2026-10论文中文摘要
将大语言模型条件化于社会价值取向可提升其在序贯社会困境中的行为对齐
Conditioning LLMs on Social Value Orientation improves behavioural alignment in a sequential social dilemma
arXiv 2610.01667 · 2026-10-02带处理时间的线路网络路由问题
Routing in Line Networks with Handling Times
arXiv 2610.01666 · 2026-10-02代码检测器有半衰期：LLM生成代码检测中的过时与度量幻觉
Code Detectors Have a Half-Life: Obsolescence and Metric Illusions in LLM-Generated Code Detection
arXiv 2610.01664 · 2026-10-02非凸-凹极小极大优化中带方差缩减的随机一阶算法的下界
Lower Bounds for Stochastic First-Order Algorithms with Variance Reduction in Nonconvex--Concave Minimax Optimization
arXiv 2610.01662 · 2026-10-02DiVid：诊断视频生成模型中的维度特定多样性崩溃
DiVid: Diagnosing Dimension-Specific Diversity Collapse in Video Generation Models
arXiv 2610.01661 · 2026-10-02三阶对称张量的 Comon 秩间隙可以超过一
The Comon Rank Gap of a Third-Order Symmetric Tensor Can Exceed One
arXiv 2610.01660 · 2026-10-02事件链蒙特卡洛中的最优采样策略
Optimal sampling strategies in event-chain Monte Carlo
arXiv 2610.01659 · 2026-10-02随机随机游走是正常的吗？
Are random random walks normal?
arXiv 2610.01658 · 2026-10-02椭圆偏微分方程均质化中平均格林函数的正则性
On regularity of averaged Green's functions in homogenization of elliptic PDE
arXiv 2610.01656 · 2026-10-02MDIRNET：基于深度展开的多退化图像恢复网络
MDIRNET: Multi-Degradation Image Restoration Network via Deep Unfolding
arXiv 2610.01655 · 2026-10-02识别面板条件效应与刷新样本：尖锐边界与设计假设
Identifying Panel Conditioning with Refreshment Samples: Sharp Bounds and Design Assumptions
arXiv 2610.01654 · 2026-10-02通过语义回滚分析的迭代式策略精化
Iterative Policy Refinement through Semantic Rollout Analysis
arXiv 2610.01652 · 2026-10-02随机梯度Langevin动力学的理论保证
Theoretical guarantees for stochastic gradient Langevin dynamics
arXiv 2610.01651 · 2026-10-02在同态加密与差分隐私的联邦学习中结合模型检查与可用性
Combining Homomorphic Encryption and Differential Privacy in Federated Learning for Model Inspection and Availability
arXiv 2610.01650 · 2026-10-02CrossGMN：用于跨架构权重空间变换的图元网络
CrossGMN: Graph Metanetworks for Cross-Architecture Weight-Space Transformations
arXiv 2610.01649 · 2026-10-02可匹配半匹配的精确局部性差距
Exact Locality Gaps for Matchable Semi-Matchings
arXiv 2610.01648 · 2026-10-02迈向智能建筑的云-雾-边缘系统
Towards a Cloud Fog Edge System for Smart Building
arXiv 2610.01647 · 2026-10-02DIADA：数据湖中的自动数据组合
DIADA: Automatic Data Composition in Data Lakes
arXiv 2610.01646 · 2026-10-02SonarVoxNet：使用3D声纳在3D边界框中进行潜水员检测
SonarVoxNet: Diver Detection in 3D Bounding Box using 3D Sonar
arXiv 2610.01644 · 2026-10-02紧致Kähler曲面上Kähler Ricci流的有限时间奇点为第一类
Finite Time Singularities of the Kähler Ricci Flow on Compact Kähler Surfaces are of Type I
arXiv 2610.01643 · 2026-10-02线性环面自同态、精确性、伯努利性与生成元
Linear toral endomorphisms, exactness, Bernoulliness and generators
arXiv 2610.01642 · 2026-10-02MCIR：一种具有可靠性保证的基于特征依赖的可解释性方法
MCIR: A Feature Dependence-Aware Explainability Method with Reliability Guarantees
arXiv 2610.01641 · 2026-10-02并非所有错误都能通过规模解决：视觉-语言推理中规模化的极限
Not All Error Yields to Scale: Where Scaling Stops in Vision-Language Inference
arXiv 2610.01640 · 2026-10-02整数上3x3矩阵乘法的22次乘法下界
Lower Bound of 22 for 3x3 Matrix Multiplication over the Integers
arXiv 2610.01639 · 2026-10-02FedSAP：面向多域异构边缘设备的结构化自适应分区联邦学习
FedSAP: Federated Learning with Structured Adaptive Partitioning for Multi-Domain Heterogeneous Edge Devices
arXiv 2610.01638 · 2026-10-02融合视觉与文本表示的多层融合Transformer用于越南语视觉问答
Fusing Visual and Textual Representations via Multi-layer Fusing Transformers for Vietnamese Visual Question Answering
arXiv 2610.01637 · 2026-10-02化学反应系统中的自催化与边界稳定性
Autocatalysis and Boundary Stability in Chemical Reaction Systems
arXiv 2610.01636 · 2026-10-02低开销 IMU 辅助预测式波束管理用于多频段低轨卫星直连设备链路
Low Overhead IMU Assisted Predictive Beam Management for Multiband LEO Direct to Device Links
arXiv 2610.01635 · 2026-10-02Yo-ByT5：约鲁巴语的高效高保真变音符号恢复
Yo-ByT5: Efficient and High-Fidelity Diacritic Restoration for Yorùbá
arXiv 2610.01634 · 2026-10-02通过量级势视角理解神经网络的泛化
Generalization in Neural Networks Through the Lens of Magnitude Potential
arXiv 2610.01633 · 2026-10-02SL-RFSIM：在OpenAirInterface中实现可扩展的多跳5G NR Sidelink网状网络
SL-RFSIM: Enabling Scalable Multi-Hop 5G NR Sidelink Mesh Networking in OpenAirInterface
arXiv 2610.01632 · 2026-10-02光学LEO卫星网络中基于DRL的联合几何与QoS感知路由
Joint Geometric and QoS-Aware Routing in Optical LEO Satellite Networks via DRL
arXiv 2610.01631 · 2026-10-02合作学习之后：多智能体强化学习中的梯度路由与优化器依赖的维持
After Cooperation Is Learned: Gradient Routing and Optimizer-Dependent Maintenance in Multi-Agent Reinforcement Learning
arXiv 2610.01630 · 2026-10-02Janson-Sobolev 与 Janson-Sobolev-Hardy 空间上的对角算子
Diagonal operators on Janson-Sobolev and Janson-Sobolev-Hardy spaces
arXiv 2610.01628 · 2026-10-02什么使问题更难？用自然语言解释问题难度
What Makes Something Hard(er)? Explaining Question Difficulty in Natural Language
arXiv 2610.01627 · 2026-10-02测量动作分块背后的稳定性假设
Measuring the Stability Assumption Behind Action Chunking
arXiv 2610.01626 · 2026-10-02超越领域级自适应：面向个性化联邦视觉-语言模型的边际导向语义-外观交互校正
Beyond Domain-Level Adaptation: Margin-Oriented Semantic-Appearance Interaction Correction for Personalized Federated Vision-Language Models
arXiv 2610.01625 · 2026-10-02分子中的相对论性 Hirshfeld 原子：一种信息论视角，及其在 Drude 振子色散模型中的应用
Relativistic Hirshfeld atoms in a molecule: An information-theoretic view, with application to Drude oscillator dispersion models
arXiv 2610.01624 · 2026-10-02开源多线SPI读出接口用于可穿戴超声探头
Open-Source Multi-Wire SPI Readout for Wearable Ultrasound Probes
arXiv 2610.01623 · 2026-10-02Buchholz 代数从普适预解代数——粒子结构、Gibbs 态与无 Fock 表示的闭路展开
The Buchholz Algebra from the Universal Resolvent Algebra--Particle Structure, Gibbs States, and Closed-Path Expansions without Fock Representations
arXiv 2610.01622 · 2026-10-02学习分类环状聚合物熔体中的穿线程化
Learning to Classify Threading in Melts of Rings
arXiv 2610.01621 · 2026-10-02FedLore：通过共享梯度低秩投影实现通信与内存高效的联邦学习
FedLore: Communication and Memory Efficient Federated Learning via Shared Gradient Low-Rank Projection
arXiv 2610.01620 · 2026-10-02揭示深度学习音频开发的成本
Exposing the Cost of Deep Learning Audio Development
arXiv 2610.01619 · 2026-10-02智能体是系统，而非模型：重新思考智能体评估
Agents Are Systems, Not Models: Rethinking Agentic Evaluation
arXiv 2610.01618 · 2026-10-02两个链的乘积的负 Schur 系数
A negative Schur coefficient for products of two chains
arXiv 2610.01617 · 2026-10-02大型语言模型能否可靠地标注生物测定元数据以提升数据就绪度？
Can LLMs Reliably Annotate Bioassay Metadata to Improve Data Readiness?
arXiv 2610.01616 · 2026-10-02三维及以上空间中薛定谔方程Dirichlet解的消失阶
Vanishing orders of Dirichlet solutions to the Schrödinger equation in dimensions three and higher
arXiv 2610.01615 · 2026-10-02Oneira：从视频世界模型中的开放式生成到开放世界交互
Oneira: From Open-Ended Generation to Open-World Interaction in Video World Models
arXiv 2610.01614 · 2026-10-02从LZ事件到大统一：重Higgsino与质子衰变
From the LZ Event to Grand Unification: Heavy Higgsinos and Proton Decay
arXiv 2610.01613 · 2026-10-02ReCo：面向腿部操作的任务一致性运动与策略感知MPC
ReCo: Response-Consistent Locomotion with Policy-Aware MPC for Legged Manipulation
arXiv 2610.01612 · 2026-10-02
