2026-10论文中文摘要
τ的轻子衰变研究
On the leptonic decays of the $τ$
arXiv 2610.10670 · 2026-10-09台站依赖噪声配置对低频段爱因斯坦望远镜科学性能的影响
Impact of site-dependent noise configurations on Einstein Telescope science at low frequency
arXiv 2610.10669 · 2026-10-092+1维拓扑有序超流体理论
Theory of Topologically Ordered Superfluids in 2+1 Dimensions
arXiv 2610.10668 · 2026-10-09闭合宇宙、量子纠缠与观测者
Closed Universe, Quantum Entanglement and Observers
arXiv 2610.10667 · 2026-10-09解释对抗训练神经网络的显著性图稀疏性
Explaining the Saliency Map Sparsity of Adversarially-Trained Neural Networks
arXiv 2610.10666 · 2026-10-09整数格中的Erdős-Falconer距离问题
The Erdős-Falconer distance problem in the integer lattice
arXiv 2610.10665 · 2026-10-09泊松-二项分布点概率中距离为2为何特殊
Why Distance Two Is Exceptional for Poisson-Binomial Point Probabilities
arXiv 2610.10664 · 2026-10-09BEANS-Next与ROOTS：拓展生物声学的音频-语言能力
BEANS-Next and ROOTS: Broadening Audio-Language Capabilities for Bioacoustics
arXiv 2610.10663 · 2026-10-09教PPG模型“不做什么”而非“做什么”：基于固定效应的ECG知识蒸馏
Teaching PPG How not Who: Fixed-Effects Distillation from ECG
arXiv 2610.10662 · 2026-10-09GL₃与GL₁的扭L-函数的同时非零性
Simultaneous non-vanishing of twists of $\mathrm{GL}_3$ and $\mathrm{GL}_1$ $L$-functions
arXiv 2610.10661 · 2026-10-09三个二次Collatz型映射的轨道分类
Classification of the Orbits of Three Quadratic Collatz-type Maps
arXiv 2610.10660 · 2026-10-09在执行点应用设计安全：受管控的安全需求如何影响AI生成代码的安全性
Applying Security by Design at the Point of Execution: How Governed Security Requirements Affect the Security of AI-Generated Code
arXiv 2610.10659 · 2026-10-09三维聚焦能量临界非线性薛定谔方程的阈值散射
Threshold Scattering for the Three-Dimensional Focusing Energy-Critical Nonlinear Schrödinger Equation
arXiv 2610.10658 · 2026-10-09超越猫头鹰：阈下学习可迁移习得的能力与后门
Beyond Owls: Subliminal Learning Can Transfer Learned Capabilities and Backdoors
arXiv 2610.10657 · 2026-10-09大型语言模型（LLMs）能否模拟新手程序员的错误概念？
Can LLMs Simulate Novice Programmers' Misconceptions?
arXiv 2610.10656 · 2026-10-09Nullify：用于无需训练的大语言模型遗忘的零空间激活引导
Nullify: Null-Space Activation Steering for Training-Free LLM Unlearning
arXiv 2610.10655 · 2026-10-09部分Twitch聊天展现出类似意识流叙事的多重分形特征
Some Twitch chats exhibit multifractal characteristics resembling stream-of-consciousness narrative
arXiv 2610.10654 · 2026-10-09PXtal：在跨模态信息不对称下学习对齐粉末X射线衍射与晶体结构
PXtal: Learning to Align Powder X-Ray Diffraction and Crystal Structures under Information Asymmetry across Modalities
arXiv 2610.10653 · 2026-10-09具有正整数质量的傅里叶拟晶体是李-杨循环的截面
Fourier quasicrystals with positive integer masses are sections of Lee-Yang cycles
arXiv 2610.10652 · 2026-10-09超越遍历壁垒：用于分析AI缩放极限与复杂度崩溃的离散几何物理沙盒
Beyond the Ergodic Wall: A Discrete Geometric Physics Sandbox for Analysing AI Scaling Limits and Complexity Collapse
arXiv 2610.10651 · 2026-10-09大语言模型辅助交通管理计划编制：以WisDOT WisTMP系统为例
Large Language Model-Assisted Preparation of Transportation Management Plans: A Case Study with WisDOT WisTMP System
arXiv 2610.10650 · 2026-10-09两个证明与一个紧的Szemerédi–Trotter定理
Two proofs and a tight Szemerédi--Trotter theorem
arXiv 2610.10649 · 2026-10-09用于阿尔茨海默病研究诊断与进展预测的泄漏控制多模态学习
Leakage-Controlled Multimodal Learning for Diagnosis and Progression Prediction in Alzheimer's Disease Research
arXiv 2610.10648 · 2026-10-09具有非准静态绝热冲程的量子热机中的效率涨落
Efficiency fluctuations in quantum heat engines with non-quasi-static adiabatic strokes
arXiv 2610.10647 · 2026-10-09带几何引导令牌搜索的掩码生成运动规划
Masked Generative Motion Planning with Geometry-Guided Token Search
arXiv 2610.10646 · 2026-10-09理想气体理论中热平衡与体积依赖熵之间的公理不相容性
Axiomatic Incompatibility Between Thermal Equilibrium and Volume-Dependent Entropy in Ideal Gas Theory
arXiv 2610.10645 · 2026-10-09SoK：通用标准产品评估中的失效模式——分类与可评估性设计指南
SoK: Failure Modes in Common Criteria Product Evaluation - A Taxonomy and Design-for-Evaluability Guidance
arXiv 2610.10644 · 2026-10-09完美长方体的秩条件
On the Rank Condition for Perfect Cuboids
arXiv 2610.10643 · 2026-10-09从对数几率到Shapley值：加权朴素贝叶斯分类器的解释性几何
From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier
arXiv 2610.10642 · 2026-10-09面向诊断预测的医学标记覆盖感知推理
Coverage-Aware Reasoning with Medical Tokens for Diagnosis Prediction
arXiv 2610.10641 · 2026-10-09截断扩散采样器必须保留多少个方向？幂律谱下的匹配界
How Many Directions Must a Truncated Diffusion Sampler Retain? Matching Bounds Under Power-Law Spectra
arXiv 2610.10640 · 2026-10-09可见推理并非通用优化器：分析代码生成中依赖角色与思维的效应
Visible Reasoning Is Not a Universal Optimizer: Persona- and Thinking-Dependent Effects in Analytics Code Generation
arXiv 2610.10639 · 2026-10-09黑洞阴影的几何热力学：综述
Geometrothermodynamics of Black Hole Shadows: A Review
arXiv 2610.10638 · 2026-10-09TacHair：用于机器人抚发与感知的触觉接触分布引导型在线校正方法
TacHair: Tactile Contact-Distribution Guided Online Correction for Robotic Hair Stroking and Perception
arXiv 2610.10637 · 2026-10-09D-SLR：不相交行稀疏加低秩分解
D-SLR: The Disjoint Row-Sparse plus Low-Rank Decomposition
arXiv 2610.10636 · 2026-10-09导航员的语言：面向冻结空中视觉语言导航（VLN）智能体的轨迹基础指令翻译
Speaking the Navigator's Language: Trajectory-Grounded Instruction Translation for Frozen Aerial VLN Agents
arXiv 2610.10635 · 2026-10-09拓扑球面λ-自展子的例子
Examples of topological sphere $λ$-self-expander
arXiv 2610.10634 · 2026-10-09大型语言模型在软件工程系统综述中的筛选性能是否已停滞？
Has LLM Screening Performance Stalled in Software Engineering Systematic Reviews?
arXiv 2610.10633 · 2026-10-09重新审视多联骨牌计数：勘误与补遗
Counting Polyominoes, Revisited: Corrigendum and Addendum
arXiv 2610.10632 · 2026-10-09无规则数字孪生：通过标准化框架与技术实现声明式决策
Ruleless Digital Twins: Toward Declarative Decision-Making Through Standardized Frameworks and Technologies
arXiv 2610.10631 · 2026-10-09Phase-HDC：在离散相位学习中用梯度阈值替代优化器历史记录
Phase-HDC: Replacing Optimizer History with Gradient Thresholds in Discrete Phase Learning
arXiv 2610.10630 · 2026-10-09仅将控制流作为唯一可变表面：信贷流程中LLM智能体的合规性受限自演化及带测量的准入门
The Harness as the Only Mutable Surface: Compliance-Bounded Self-Evolution of LLM Agents in Credit Pipelines, with a Measured Admission Gate
arXiv 2610.10629 · 2026-10-09Agent4RE：用于端到端软件需求工程与基准测试的自优化多智能体框架
Agent4RE: A Self-Refining Multi-agent Framework for End-to-End Software Requirements Engineering and Benchmarking
arXiv 2610.10628 · 2026-10-09柯尔莫哥洛夫-阿诺德网络的样本效率
Sample-Efficiency of Kolmogorov-Arnold Networks
arXiv 2610.10627 · 2026-10-09适用于旋转鲁棒神经动力学的精确SO(3)等变各向同性核
Exact SO(3)-Equivariant Isotropic Kernels for Rotation-Robust Neural Dynamics
arXiv 2610.10626 · 2026-10-09在一个循环中安全，在另一个循环中危险：在循环语言模型的循环深度间对齐安全性
Safe at One Loop, Risky at Another: Aligning Safety Across Recurrent Depths in Looped Language Models
arXiv 2610.10625 · 2026-10-09SAVU-BENCH：用于空间视听理解的真实世界基准
SAVU-BENCH: A Real-World Benchmark for Spatial Audio-Visual Understanding
arXiv 2610.10624 · 2026-10-09循环自改进：面向循环语言模型的动态跨循环在线蒸馏
Recurrent Self-Improvement: Dynamic Cross-Loop On-Policy Distillation for Looped Language Models
arXiv 2610.10623 · 2026-10-09WorldBench：在Three.js体素世界生成任务上评估大语言模型
WorldBench: Evaluating LLMs on Three.js Voxel World Generation
arXiv 2610.10622 · 2026-10-09受约束几何辐射的自组织
Self-Organization from Constrained Geometric Radiation
arXiv 2610.10621 · 2026-10-09
