2026-08论文中文摘要
SPECTRA：通过谱变换编码将KV缓存推向2比特极限之外
SPECTRA: Pushing the KV Cache Beyond the 2-Bit Cliff via Spectral Transform Coding
arXiv 2608.07915 · 2026-08-11基准污染何时可检测？信息极限与功率校准审计
When Is Benchmark Contamination Detectable? Information Limits and Power-Calibrated Audits
arXiv 2608.07914 · 2026-08-11面向联邦检索增强生成的私有任意时刻选择性风险认证：保证与经验极限
Private Anytime Selective-Risk Certification for Federated Retrieval-Augmented Generation: Guarantees and Empirical Limits
arXiv 2608.07913 · 2026-08-11霍施塔特$ab-1$问题的正下密度
Positive Lower Density for Hofstadter's $ab-1$ Problem
arXiv 2608.07910 · 2026-08-11史瓦西时空的映射：从克鲁斯卡（Kruskal）到菱形（Diamond）及黄金表示
Mapping Schwarzschild Spacetime: From Kruskal to Diamond and Golden Representations
arXiv 2608.07909 · 2026-08-11计算神经网络控制系统的最大受控不变集
Computing the Maximal Controlled Invariant Set for Neural Network Control Systems
arXiv 2608.07908 · 2026-08-11弯曲时空中的四费米子凝聚：一种泛函方法
Four-Fermion Condensates in Curved Spacetimes: A Functional Approach
arXiv 2608.07907 · 2026-08-11无共线三点的长格路
Long Lattice Paths with No Three Collinear Vertices
arXiv 2608.07906 · 2026-08-11GraphThink：用于长视距具身任务规划的图增强大语言模型思维
GraphThink: Graph-Enhanced LLM Thinking for Long-Horizon Embodied Task Planning
arXiv 2608.07905 · 2026-08-11TopoBudget：用于可复用社区分析的持久连通性保持Web图稀疏化
TopoBudget: Persistent-Connectivity-Preserving Web Graph Sparsification for Reusable Community Analytics
arXiv 2608.07903 · 2026-08-11超越“我无法对此提供帮助”：儿童安全专家如何评估AI聊天机器人的安全性
Beyond "I Can't Help With That": How Child Safety Experts Evaluate AI Chatbot Safety
arXiv 2608.07902 · 2026-08-11基于DVB的波束跳变低轨卫星网络中需求预测对时延和抖动的影响
The Impact of Demand Forecasting on Delay and Jitter in DVB-Based Beam-Hopping LEO Networks
arXiv 2608.07901 · 2026-08-11利用可印刷六方氮化硼离子凝胶实现外延Bi (111)薄膜的门可调电子性质
Gate-tunable electronic properties of epitaxial Bi (111) films using a printable hexagonal boron nitride ionogel
arXiv 2608.07900 · 2026-08-11TelemetrySuffBench：智能体遥测是否足以用于故障起源诊断？
TelemetrySuffBench: Is Agent Telemetry Sufficient for Failure-Origin Diagnosis?
arXiv 2608.07899 · 2026-08-11用于上行非正交传输的语义辅助迭代解码
Semantic-Aided Iterative Decoding for Uplink Non-Orthogonal Transmission
arXiv 2608.07897 · 2026-08-11用于性能-安全权衡的混合ASIC-FPAA架构
Hybrid ASIC-FPAA Fabric for Performance Security Trade-off
arXiv 2608.07896 · 2026-08-11多模态机器人演示中指令-轨迹不匹配的审计
Auditing Instruction-Trajectory Mismatches in Multimodal Robot Demonstrations
arXiv 2608.07895 · 2026-08-11基于大语言模型嵌入的程序分析与优化
LLM-Based Embeddings for Program Analysis and Optimization
arXiv 2608.07894 · 2026-08-11能力路由守卫：防御大型推理模型免受以推理为中心的越狱攻击
Capability-Routed Guard: Defending Large Reasoning Models Against Reasoning-Centric Jailbreaks
arXiv 2608.07892 · 2026-08-11立法证词中的自我介绍检测
Detection of Self-Introductions in Legislative Testimony
arXiv 2608.07891 · 2026-08-11从年度吞吐量到船舶班期：一种用于中转枢纽仿真的随机生成器
From Annual Throughput to Vessel Schedules: A Stochastic Generator for Transshipment Hub Simulation
arXiv 2608.07889 · 2026-08-11改进型^2环境分类器
Refined^2 Environment Classifiers
arXiv 2608.07888 · 2026-08-11混合刚度仿人指尖扩大硬币抓取的工作范围
A Mixed-Stiffness Anthropomimetic Fingertip Broadens the Operating Range for Coin Grasping
arXiv 2608.07887 · 2026-08-11视觉-语言 Grounding 作为双向概念对应
Vision-Language Grounding as Bidirectional Concept Correspondence
arXiv 2608.07886 · 2026-08-11广泛推理，而非深度推理：将推理溢价分摊到提炼技能中
Reason Wide, Not Deep: Amortizing the Reasoning Premium into Distilled Skills
arXiv 2608.07885 · 2026-08-11高能量晶面的本征二维贵金属的亚稳态极限：Bell-Evans-Polanyi原理的主导作用
Metastability limit of pristine 2D noble metals with high-energy facet: dominance of the Bell-Evans-Polanyi principle
arXiv 2608.07884 · 2026-08-11随机环境中的随机伯努利测度
Random Bernoulli measures in a random environment
arXiv 2608.07883 · 2026-08-11GRACE：用于可扩展混合数据聚类的大语言模型（LLM）基础语义度量空间
GRACE: LLM-Grounded Semantic Metric Spaces for Scalable Mixed-Data Clustering
arXiv 2608.07881 · 2026-08-11通过Ringel–Hall李代数构造GIM和椭圆李代数
GIM and Elliptic Lie algebras via Ringel--Hall Lie algebras
arXiv 2608.07877 · 2026-08-11SurgLAT：面向深度感知的机器人腹腔镜控制的手术隐式注意力跟踪
SurgLAT: Surgical Latent Attention Tracking for Depth-Aware Robotic Laparoscope Control
arXiv 2608.07876 · 2026-08-11从函数重整化群看有效引力常数的标度依赖性
Scale dependence of the effective gravitational constant from functional renormalization group
arXiv 2608.07875 · 2026-08-11回到未来：用于电子表格创建基准测试的工作簿时光机
Back to the Future: A workbook time machine for spread sheet creation benchmarks
arXiv 2608.07873 · 2026-08-11高频下振动的夹紧板类似于振动的膜
A Vibrating Clamped Plate Resembles a Vibrating Membrane at High Frequency
arXiv 2608.07872 · 2026-08-11众包地理收入：使用谷歌地图兴趣点作为巴西圣保罗市亚市政级收入估算的高频代理
Crowd-Sourced Geographies of Income: Using Google Maps Points of Interest as High-Frequency Proxies for Sub-Municipal Income Estimation in Sao Paulo, Brazil
arXiv 2608.07871 · 2026-08-11V-Simba：释放强化学习在视觉连续控制中的架构潜力
V-Simba: Unleashing the Architectural Potential of RL in Visual Continuous Control
arXiv 2608.07870 · 2026-08-11因果菱形热化诱导N部分量子系统中的非可分性
Causal-diamond thermalization induces nonseparability in N-partite quantum systems
arXiv 2608.07869 · 2026-08-11不等强度光源的亚瑞利成像：近量子极限的多参量估计
Sub-Rayleigh Imaging of Unequal-Intensity Sources: Near-Quantum-Limit Multiparameter Estimation
arXiv 2608.07868 · 2026-08-11CONFER：面向多模态情感识别中规则校准弱监督的冲突感知证据协商框架
CONFER: Conflict-Aware Evidence Negotiation for Regime-Calibrated Weak Supervision in Multimodal Emotion Recognition
arXiv 2608.07867 · 2026-08-11IDH野生型胶质母细胞瘤中MAPK通路活性与血红素生物合成基因表达：一项纯度校正的发现-验证分析
MAPK Pathway Activity and Heme Biosynthesis Gene Expression in IDH-Wildtype Glioblastoma: A Purity-Adjusted, Discovery-Validation Analysis
arXiv 2608.07865 · 2026-08-11LHSDet：基于视觉问答的高分辨率AI生成图像检测方法
LHSDet: High-Resolution AI-Generated Image Detection via Visual Question Answering
arXiv 2608.07863 · 2026-08-11SurakshaEval：面向多语言大语言模型的印度语安全基准
SurakshaEval: An Indic Safety Benchmark for Multilingual LLMs
arXiv 2608.07862 · 2026-08-11回答我的问题需要多少成本？基于云VLM的VQA系统基准测试
How Much Does It Cost to Answer My Question? Benchmarking Cloud VLM-based VQA Systems
arXiv 2608.07861 · 2026-08-11独立性并非总是可一致检验的
Independence Is Not Always Consistently Testable
arXiv 2608.07860 · 2026-08-11自适应KappaSharp：用于偏好贝叶斯优化的条件数塑形
Adaptive KappaSharp: Condition-Number Shaping for Preferential Bayesian Optimization
arXiv 2608.07859 · 2026-08-11特斯拉级磁场中HRPPD的性能
Performance of an HRPPD in Tesla-scale magnetic fields
arXiv 2608.07858 · 2026-08-11将CT基础模型蒸馏为可编辑概念瓶颈用于肺结节恶性程度预测
Distilling CT Foundation Models into Editable Concept Bottlenecks for Lung Nodule Malignancy Prediction
arXiv 2608.07857 · 2026-08-11常压下锰的自旋选择性非范德华电子化物特性
Spin selective non-van der Waal electride nature in manganese under ambient pressure
arXiv 2608.07856 · 2026-08-11CommitKV：面向多轮智能体的、基于提交转换的生命周期感知KV缓存压缩
CommitKV: Lifecycle-Aware KV Cache Compression via Commit Transitions for Multi-Turn Agents
arXiv 2608.07855 · 2026-08-11一致可测试性蕴含非对称可测试性
Uniform Testability Implies Asymmetric Testability
arXiv 2608.07854 · 2026-08-11关于古谢尔-穆凯三维簇上的布里奇兰德模空间几何的注记
A note on geometry of Bridgeland Moduli spaces on Gushel-Mukai threefolds
arXiv 2608.07853 · 2026-08-11
