstract Reasoning Inside VLMs
arXiv 2610.07646 · 2026-10-07SkillPoison：通过成功经验进行渐进式技能投毒
SkillPoison: Progressive Skill Poisoning via Successful Experiences
arXiv 2610.07645 · 2026-10-07从ASIC到机群：构建和运营超大规模数据中心NIC的经验教训
From ASIC to Fleet: Lessons from Building and Operating a Hyperscaler NIC
arXiv 2610.07644 · 2026-10-07KV缓存驱逐的蒙特卡洛估计
Monte Carlo Estimation for KV Cache Eviction
arXiv 2610.07643 · 2026-10-07冷原子里德伯阵列与时空复用技术用于高灵敏度无线通信
Cold-Atom Rydberg Array with Spatiotemporal Multiplexing for High-Sensitivity Wireless Communications
arXiv 2610.07642 · 2026-10-07通过推测执行隐藏端侧级联语音代理中的工具延迟
Hiding Tool Latency in On-Device Cascaded Voice Agent through Speculative Execution
arXiv 2610.07641 · 2026-10-07迈向可解释国际象棋战术的自动综合
Towards the Automatic Synthesis of Interpretable Chess Tactics
arXiv 2610.07640 · 2026-10-07HarnessSecurity-Bench：安全机制真的能保护编码智能体框架吗？
HarnessSecurity-Bench: Do Security Mechanisms Really Protect Coding Agent Harnesses?
arXiv 2610.07639 · 2026-10-07学习复杂游戏策略的可解释表示
Learning Explainable Representations of Complex Game-playing Strategies
arXiv 2610.07638 · 2026-10-07逐项独立同分布重尾数据上经验风险最小化的渐近分析
Asymptotic Analysis of Empirical Risk Minimization on Entry-wise i.i.d. Heavy-Tailed Data
arXiv 2610.07637 · 2026-10-07哈勃空间望远镜对活跃小行星(101955) Bennu的观测
Hubble Space Telescope Observations of Active Asteroid (101955) Bennu
arXiv 2610.07636 · 2026-10-07CISB-Bench：一个可审计的编译器引入安全漏洞的源代码-IR数据集
CISB-Bench: An Auditable Source--IR Dataset of Compiler-Introduced Security Bugs
arXiv 2610.07635 · 2026-10-07测量 Twitter 和 Reddit 档案中的气候反弹：词汇定义、记录响应与参与者更替
Measuring climate backlash in Twitter and Reddit archives: Lexical definitions, recorded responses and participant turnover
arXiv 2610.07634 · 2026-10-07超表面吸收体非平衡热提取的直接可视化
Direct visualization of non-equilibrium heat extraction by a metasurface absorber
arXiv 2610.07633 · 2026-10-07第一海森堡群中的尖锐等周不等式
The Sharp Isoperimetric Inequality in the First Heisenberg Group
arXiv 2610.07632 · 2026-10-07有限几何障碍与Buchstaber不变量的多面体分离
A Finite-Geometric Obstruction and a Polytopal Separation of Buchstaber Invariants
arXiv 2610.07631 · 2026-10-07用协作选择数据测量谈判特质：一种两阶段潜变量模型
Measuring Negotiation Traits with Collaborative Choice Data: A Two-Stage Latent Variable Model
arXiv 2610.07630 · 2026-10-07超越任务奖励：一种用于评估具身依赖能力的控制器限制协议
Beyond Task Reward: A Controller-Restriction Protocol for Evaluating Embodiment-Dependent Competence
arXiv 2610.07629 · 2026-10-07互补的监督与自监督表示用于图分布外学习
Complementary Supervised and Self-Supervised Representations for Out-of-Distribution Graph Learning
arXiv 2610.07628 · 2026-10-07学习超越理论：在初始假设空间之外的实验发现
Learning to Outgrow a Theory: Experimental Discovery Beyond the Initial Hypothesis Space
arXiv 2610.07627 · 2026-10-07一种利用辅助词重音建模和损失优化的新型句子重音检测框架
A Novel Sentence Stress Detection Framework Leveraging Auxiliary Word-Stress Modeling and Loss Optimization
arXiv 2610.07626 · 2026-10-07无状态语言智能体：扩展长时程自动化研究
Stateless Language Agents: Scaling Long-Horizon Automated Research
arXiv 2610.07625 · 2026-10-07辐射压驱动类星体宽吸收线风，但未能驱动星系反馈
Radiation pressure powers quasar broad absorption line winds but fails to drive galaxy feedback
arXiv 2610.07624 · 2026-10-07超越 $T^{2/3}$ 的序贯校准显式渐近界
Explicit Asymptotic Bounds for Sequential Calibration Beyond $T^{2/3}$
arXiv 2610.07623 · 2026-10-07DBRAG：面向复杂数据库查询的多表检索增强生成
DBRAG: Multi-Table Retrieval-Augmented Generation for Complex Database Queries
arXiv 2610.07622 · 2026-10-07TacZero：使用带触觉反馈的通用视觉语言模型实现免训练插销任务
TacZero: Training-Free Peg Insertion Using a General-Purpose Vision-Language Model with Tactile Feedback
arXiv 2610.07621 · 2026-10-07探索，然后提交：利用语言模型进行测量高效的科学定律发现
Explore, Then Commit: Measurement-Efficient Scientific Law Discovery with Language Models
arXiv 2610.07620 · 2026-10-07
