2026-09论文中文摘要
BaZr0.2Ti0.8O3陶瓷的热波热释电探测
Pyroelectric thermo-wave probing of BaZr0.2Ti0.8O3 ceramics
arXiv 2609.35667 · 2026-09-30大型语言模型激活中的语义搜索特征
Signatures of semantic search in the activations of large language models
arXiv 2609.35599 · 2026-09-30F4R：面向持续机器人自我改进的失败驱动识别、重建、精炼与重新部署
F4R: Failure-Driven Recognition, Reconstruction, Refinement, and Redeployment for Continual Robot Self-Improvement
arXiv 2609.35575 · 2026-09-30RareDx：面向罕见病诊断的受控知识整合与图基策略优化
RareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis
arXiv 2609.35549 · 2026-09-30重构SIXTE模拟器：迈向更模块化的代码库
Refactoring the SIXTE simulator: Towards a more modular code base
arXiv 2609.35494 · 2026-09-30非CM椭圆曲线在$\Q$上的$7$-adic Galois像的完全分类
Complete classification of $7$-adic Galois images for non-CM elliptic curves over $\Q$
arXiv 2609.35460 · 2026-09-30NeuronSifter：中枢神经系统微环境中的干预规划
NeuronSifter: Intervention Planning in CNS Microenvironments
arXiv 2609.35445 · 2026-09-30只需初始化：面向大规模路由优化的免训练初始化组件
Just Initialize: A Training-Free Initialization Component for Large-Scale Routing Optimization
arXiv 2609.35443 · 2026-09-30构建黎曼神经网络变换层
Building Transformation Layers for Riemannian Neural Networks
arXiv 2609.35436 · 2026-09-30带光滑外力的Navier-Stokes方程的有限时间爆破，第一部分：具有可容许应力与平坦余项的自相似解的构造
Finite-Time Blowup for Navier-Stokes with Smooth Forcing, Part I: Construction of Self-Similar Solutions with Admissible Stress and Flat Remainder
arXiv 2609.35406 · 2026-09-30Jordan 归纳法对于 $\mathrm{GL}_n(\mathcal{O})$
Jordan induction for $\mathrm{GL}_n(\mathcal{O})$
arXiv 2609.35393 · 2026-09-30MCP 错误信息面向开发者编写，却对最强智能体伤害最大
MCP Error Messages Written for Developers Hurt the Most Capable Agents Most
arXiv 2609.35381 · 2026-09-30谁驱动系统？从轨迹数据对平均场粒子系统进行分类
Who Drives the System? Classifying Mean-Field Particle Systems from Trajectory Data
arXiv 2609.35374 · 2026-09-30$λ$-JEPA：用于自监督学习的谱抗坍缩正则化
$λ$-JEPA Spectral Anti-Collapse Regularization for Self-Supervised Learning
arXiv 2609.35288 · 2026-09-30超越选择：面向极端视觉令牌压缩的令牌参数化
Beyond Selection: Token Parameterization for Extreme Visual Token Compression
arXiv 2609.35232 · 2026-09-30生成式机器人策略的零样本反应式避障
Zero-Shot Reactive Obstacle Avoidance for Generative Robot Policies
arXiv 2609.35231 · 2026-09-30面向文化适配的中文语言智能体：中德跨文化互动中非语言行为的绿野仙踪研究
Toward a Culturally Adapted Chinese Language Agent: A Wizard-of-Oz Study of Nonverbal Behavior in Chinese-German Intercultural Interaction
arXiv 2609.35150 · 2026-09-30水及其分子离子直至H$_2$O$^{3+}$的势能面
Potential-energy surfaces of water and its molecular ions up to H$_2$O$^{3+}$
arXiv 2609.35144 · 2026-09-30FlexiWorld：通过跨多时间尺度的灵活动作块进行学习与规划
FlexiWorld: Learning and Planning via Flexible Action Chunks Across Multiple Time Scales
arXiv 2609.35138 · 2026-09-30Landau-de Gennes 对 Oseen-Frank 极限的修正：锚定诱导的倾斜模式
Landau-de Gennes corrections to the Oseen-Frank limit: Anchoring-induced tilt modes
arXiv 2609.35127 · 2026-09-30THEIA：面向布局视觉-语言分析的多模态数据集与基准
THEIA: A Multimodal Dataset and Benchmark for Vision-Language Analysis of Layout
arXiv 2609.35035 · 2026-09-30什么使世界动作模型具有泛化能力？测试时未来建模的实证研究
What Makes World Action Models Generalize? An Empirical Study of Test-Time Future Modeling
arXiv 2609.34981 · 2026-09-30医学时间序列扩散的循环平稳相位条件化
Cyclostationary Phase Conditioning for Medical Time Series Diffusion
arXiv 2609.34965 · 2026-09-30ALICE：上下文内、零样本互信息估计
ALICE: In-context, Zero-shot, Mutual Information Estimation
arXiv 2609.34962 · 2026-09-30光锥夸克模型中零斜度海夸克的Twist-4 GTMDs
Twist-4 GTMDs of sea quarks at zero skewness in the light-cone quark model
arXiv 2609.34937 · 2026-09-30弱压力凸起中流动不稳定性促进星子形成
Planetesimal formation facilitated by streaming instability in weak pressure bumps
arXiv 2609.34932 · 2026-09-30不要丢弃尾部：通过动作升级加速策略
Don't Throw Away the Tail: Action Upcycling for Policy Acceleration
arXiv 2609.34911 · 2026-09-30LVMT：用于长时视频分割的视频掩码Transformer
LVMT: Video Mask Transformer for Long-term Video Segmentation
arXiv 2609.34895 · 2026-09-30一次读出，多次修复：面向工具智能体修复的扩散引导分层搜索
One Readout, Many Repairs: Diffusion-Guided Hierarchical Search for Tool-Agent Repair
arXiv 2609.34879 · 2026-09-30上下文依赖的时间序列预测：基于HyperReservoirs方法
Context-dependent time-series prediction via HyperReservoirs
arXiv 2609.34847 · 2026-09-30AGRO-SUVIDE：用于外科粘弹性清创的智能体机器人技术
AGRO-SUVIDE: Agentic Robotics for Surgical Viscoelastic Debridement
arXiv 2609.34823 · 2026-09-30加速器选择并非全部：AlphaFold2 在云 TPU 上的推理
Accelerator Choice Is Not Enough: AlphaFold2 Inference on Cloud TPUs
arXiv 2609.34818 · 2026-09-30实射影平面上的半经典刘维尔理论：自举的复解释
Semiclassical Liouville Theory on the Real Projective Plane: A Complex Interpretation of the Bootstrap
arXiv 2609.34811 · 2026-09-30STRIDE：跨多样情境的文本到轨迹对齐自动化评估
STRIDE: Automated Evaluation of Text-to-Trajectory Alignment across Diverse Contexts
arXiv 2609.34799 · 2026-09-30CoSec：社区中智能体安全性的基准测试
CoSec: Benchmarking Agent Security in Communities
arXiv 2609.34790 · 2026-09-30在医学中应用语言模型：近期趋势与展望
Applying Language Models in Clinical Medicine: Recent Trends and Perspectives
arXiv 2609.34780 · 2026-09-30从像素生成到拓扑推断：用于可信跨物理域骨小梁形态学习的结构双重超分辨率
From Pixel Generation to Topological Inference: Structural Dual Super-Resolution for Trustworthy Cross-Physical-Domain Trabecular Morphology Learning
arXiv 2609.34716 · 2026-09-30CapField-OPD：通过联合锚定多教师在线策略蒸馏学习连续能力场用于流模型
CapField-OPD: Learning Continuous Capability Fields via Joint-Anchored Multi-Teacher On-Policy Distillation for Flow Models
arXiv 2609.34658 · 2026-09-30OmniTide：面向高效端侧全模态流式推理的算法与系统协同设计
OmniTide: Co-Designing Algorithms and Systems for Efficient On-Device Omni-LLM Streaming
arXiv 2609.34653 · 2026-09-30DirectUV：基于图像条件的UV纹理生成与表面感知位置编码
DirectUV: Image-Conditioned UV Texture Generation with Surface-Aware Positional Encoding
arXiv 2609.34651 · 2026-09-30修复之后：修正后的智能体历史如何迁移到相关任务
After the Fix: Transfer of Corrected Agent Experience
arXiv 2609.34603 · 2026-09-30用于时间视频定位的反事实注意力策略蒸馏
Counterfactual Attention Policy Distillation for Temporal Video Grounding
arXiv 2609.34581 · 2026-09-30GenNVS：通过解耦3D先验的几何增强新视角合成
GenNVS: Geometry-enhanced Novel View Synthesis via Disentangled 3D Prior
arXiv 2609.34579 · 2026-09-30通过塞贝克系数探测石墨烯和夸克物质中的流体动力学
Probing hydrodynamics in graphene and quark matter via Seebeck coefficient
arXiv 2609.34544 · 2026-09-30如何驯服多头九头蛇？面向大语言模型的自适应多类别安全引导
How to Tame a Multi-Headed Hydra? Adaptive Multi-Category Safety Steering for Large Language Models
arXiv 2609.34514 · 2026-09-30HALO：通过超球面潜变量与掩码自回归建模增强时间序列生成
HALO: Enhancing Time Series Generation via Hyperspherical Latents and Masked AutoregRessive Modeling
arXiv 2609.34511 · 2026-09-30利用图神经网络进行高粒度量能器中重叠物体的分割
Using Graph Neural Networks for the segmentation of overlapping objects in high granularity calorimeters
arXiv 2609.34500 · 2026-09-30对齐引导的流变换器用于高效视觉-语言-动作策略学习
Alignment-Guided Flow Transformer for Efficient Vision-Language-Action Policy Learning
arXiv 2609.34467 · 2026-09-30Q-learning 惩罚 Transformer 用于安全离线强化学习
Q-learning Penalized Transformer for Safe Offline Reinforcement Learning
arXiv 2609.34426 · 2026-09-30CAR-VLA：自动驾驶的复杂度感知与风险自适应推理
CAR-VLA: Complexity-Aware and Risk-Adaptive Reasoning for Autonomous Driving
arXiv 2609.34387 · 2026-09-30
