2026-09论文中文摘要
风格胜于实质：内容不变包装器翻转LLM安全裁判的判定
Style Over Substance: Content-Invariant Wrappers Flip LLM Safety-Judge Verdicts
arXiv 2609.08236 · 2026-09-09剪切和拉伸流动下蠕虫状胶束流动诱导断裂的统一描述
A unified description of flow-induced scission of wormlike micelles under shear and extensional flows
arXiv 2609.08235 · 2026-09-09变点数量的无分布推断
Distribution-free inference on the number of changepoints
arXiv 2609.08234 · 2026-09-09Transversal Fanout 用于容错分布式量子计算：分析与应用
Transversal Fanout for Fault Tolerant Distributed Quantum Computing: Analysis and Application
arXiv 2609.08233 · 2026-09-09使用LSTM的基于历史感知的离线强化学习路由稠密布局
Routing Dense Layouts with History-Aware Offline Reinforcement Learning using LSTM
arXiv 2609.08232 · 2026-09-09SemBridge：将消费者观察编译为跨栈通信计划
SemBridge: Compiling Consumer Observations into Cross-Stack Communication Plans
arXiv 2609.08231 · 2026-09-09源到观测电磁散射格林算子的有限模态实现与算子范数收敛
Finite-Modal Realization and Operator-Norm Convergence of a Source-to-Observation Electromagnetic Scattering Green Operator
arXiv 2609.08229 · 2026-09-09关于Camassa-Holm方程的奇异极限
On the singular limit of the Camassa-Holm equation
arXiv 2609.08227 · 2026-09-09TTGBench：文本属性时序图中的拓扑演化与语义漂移基准测试
TTGBench: Benchmarking Topological Evolution and Semantic Drift in Text-attributed Temporal Graphs
arXiv 2609.08226 · 2026-09-09多目标复合纵向生物标志物评分用于改进癌症风险评估
Multi-Objective Composite Longitudinal Biomarker Scores for Improved Cancer Risk Assessment
arXiv 2609.08225 · 2026-09-093DWay：通过3D一致路径点实现机器人操作泛化
3DWay: Generalizing Robot Manipulation via 3D Consistent Waypoints
arXiv 2609.08224 · 2026-09-09具有SU(N)非阿贝尔规范场的方格系统中的大陈数平带、反常狄拉克锥和非常规超流性
Large-Chern-number flat bands, anomalous Dirac cones, and unconventional superfluidity in square-lattice systems with SU(N) non-Abelian gauge fields
arXiv 2609.08223 · 2026-09-09三角条带格上的平衡生成树
Balanced Spanning Trees for Triangular Strip Lattices
arXiv 2609.08222 · 2026-09-09SoftRerank：面向长尾微动作识别的候选标签重排序分层软融合方法
SoftRerank: Hierarchical Soft Fusion with Candidate-Label Reranking for Long-Tailed Micro-Action Recognition
arXiv 2609.08221 · 2026-09-09弥合语言与物理：利用大型语言模型自动设计连续体机器人
Bridging Language and Physics: Automated Design of Continuum Robots with Large Language Models
arXiv 2609.08220 · 2026-09-09随机学习动力学中信息获取的速度极限
Speed Limit for Information Acquisition in Stochastic Learning Dynamics
arXiv 2609.08219 · 2026-09-09通过后见之明与预见之明驾驶：面向自动驾驶的层次化记忆上的工具接地协同推理
Drive by Hindsight and Foresight: Tool-Grounded Synergistic Reasoning over Hierarchical Memory for Autonomous Driving
arXiv 2609.08217 · 2026-09-09视觉：面向鲁棒且可解释智能体AI的数据中心锚定
Vision: Data-Centric Anchoring for Robust and Interpretable Agentic AI
arXiv 2609.08216 · 2026-09-09PhysFlow：面向运动可控视频生成的物理感知光流
PhysFlow: Physics-Aware Optical Flow for Motion Controllable Video Generation
arXiv 2609.08215 · 2026-09-09TacClip：一种不覆盖指腹的夹持式传感器，用于测量动态接触力
TacClip: a clip-on sensor measures dynamic contact forces without covering the fingerpads
arXiv 2609.08214 · 2026-09-09DRIFT：通过偏转生成轨迹去除扩散水印
DRIFT: Removing Diffusion Watermarks by Deflecting the Generative Trajectory
arXiv 2609.08213 · 2026-09-09三类带自由边界的反应-扩散方程的有限差分方法
Finite difference methods for three kinds of reaction-diffusion equations with free boundaries
arXiv 2609.08212 · 2026-09-09更好的助推应始于每个目标
A Better Spur Should Start From Each Objective
arXiv 2609.08211 · 2026-09-09交换两个间隔一个格点的区间互信息中的极限次序
Interchanging limits in the mutual information of two intervals separated by one site
arXiv 2609.08210 · 2026-09-09$^{39}\mathrm{K}$玻色-爱因斯坦凝聚实验中磁阱装载效率的提升
Increase of Magnetic Trap Loading Efficiency for $^{39}\mathrm{K}$ Bose--Einstein Condensation Experiments
arXiv 2609.08208 · 2026-09-09三维无线电断层成像的正则化方法基准测试
Benchmarking Regularization Methods For 3D Radio Tomographic Imaging
arXiv 2609.08207 · 2026-09-09由小量子群控制的量子群的同调性质
Homological properties of quantum groups governed by small quantum groups
arXiv 2609.08205 · 2026-09-09稳定指令监督用于Instruct-TTS：通过可控多样化与漂移过滤
Stabilizing Instruction Supervision for Instruct-TTS via Controllable Diversification and Drift Filtering
arXiv 2609.08204 · 2026-09-09粘弹性重塑旋转磁性粒子的频率响应
Viscoelasticity reshapes the frequency response of a rotating magnetic particle
arXiv 2609.08202 · 2026-09-09非几何二次诱导的Lang-Trotter问题
A Lang-Trotter Problem for Non-Geometric Quadratic Inductions
arXiv 2609.08201 · 2026-09-09SIM：基于子空间交互的令牌级文本异常检测方法
SIM: Subspace Interaction-based Method for Token-Level Text Anomaly Detection
arXiv 2609.08200 · 2026-09-09QoS感知的RACH前导码切片：基于配额投影的分支深度强化学习
QoS-Aware RACH Preamble Slicing via Quota-Projected Branching Deep Reinforcement Learning
arXiv 2609.08199 · 2026-09-09自我实现的预言、准非遍历性与财富不平等：一个评论
Self-Fulfilling Prophecies, Quasi Nonergodicity, and Wealth Inequality: A Comment
arXiv 2609.08198 · 2026-09-09过渡金属中原子体积系统学的压力演化
Pressure Evolution of Atomic Volume Systematics in Transition Metals
arXiv 2609.08197 · 2026-09-09Qiushi Engine 在 AstaBench E2E-Bench-Hard 上的表现
Qiushi Engine on AstaBench E2E-Bench-Hard
arXiv 2609.08196 · 2026-09-09几何对单侧布朗环境中不连通分形集上扩散过程极限行为的影响
Geometric influence on the limiting behavior of diffusion processes in one-sided Brownian environments on disconnected fractal sets
arXiv 2609.08195 · 2026-09-09固定边密度图中最大化 $K_r + I_r$
Maximizing $K_r + I_r$ in graphs with fixed edge density
arXiv 2609.08194 · 2026-09-09Fefferman--Stein 型不等式：通过 Schrödinger 算子的面积函数与极大函数及其应用
Fefferman--Stein type inequalities via area and maximal functions for Schrödinger operators with applications
arXiv 2609.08193 · 2026-09-09集成yoctosecond精度定时探测器
Integrated yoctosecond-precision timing detector
arXiv 2609.08192 · 2026-09-09分离单词的 $\widetilde O(n^{1/3})$ 界的一个初等证明
An Elementary Proof of the $\widetilde O(n^{1/3})$ Bound for Separating Words
arXiv 2609.08191 · 2026-09-09关于元组格筛法的球堆积界的一个注记
A Note on Sphere Packing Bounds for Tuple Lattice Sieving
arXiv 2609.08190 · 2026-09-09动态路由器需要记忆吗？HeRo：面向高效LLM推理的历史感知路由
Do Dynamic Routers Need Memory? HeRo: History-Aware Routing for Efficient LLM Inference
arXiv 2609.08189 · 2026-09-09弥合多模态RAG中的语义-效用差距：基于生成器在环对齐
Bridging the Semantic-Utility Gap in Multimodal RAG via Generator-in-the-Loop Alignment
arXiv 2609.08188 · 2026-09-09BeamFocusNet：基于波束形成的显式空间信号聚焦用于低信噪比和单快照条件下的稳健DoA估计
BeamFocusNet: Beamforming-based Explicit Spatial Signal Focusing for Robust DoA Estimation under Low-SNR and Single-Snapshot Conditions
arXiv 2609.08187 · 2026-09-09更深层推理会损害对齐吗？大型推理模型中对齐崩溃的揭示与缓解
Does Deeper Reasoning Compromise Alignment? Revealing and Mitigating of Alignment Collapse in Large Reasoning Models
arXiv 2609.08186 · 2026-09-09Cassette：面向高效法律案例检索的案例到案例结构蒸馏
Cassette: Case-to-Case Structural Distillation for Efficient Legal Case Retrieval
arXiv 2609.08185 · 2026-09-09UnespDataLens-RM：一种面向分析性数据工程的参考模型，涵盖治理、质量、溯源与可复现性
UnespDataLens-RM: A Reference Model for Analytical Data Engineering with Governance, Quality, Provenance, and Reproducibility
arXiv 2609.08184 · 2026-09-09NeoHorse-1：通过带路由框架的智能体后训练实现递归自我改进
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
arXiv 2609.08183 · 2026-09-09复折射率调制下光子时间晶体的拓扑性质
Topology of Photonic Time Crystals under Complex Refractive Index Modulation
arXiv 2609.08181 · 2026-09-09相对 $(φ, Γ)$-模与 $p$-adic 微分方程
Relative $(φ, Γ)$-modules and $p$-adic differential equations
arXiv 2609.08179 · 2026-09-09
