2026-10论文中文摘要
具有不变除数的通用柔性簇
Generically Flexible Varieties with an Invariant Divisor
arXiv 2610.05344 · 2026-10-06MemStrata：本地 Qwen 3.8 27B Q4_K_M 阅读器在 LongMemEval-500 和 LoCoMo-1540 上实现 95% 和 90.91% 的源感知准确率
MemStrata: 95% and 90.91% Source-Aware Accuracy on LongMemEval-500 and LoCoMo-1540 with a Local Qwen 3.8 27B Q4_K_M Reader
arXiv 2610.05343 · 2026-10-06IRSTD-Agent：基于缩放引导交互学习的智能体红外小目标检测
IRSTD-Agent: Agentic Infrared Small Target Detection via Zoom-Guided Interaction Learning
arXiv 2610.05342 · 2026-10-06WILLIE：用于伤口分类、分割和定位的统一框架与基准
WILLIE: A Unified Framework and Benchmark for Wound Classification, Segmentation, and Localization
arXiv 2610.05341 · 2026-10-06重叠多步预测的贝叶斯协调：跨来源一致性的后处理框架
Bayesian Reconciliation of Overlapping Multi-Step Forecasts: A Post-Processing Framework for Cross-Origin Coherence
arXiv 2610.05340 · 2026-10-06用于模式发现的非线性张量分解
Nonlinear Tensor Decomposition for Pattern Discovery
arXiv 2610.05339 · 2026-10-06分层强化学习中的半马尔可夫次优性
On Semi-Markov Suboptimality in Hierarchical Reinforcement Learning
arXiv 2610.05338 · 2026-10-06绿色路由神经算子：物理决定网络读取位置
Green-Routed Neural Operators:\\Physics Determines Where the Network Reads
arXiv 2610.05337 · 2026-10-06SheetSage2：基于合成监督的连贯主旋律谱转录
SheetSage2: Coherent Lead-Sheet Transcription with Synthetic Supervision
arXiv 2610.05336 · 2026-10-06StateSync-GKR：从稀疏默克尔状态转换到GKR验证的信任链的机器验证
StateSync-GKR: Machine-Checking the Trust Chain from Sparse-Merkle State Transitions to GKR Verification
arXiv 2610.05335 · 2026-10-06AgentDiscover：以最小搜索脚手架实现自主发现
AgentDiscover: Autonomous Discovery with Minimal Search Scaffolding
arXiv 2610.05334 · 2026-10-06扩散变换器是可证明最优的上下文生成器
Diffusion Transformers are Provably Optimal In-context Generators
arXiv 2610.05333 · 2026-10-06多分类中的紧致集值深度集成
Compact set-valued deep ensembling in multi-class classification
arXiv 2610.05332 · 2026-10-06VAMPS: 基于采样规划的视觉与运动策略
VAMPS: Visual and Motor Policies from Sampling-Based Planning
arXiv 2610.05331 · 2026-10-06CT-Miner：基于笛卡尔树的快速粗粒度时间序列模式挖掘
CT-Miner: Fast and Coarse-Grained Time-Series Pattern Mining via Cartesian Trees
arXiv 2610.05330 · 2026-10-06理解大语言模型训练中的权重平均机制用于训练后量化
Understanding the Weight Averaging Mechanism in LLM Training for Post-Training Quantization
arXiv 2610.05329 · 2026-10-06分层Pitman-Yor模型中随机索引簇计数的中偏差
Moderate Deviations for Random-Indexed Cluster Counts in Hierarchical Pitman-Yor Models
arXiv 2610.05328 · 2026-10-06近似消息传递在秩一二次感知中的定量普适性
Quantitative Universality of Approximate Message Passing for Rank-One Quadratic Sensing
arXiv 2610.05327 · 2026-10-06Kendall空间中胼胝体的黎曼形状分析：衰老与阿尔茨海默病
Riemannian Shape Analysis of the Corpus Callosum in Kendall Space: Aging and Alzheimer's Disease
arXiv 2610.05326 · 2026-10-06阴影特征细化网络：基于知识蒸馏的渐进式特征细化用于高效阴影去除
Shadow Feature Refinement Network: Progressive Feature Refinement based on Knowledge Distillation for Effective Shadow Removal
arXiv 2610.05325 · 2026-10-06CoDance：从视频中学习反应性与顺从性的人-人形机器人交互
CoDance: Learning Reactive and Compliant Human-Humanoid Interaction from Video
arXiv 2610.05324 · 2026-10-06语法引导的绿色温度代码水印
Grammar-Guided Code Watermarking with Green Temperature
arXiv 2610.05323 · 2026-10-06更长推理何时有帮助？通过发现与执行预测数学推理
When Does Longer Reasoning Help? Predicting Mathematical Reasoning Through Discovery and Execution
arXiv 2610.05322 · 2026-10-06准一维双带电子-声子模型中电荷密度波与超导态的竞争
Competing charge density-wave and superconducting states in a quasi-one dimensional two-band electron-phonon model
arXiv 2610.05321 · 2026-10-06加倍测度的弱反向Hölder不等式
Weak reverse Hölder inequalities for doubling measures
arXiv 2610.05320 · 2026-10-06鲁棒的模拟硬件参数高效大语言模型适配
Robust Parameter-Efficient LLM Adaptation on Analog Hardware
arXiv 2610.05318 · 2026-10-06打破双矩阵博弈近似纳什均衡1/3壁垒的简单算法
A Simple Algorithm Breaking the 1/3 Barrier for Approximate Nash Equilibria in Bimatrix Games
arXiv 2610.05317 · 2026-10-06利用SMT求解实现更强大的常量值检查
More Powerful Constant Value Checking with SMT Solving
arXiv 2610.05316 · 2026-10-06计算复杂度中相对于随机谕示的免疫分离
Separations with Immunity Relative to a Random Oracle in Computational Complexity
arXiv 2610.05315 · 2026-10-06多元正交多项式两个族的加法公式
Addition formula for two families of orthogonal polynomials in several variables
arXiv 2610.05314 · 2026-10-06Grad--Mercier 方程的一种有限元方法
A Finite Element Method for the Grad--Mercier Equation
arXiv 2610.05313 · 2026-10-06具有密度依赖体积粘性的二维可压缩磁流体力学方程的全局解
Global solutions to the two-dimensional compressible MHD equations with density-dependent bulk viscosity
arXiv 2610.05312 · 2026-10-06与斜多项式代数及曲面三角剖分相关的超图
Hypergraphs associated to skew polynomial algebras and surface triangulations
arXiv 2610.05311 · 2026-10-06语音基础模型的可迁移对抗鲁棒性：基于层级稳定化方法
Transferable Adversarial Robustness for Speech Foundation Models via Hierarchical Stabilization
arXiv 2610.05310 · 2026-10-06拟点精确性与挠理论
Quasi-pointed exactness and torsion
arXiv 2610.05309 · 2026-10-06RubricArmor：对抗演化改进基于LLM的评分标准生成
RubricArmor: Adversarial Evolution Improves LLM-Based Rubric Generation
arXiv 2610.05308 · 2026-10-06非阿贝尔双块群代数码的提升手术
Lifted surgery for non-Abelian two-block group-algebra codes
arXiv 2610.05307 · 2026-10-06以运动学为中心的连续手语检索：基于手语词汇引导的边界感知对齐
Kinematics-Centric Continuous Sign Language Retrieval with Gloss-Guided Boundary-Aware Alignment
arXiv 2610.05306 · 2026-10-06大语言模型推理中的并行策略表征：基础计算-通信权衡
Characterizing Parallelism Strategies in LLM Inference: Fundamental Compute-Communication Trade-offs
arXiv 2610.05305 · 2026-10-06结构化智能体工作流在应用量子计算研究中的演示
Demonstration of a Structured Agentic Workflow for Applied Quantum Computing Research
arXiv 2610.05304 · 2026-10-06ASCENT：通过验证经验的自蒸馏实现长程智能体的在线测试时训练
ASCENT: Online Test-Time Training of Long-Horizon Agents via Self-Distillation of Verified Experience
arXiv 2610.05303 · 2026-10-06非相干滤波器组
Incoherent filter banks
arXiv 2610.05302 · 2026-10-06从可及性到实现性可供性：大学生在不同语言与社会技术情境中的生成式人工智能参与
From Access to Realized Affordances: University Students' Generative AI Engagement across Linguistic and Sociotechnical Contexts
arXiv 2610.05301 · 2026-10-06MESH-Harness：通过波段引导的组合进化实现自我改进的智能体框架
MESH-Harness: Self-Improving Agent Harnesses via Bandit-Guided Compositional Evolution
arXiv 2610.05300 · 2026-10-06计算 Hecke 范畴中的双叶系数
Calculating double-leaf coefficients in the Hecke category
arXiv 2610.05299 · 2026-10-06重新审视高速视频的地面真值合成：精确有效性条件与审计的消费级采集语料库
Revisiting Ground-Truth Synthesis from High-Speed Video: Exact Validity Conditions and an Audited Consumer Capture Corpus
arXiv 2610.05298 · 2026-10-06神经网络能否学习保结构映射？潜在空间到希尔伯特空间嵌入的案例研究
Do Neural Networks Learn Structure-Preserving Maps? A Case Study in Latent-to-Hilbert Embeddings
arXiv 2610.05297 · 2026-10-06FlexCast：从任意字段集进行自适应天气预报
FlexCast: Adaptive Weather Forecasting from Arbitrary Field Sets
arXiv 2610.05296 · 2026-10-06可读先于可行动：间接提示注入的因果追踪
Readable Before Actionable: Causal Tracing of Indirect Prompt Injection
arXiv 2610.05295 · 2026-10-06利用宇宙射线缪子精确测量海洋潮汐
Precision measurement of ocean tides using cosmic-ray muons
arXiv 2610.05294 · 2026-10-06
