2026-09论文中文摘要
HALO-WA：用于世界-动作模型的混合注意力潜在引导在线强化学习
HALO-WA: Hybrid-Attention Latent-Guided Online Reinforcement Learning for World-Action Models
arXiv 2607.04265 · 2026-09-22具有通用软超对称能谱的最小超对称SO(10)的更新分析
Updated analysis of minimal supersymmetric SO(10) with a universal soft spectrum
arXiv 2607.04250 · 2026-09-22Why3-py：用于Python中假设检验和元分析形式验证的工具
Why3-py: A Tool for Formal Verification of Hypothesis Testing and Meta-Analysis in Python
arXiv 2607.03951 · 2026-09-22量子通信信道偏振模色散的宽带特性表征
Broadband characterization of polarization mode dispersion for quantum communication channels
arXiv 2607.03202 · 2026-09-22全息光瞬态显微镜中的双向相位灵敏度
Bidirectional phase sensitivity in holographic phototransient microscopy
arXiv 2607.03170 · 2026-09-22超越点附着语义：用于可泛化操作的以对象为中心的语义场
Beyond Point-Attached Semantics: Stable Object-Centric Semantic Fields for Robust Manipulation
arXiv 2607.03163 · 2026-09-22自旋-1 三角晶格反铁磁体的量子磁性
Quantum magnetism of the spin-1 kagome-lattice antiferromagnet
arXiv 2607.03086 · 2026-09-22逆问题的方向加权全变差
Directionally Weighted Total Variation for Inverse Problems
arXiv 2607.03054 · 2026-09-22跨度内学习：利用自身预测调整降阶模型
In-context learning from self-generated trajectories for adaptive model reduction
arXiv 2607.02937 · 2026-09-22语言模型中风险规避的分布外泛化
Out-of-Distribution Generalization of Risk Aversion in Language Models
arXiv 2607.02755 · 2026-09-22标量张量理论中的黑洞持续性
Black Hole Persistence in Scalar Tensor Theories
arXiv 2607.02409 · 2026-09-22CSI仿真：为什么加性噪声失败以及如何修复
CSI Simulation: Why Additive Noise Fails and How to Fix It
arXiv 2607.01882 · 2026-09-22重新思考语音-LLM集成用于ASR：通过交错实现有效的联合语音-文本训练
Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving
arXiv 2607.01733 · 2026-09-22自指 $K$-SAT 与哥德尔不完备定理的有限模拟
Self-Referential $K$-SAT and the Finite Analogue of Gödel's Incompleteness Theorem
arXiv 2607.01671 · 2026-09-22在质子-质子碰撞中 $\sqrt{s}$ = 13 TeV 下研究 bb$\tau\tau$ 末态的 ZZ 和 ZH 产生并寻找高自旋 0 和自旋 1 共振
Study of ZZ and ZH production in the bb$ττ$ final state and search for high-mass spin-0 and spin-1 resonances in proton-proton collisions at $\sqrt{s}$ = 13 TeV
arXiv 2607.01589 · 2026-09-22拓扑弦的模恢复
Modularity of resurgent topological string
arXiv 2607.00880 · 2026-09-22实用的带推断的范围细化类型
Practical Range Refinement Types with Inference
arXiv 2607.00824 · 2026-09-22多早才算足够早？订阅流失预测中依赖于设计的观察窗口充分性
How Early Is Early Enough? Design-Dependent Observation-Window Sufficiency in Subscription Churn Prediction
arXiv 2607.00473 · 2026-09-22Designer-RSI：从用户流量中演化程序性记忆用于智能体图形设计
Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design
arXiv 2609.22086 · 2026-09-21多边界旋转 $AdS_5$ 黑洞、强耦合等离子体球及对偶的局部 de Sitter 时空
Multi-Boundary Spinning $AdS_5$ Black Holes, Strongly Coupled Plasma Balls and a dual locally de Sitter spacetime
arXiv 2609.22084 · 2026-09-21MintAct：面向数字环境的统一视觉智能体
MintAct: A Unified Visual Agent for Digital Environments
arXiv 2609.22083 · 2026-09-21关于几何生成子图的（有向）宽度参数
On (Directed) Width-Parameters of Geometric Spanners
arXiv 2609.22082 · 2026-09-21旋转中子星迁移作为3+1数值相对论的标准测试
Rotating Neutron Star Migrations as a Standardized Test for 3+1 Numerical Relativity
arXiv 2609.22080 · 2026-09-21非厄米随机矩阵的普适特征向量统计
Universal Eigenvector Statistics of Non-Hermitian Random Matrices
arXiv 2609.22079 · 2026-09-21单位球面上八个点的能量最小化
Energy minimization for eight points on the sphere
arXiv 2609.22077 · 2026-09-21APort Vault：使用开放智能体护照基准测试AI智能体支付授权
APort Vault: Benchmarking AI Agent Payment Authorization with the Open Agent Passport
arXiv 2609.22076 · 2026-09-21LIMBO：学习并内化无模型障碍函数目标以实现敏捷且安全的全身控制
LIMBO: Learning and Internalizing Model-Free Barrier Objectives for Agile and Safe Whole-Body Control
arXiv 2609.22075 · 2026-09-21三维中的活动星系核通道：散射带与黑洞族群偏心率的的重要性
The AGN Channel in 3D: Scattering Belts and the Importance of Eccentricity in the Black Hole Population
arXiv 2609.22074 · 2026-09-21占空比预测跨步态类型的鲁棒约束四足运动
Duty Factor Predicts Robust Constrained Quadrupedal Locomotion Across Gait Types
arXiv 2609.22073 · 2026-09-21行星质量伴星Ross 458C自转调制的多历元稳定性
Multi-Epoch Stability in the Rotational Modulation of the Planetary-Mass Companion Ross 458C
arXiv 2609.22072 · 2026-09-21发现、表征及邻近LMC质量星系NGC 55恒星晕中一条星流的可能起源
Discovery, Characterization, and Potential Origins of a Stream in the Stellar Halo of Nearby LMC-Mass Galaxy NGC 55
arXiv 2609.22071 · 2026-09-21卫生系统中算法的社会技术审视：技术、成本与以人为本的考量
A Sociotechnical Review of Algorithms in Health Systems: Technical, Cost, and Human-Centered Considerations
arXiv 2609.22070 · 2026-09-21OmniVBench：面向全参考到视频生成的基准与大规模数据集
OmniVBench: A Benchmark and Large-Scale Dataset for Omni Reference-to-Video Generation
arXiv 2609.22069 · 2026-09-21CodeMidas：从代码本身扩展智能体编码强化学习环境
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
arXiv 2609.22068 · 2026-09-21日常AI智能体使用中的价值敏感委托：来自OpenClaw的证据
Value-Sensitive Delegation in Everyday AI Agent Use: Evidence from OpenClaw
arXiv 2609.22067 · 2026-09-21空间非局部非线性微分方程的广义哈密顿形式体系
Generalized Hamiltonian formalism for spatially nonlocal nonlinear differential equations
arXiv 2609.22066 · 2026-09-21Hermite-Fisher 界与最小熵幂不等式稳定性
Hermite-Fisher bounds and stability for min-entropy power inequalities
arXiv 2609.22065 · 2026-09-21BrainWideBench：多区域神经记录中的大规模预训练与跨动物迁移基准
BrainWideBench: Benchmarking large-scale pretraining and across-animal transfer in multi-region neural recordings
arXiv 2609.22064 · 2026-09-21通过II型跷跷板机制复活电弱暗物质：基于近期LZ事件的研究
Resurrecting Electroweak Dark Matter via Type-II Seesaw in light of recent LZ Event
arXiv 2609.22063 · 2026-09-21抓爪感知的不规则物体自动密集装箱
Gripper-Aware Automatic Dense Packing of Irregular Objects
arXiv 2609.22062 · 2026-09-21伪曲面中图嵌入的对偶性与 minors
Duality and minors for embeddings of graphs in pseudosurfaces
arXiv 2609.22061 · 2026-09-21用于自动驾驶的基于分支YOLOv2与几何特征的交通标志识别
Traffic Sign Recognition for Autonomous Driving Using Branched YOLOv2 and Geometric Features
arXiv 2609.22060 · 2026-09-21维数至少为四的每个双曲空间中强稳定的常平均曲率为一的超曲面
Strongly stable CMC-one hypersurfaces in every hyperbolic space of dimension at least four
arXiv 2609.22059 · 2026-09-21非线性晕偏置用于宇宙红外背景各向异性的精确建模
Non-linear halo bias for accurate modelling of cosmic infrared background anisotropies
arXiv 2609.22057 · 2026-09-21多跳检索中的可预测失败：分数分布置信度评分与弃权（不执行）
Predictable Failure in Multi-Hop Retrieval: Score-Distributional Confidence Scoring and Abstention
arXiv 2609.22056 · 2026-09-21组合任务持续学习的世界模型基准测试
Benchmarking World Models for Continual Learning on Compositional Tasks
arXiv 2609.22055 · 2026-09-21$\langle 3,2,m \rangle$ 矩阵乘法的下界
A lower bound for $\langle 3,2,m \rangle$ matrix multiplication
arXiv 2609.22054 · 2026-09-21粒子竞争与合作用于标签噪声下鲁棒图卷积网络学习
Particle Competition and Cooperation for Robust Graph Convolutional Network Learning Under Label Noise
arXiv 2609.22053 · 2026-09-21透明参数化模型损失巨灾债券的设计与定价：以德国风暴为例
Design and pricing of a transparent parametric-modeled loss CAT bond: application to German windstorm
arXiv 2609.22052 · 2026-09-21超导铜氧化物/锰氧化物界面处的关联增强磁性
Correlation enhanced magnetism at a superconducting cuprate/manganite interface
arXiv 2609.22051 · 2026-09-21
