2026-09论文中文摘要
具有发散任务证据的稳定几何结构用于高效长时程智能体压缩
Stable Geometry with Divergent Task Evidence for Efficient Long-Horizon Agent Compression
arXiv 2609.27332 · 2026-09-24JEV-Star：基于语言模型规划的快速、低成本星际争霸II控制
JEV-Star: Fast, Low-Cost StarCraft II Control with Language-Model Planning
arXiv 2609.27331 · 2026-09-24一种基于采样的概率占用栅格层次信息论压缩方法
A Sample-Based Approach for Hierarchical Information-Theoretic Compression of Probabilistic Occupancy Grids
arXiv 2609.27330 · 2026-09-24随机增广拉格朗日框架：非凸期望约束优化的二阶收敛保证
Stochastic Augmented Lagrangian Framework with Second-Order Convergence Guarantees for Nonconvex Expectation-Constrained Optimization
arXiv 2609.27329 · 2026-09-24并行多保真度期望改进方法用于高效全局优化
Parallel Multi-Fidelity Expected Improvement Method for Efficient Global Optimization
arXiv 2609.27328 · 2026-09-24视觉语言模型能否分析以人为中心的视频？映射模型能力与人机协作工作流
Can Vision-Language Models Analyze Human-Centered Video? Mapping Model Capabilities and Human-AI Collaborative Workflows
arXiv 2609.27327 · 2026-09-24M87中的吸积流是否处于磁停滞状态？
Is accretion flow magnetically arrested in M87?
arXiv 2609.27326 · 2026-09-24椭圆界面问题的一种混合迭代深度Ritz方法
A Hybrid Iterative Deep Ritz Method for Elliptic Interface Problems
arXiv 2609.27325 · 2026-09-24无序轴子泵浦中的迁移率隙鲁棒性与Středa-Hall分离
Mobility-Gap Robustness and Středa-Hall Separation in Disordered Axion Pumping
arXiv 2609.27324 · 2026-09-24DNNsolver：偏振敏感衍射神经网络的精确高效深度学习建模
DNNsolver: Accurate and Efficient Deep Learning Modeling of Polarization-Sensitive Diffractive Neural Networks
arXiv 2609.27323 · 2026-09-24几何提升与紧致连通阿贝尔群中Freiman的$3k-4$定理
Geometric lifting and Freiman's $3k-4$ theorem in compact connected abelian groups
arXiv 2609.27322 · 2026-09-24可验证的隐藏动态玩法：从已求解机制生成智能体强化学习环境
Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms
arXiv 2609.27321 · 2026-09-24任意尺寸下永久型簇的维数
Dimensions of permanental varieties in arbitrary size
arXiv 2609.27320 · 2026-09-24利用行间压缩 LLC 中的解压缩延迟构建隐蔽信道
Exploiting Decompression Latency for Covert Channels in Inter-Line-Compressed LLCs
arXiv 2609.27319 · 2026-09-24打破天气-内容耦合：类型-严重度引导的渐进解耦用于全合一红外图像恢复
Breaking Weather-Content Coupling: Type-Severity Guided Progressive Disentanglement for All-in-One Infrared Restoration
arXiv 2609.27317 · 2026-09-24两组分玻色-爱因斯坦凝聚体的临界曲面
Critical surface for two component Bose-Einstein Condensates
arXiv 2609.27316 · 2026-09-24超越DER：拥挤场景下端到端说话人日志中的说话人计数
Beyond DER: Speaker Counting in Crowded End-to-End Diarization
arXiv 2609.27315 · 2026-09-24强轨道等价类中的熵与复杂性：基于伪Toeplitz子移位
Entropy and complexity in a strong orbit equivalence class via pseudo-Toeplitz subshifts
arXiv 2609.27313 · 2026-09-24将安全转化为能力：通过安全过滤强化学习实现最小可被利用的机器人策略
Turning Safety into Competence: Minimally Exploitable Robot Policies via Safety-Filtered Reinforcement Learning
arXiv 2609.27312 · 2026-09-24多视图融合用于加密C2检测：泄漏受控的评估陷阱测量研究
Multi-View Fusion for Encrypted C2 Detection: A Leakage-Controlled Measurement Study of Evaluation Pitfalls
arXiv 2609.27311 · 2026-09-24希尔伯特模簇中低度完备上同调
Completed cohomology of Hilbert modular varieties below middle degree
arXiv 2609.27310 · 2026-09-24FairTest：多智能体强化学习系统的基于搜索的公平性测试
FairTest: Search-Based Fairness Testing for Multi-Agent Reinforcement Learning Systems
arXiv 2609.27309 · 2026-09-24从自身交互中学习如何行动：面向GUI智能体的在线策略自蒸馏
Learn How to Act from Your Own Interactions: On-Policy Self-Distillation for GUI Agents
arXiv 2609.27307 · 2026-09-24通过演化变分自回归网络的离散扩散模型
Discrete Diffusion Models via Evolving Variational Autoregressive Networks
arXiv 2609.27306 · 2026-09-24van der Pol 方程的振幅严格下界及对称 Liénard 系统的推广
A strict amplitude lower bound for the van der Pol equation and extensions to symmetric Liénard systems
arXiv 2609.27305 · 2026-09-24Live Assistant：学习在真实直播社交流中是否、何时以及向谁提供协助
Live Assistant: Learning Whether, When, and Whom to Assist in Real-World Live Social Streams
arXiv 2609.27303 · 2026-09-24未建模状态与不确定动作结果在智能体扫描隧道显微镜中的应用
Unmodeled states and uncertain action outcomes in agentic scanning tunneling microscopy
arXiv 2609.27302 · 2026-09-24SoK：你找到你所寻求的：重新思考硬件模糊测试中的预言机、引导和输入生成
SoK: You Find What You Seek: Rethinking Oracles, Guidance, and Input Generation in Hardware Fuzzing
arXiv 2609.27300 · 2026-09-24超越权力幻觉：观测性信息系统中的准实验校准
Beyond the Illusion of Power: Calibrating Quasi-Experiments in Observational IS
arXiv 2609.27299 · 2026-09-24StateComp：学习在长时程智能体中何时压缩历史
StateComp: Learning When to Compress History in Long Horizon Agents
arXiv 2609.27298 · 2026-09-24Teach-to-Crash：一种用于碰撞诱发测试场景生成的闭环学生-教师大语言模型框架
Teach-to-Crash: A Closed-Loop Student-Teacher LLM Framework for Collision-Inducing Test Scenario Generation
arXiv 2609.27296 · 2026-09-24每个枢轴尖点融合范畴的图解表示
A diagrammatic presentation for every pivotal pointed fusion category
arXiv 2609.27295 · 2026-09-24KITE：KV不变变压器扩展用于高效智能体LLM扩展
KITE: KV-Invariant Transformer Expansion for Efficient Agentic LLM Scaling
arXiv 2609.27294 · 2026-09-24匹配中的单例可达性与透明访问
Singleton-Attainability and Transparent Access in Matching
arXiv 2609.27293 · 2026-09-24二维聚合物湍流中从弱弹性到弹性激活状态的转变
Transition from a weak-elastic to an elastic-activated regime in two-dimensional polymeric turbulence
arXiv 2609.27292 · 2026-09-24NGN：将神经网络规模学习为可微计数
NGN: Learning Neural Network Size as a Differentiable Count
arXiv 2609.27291 · 2026-09-24基于物理信息神经网络的稀疏观测大气热力预报用于气候感知数字孪生
Sparse-Observation Atmospheric Thermal Forecasting with Physics-Informed Neural Networks for Climate-Aware Digital Twins
arXiv 2609.27290 · 2026-09-24Ruby-ASR：保留证据的联合正字法与词汇读音识别监督
Ruby-ASR: Evidence-Preserving Supervision for Joint Orthographic and Lexical-Reading Recognition
arXiv 2609.27289 · 2026-09-24PotARCin：抽象推理任务中技能习得的多维评估
PotARCin: Multi-Dimensional Evaluation of Skill Acquisition in Abstract Reasoning Tasks
arXiv 2609.27288 · 2026-09-24长时程智能体中的记忆控制信号在行动之前出现
Memory Control Signals Emerge Before Action in Long Horizon Agents
arXiv 2609.27286 · 2026-09-24抽象机制中的节奏均衡
Pacing Equilibria in Abstract Mechanisms
arXiv 2609.27285 · 2026-09-24Hunyuan-A13B 技术报告
Hunyuan-A13B Technical Report
arXiv 2609.27284 · 2026-09-24多波COVID-19纽约市发病率的自适应行为SIR模型的贝叶斯校准
Bayesian calibration of adaptive-behavior SIR models for multi-wave COVID-19 incidence in New York City
arXiv 2609.27283 · 2026-09-24对FRB族群的多望远镜拟合，允许FRB重复
A multi-telescope fit to the FRB population, allowing for FRB repetition
arXiv 2609.27282 · 2026-09-24多任务回归与成对融合
Multitask Regression with Pairwise Fusion
arXiv 2609.27280 · 2026-09-24稀缺数据的谱连通性先验图学习
Graph Learning with Spectral Connectivity Priors for Scarce Data
arXiv 2609.27278 · 2026-09-24TimeEvo：时间序列智能体的失败驱动自我进化
TimeEvo: Failure-Driven Self-Evolution of a Time Series Agent
arXiv 2609.27277 · 2026-09-24DRSR：学习集合级删除风险以实现高效长时程智能体
DRSR: Learning Set-Level Deletion Risk for Efficient Long-Horizon Agents
arXiv 2609.27276 · 2026-09-24BranchDrive：用于动作条件驾驶预测的分支结构数据集
BranchDrive: A Branch-Structured Dataset for Action-Conditioned Driving Prediction
arXiv 2609.27275 · 2026-09-24高动态范围视频重建：基于单曝光原始序列
High Dynamic Range Video Reconstruction from Single-Exposure Raw Sequences
arXiv 2609.27274 · 2026-09-24
