2026-10论文中文摘要
DuplexAct-Bench：面向多样化行为需求的主动交互，拓宽全双工语音评估
DuplexAct-Bench: Broadening Full-Duplex Speech Evaluation toward Proactive Interaction across Diverse Behavioral Requirements
arXiv 2609.39446 · 2026-10-01原始路由混合适配器：时间序列基础模型中路由崩溃的因果干预
Raw-Routed Mixture of Adapters: A Causal Intervention for Routing Collapse in Time Series Foundation Models
arXiv 2609.39445 · 2026-10-01具有周期平流项的双稳脉动波：均匀化与尖锐速度渐近
Bistable pulsating waves with periodic advection: homogenization and sharp speed asymptotics
arXiv 2609.39444 · 2026-10-01几何精确有限体积梁与不可压缩及自由表面流体的双向致动线耦合
Two-way actuator-line coupling of a geometrically exact finite-volume beam with incompressible and free-surface flow
arXiv 2609.39443 · 2026-10-01AI中介决策环境中的高管判断：形成、资格损耗与授权的过程理论
Executive Judgement in AI-Mediated Decision-Making Environments: A Process Theory of Formation, Qualification Attrition and Authorisation
arXiv 2609.39442 · 2026-10-01CAST：基于因果优势结构训练与空间接地组合奖励的扩散模型方法
CAST: Causal Advantage-Structured Training with Spatially Grounded Compositional Rewards for Diffusion Models
arXiv 2609.39441 · 2026-10-01主成分回归在所有单调谱滤波器中占优（用于线性回归）
Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression
arXiv 2609.39440 · 2026-10-01主动旋转器驱动被动物质进入手性转子
Active spinners drive passive matter into chiral rotors
arXiv 2609.39439 · 2026-10-01多运营商ISAC系统中无人机跟踪的运动不确定性联邦校准
Federated Calibration of Motion Uncertainty for UAV Tracking in Multi-Operator ISAC Systems
arXiv 2609.39438 · 2026-10-01基于网络的空间上下文检索用于开放权重大语言模型：一个面向扎根地理推理的忠实性基准
Network-based Spatial Context Retrieval for Open-weight LLMs: A Faithfulness Benchmark for Grounded Geographic Reasoning
arXiv 2609.39437 · 2026-10-01从模仿到奖励发现：智能体强化学习的同策略预热
From Imitation to Reward Discovery: On-Policy Warmup for Agentic RL
arXiv 2609.39436 · 2026-10-01巨型光磁二极管效应
Giant PhotoMagnetoDiode Effect
arXiv 2609.39435 · 2026-10-01仿射正交箭图的箭图簇
Quivers variety for affine orthogonal quiver
arXiv 2609.39434 · 2026-10-01教学中的AI危机：“训练你自己的神经网络！”
The AI crisis for teaching: "Train your own neural network!"
arXiv 2609.39433 · 2026-10-01探索暗时代21厘米涨落中的原初功率谱
Exploring the Primordial Power Spectrum with Dark-Age 21 cm fluctuations
arXiv 2609.39432 · 2026-10-01耦合刚体运动与二元相变：一种尖锐水平集/嵌入边界方法
Coupled Rigid-Body Motion and Binary Phase Change: A Sharp Level-Set/Embedded-Boundary Method
arXiv 2609.39431 · 2026-10-01Carter 型定理用于斜辫子（skew braces）
Carter-like theorems for skew braces
arXiv 2609.39430 · 2026-10-01迈向可信赖的胶质瘤诊断人工智能：面向任务的不确定性量化评估
Towards Trustworthy AI for Glioma Diagnosis: A Task-Aware Evaluation of Uncertainty Quantification
arXiv 2609.39429 · 2026-10-01孤立中子星表面磁场衰减与内部磁场深度之间的关联
Correlating Surface Magnetic Field Decay with Internal Field Depth of Isolated Neutron Stars
arXiv 2609.39428 · 2026-10-01PCB-MC：印刷电路板中的缺失元件分析
PCB-MC: Missing Component Analysis in Printed Circuit Boards
arXiv 2609.39427 · 2026-10-01蜈蚣博弈中公平结果的预期触发亲社会行为
The prospect of a fair outcome triggers prosocial behaviour in the Centipede Game
arXiv 2609.39426 · 2026-10-01学校选择中基于排名的优先级的局限性
The Limits of Rank-Dependent Priorities in School Choice
arXiv 2609.39425 · 2026-10-01局部谱隙与Stuck–Zimmer猜想
Local spectral gap and the Stuck--Zimmer conjecture
arXiv 2609.39424 · 2026-10-01多晶达夫毛石的弹性约束其在下地幔中的晶粒尺寸
Elasticity of polycrystalline davemaoite constrains its grain size in the lower mantle
arXiv 2609.39423 · 2026-10-01FissionReady：面向小型模块化反应堆供电数据中心的联合工作负载与电力调度
FissionReady: Joint Workload and Power Scheduling for Data Centers Powered by Small Modular Reactors
arXiv 2609.39422 · 2026-10-01IDEAL：一种用于脑电图-眼动情绪识别的多模态域自适应框架
IDEAL: A Multimodal Domain Adaptation Framework for EEG-Eye Emotion Recognition
arXiv 2609.39421 · 2026-10-01QuantCode模型：面向可执行算法交易代码的语言模型特化
QuantCode Model: Specializing Language Models for Executable Algorithmic Trading Code
arXiv 2609.39420 · 2026-10-01波动增长率与空间扩散塑造浮游生物多样性
Fluctuating growth rate and spatial diffusion shape plankton diversity
arXiv 2609.39419 · 2026-10-01基于标记多伯努利滤波的部分观测下无通信分布式多机器人任务分配
Communication-Free Distributed Multi-Robot Task Allocation under Partial Observations Using Labeled Multi-Bernoulli Filtering
arXiv 2609.39416 · 2026-10-01半导体自旋量子比特的无储层、无测量微波初始化
Reservoir- and Measurement-free Microwave Initialization of Semiconductor Spin Qubits
arXiv 2609.39415 · 2026-10-01SEW：LLM生成代码的风格编码水印
SEW: Style-Encoded Watermarking of LLM-Generated Code
arXiv 2609.39414 · 2026-10-01对齐不完整：多模态脑电-眼动情绪识别的联合分布校准
Aligning the Incomplete: Joint Distribution Calibration for Multimodal EEG-Eye Emotion Recognition
arXiv 2609.39413 · 2026-10-01画笔隐藏的一面：由幂律塑造的几何形状
The hidden side of paintbrushes: a geometry shaped by power laws
arXiv 2609.39410 · 2026-10-01正则点与一般光滑域中次椭圆过程的边值问题
Regular points and the boundary-value problem for a hypoelliptic process in a general smooth domain
arXiv 2609.39409 · 2026-10-01佛陀的觉醒：种群损失平台期中的子空间学习
Awakening of the Buddha: Subspace Learning During Population-Loss Plateaus
arXiv 2609.39408 · 2026-10-01Nice 划分、超可解性与图形排列形变中的自由性
Nice Partitions, Supersolvability, and Freeness in Deformations of Graphic Arrangements
arXiv 2609.39407 · 2026-10-01推断两个事件序列之间的因果关系：基于语言模型的方法
Inferring Causal Relations between Two Sequences of Events with Language Models
arXiv 2609.39406 · 2026-10-01没有任务向量是孤岛：关于从策略蒸馏任务向量可组合性的综合研究
No Task Vector Is an Island: A Comprehensive Study on the Composability of Task Vectors from On-Policy Distillation
arXiv 2609.39405 · 2026-10-01量子击穿模型中的手性量子混沌与指数多个零模
Chiral quantum chaos around exponentially many zero modes in the quantum breakdown model
arXiv 2609.39404 · 2026-10-01IronMind：通过相机空间自我中心预训练扩展人形机器人灵巧操作能力
IronMind: Scaling Humanoid Dexterous Manipulation via Camera-Space Ego-Centric Pretraining
arXiv 2609.39403 · 2026-10-01原初磁场对星系形成的影响
The impact of primordial magnetic fields on the formation of galaxies
arXiv 2609.39399 · 2026-10-01共享聚合收益：合作性不平衡成本分配
Sharing the Gains of Aggregation: Cooperative Imbalance Cost Allocation
arXiv 2609.39398 · 2026-10-01面向需求截断下的最优库存控制：一种有偏样本平均近似方法
Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach
arXiv 2609.39397 · 2026-10-01学习固定：面向加速机组组合的优化感知机器学习
Learning to Fix: Optimisation-Aware Machine Learning for Accelerated Unit Commitment
arXiv 2609.39396 · 2026-10-01来自先前问题的计算能否帮助LLMs解决新问题？
Can Computation from Earlier Problems Help LLMs Solve New Ones?
arXiv 2609.39394 · 2026-10-01通过非谐中介体的相互作用量子电池的光谱选择性充电
Spectrally Selective Charging of an Interacting Quantum Battery via an Anharmonic Mediator
arXiv 2609.39393 · 2026-10-01自主研究的实验经验建模
Experimental Experience Modeling for Autonomous Research
arXiv 2609.39392 · 2026-10-01降水阻塞流体流动的伪三维建模的严格垂直平均化
Rigorous vertical averaging for pseudo-3D modeling of precipitation obstructed fluid flow
arXiv 2609.39391 · 2026-10-01解耦与蒸馏：面向少样本类增量学习的任务自适应LoRA教师与集成知识迁移
Decoupled and Distilled: Task-Adaptive LoRA-Teachers with Ensemble Knowledge Transfer for Few-Shot Class-Incremental Learning
arXiv 2609.39390 · 2026-10-01多项式的自然重参数化与解映射的连续性
Natural reparameterization and continuity of the solution map for polynomials
arXiv 2609.39389 · 2026-10-01
