2026-10论文中文摘要
RAPO-Sol：面向仓库级Solidity代码生成的检索增强偏好优化
RAPO-Sol: Retrieval-Augmented Preference Optimization for Repository-Level Solidity Code Generation
arXiv 2610.08429 · 2026-10-08HE-OFT：同态加密下的隐私保护一次性联邦微调
HE-OFT: Privacy-Preserving One-Shot Federated Fine-Tuning under Homomorphic Encryption
arXiv 2610.08255 · 2026-10-08OSFP4：NVFP4量化的对角平滑与块缩放联合优化
OSFP4: Joint Optimization of Diagonal Smoothing and Block Scales for NVFP4 Quantization
arXiv 2610.08231 · 2026-10-08寻找语言模型中负责网络信息检索的注意力头与神经元
Finding the Heads and the Neurons Responsible for Network Information Retrieval in Language Models
arXiv 2610.08200 · 2026-10-08社会物理学：宣言
Social Physics: A manifesto
arXiv 2610.08117 · 2026-10-08无限 Cuntz 代数的表示
Representations of the infinite Cuntz algebra
arXiv 2610.08094 · 2026-10-08基于控制动态优化的多样化运动定制
Diverse Motion Customization via Control-based Dynamic Optimization
arXiv 2610.07911 · 2026-10-08检索不够：为冻结的时间序列预测器刷新记忆
Retrieval Is Not Enough: Refreshing Memory for Frozen Time-Series Forecasters
arXiv 2610.07834 · 2026-10-08虚转动下QCD的分形相结构
Fractal phase structure of QCD under imaginary rotation
arXiv 2610.07779 · 2026-10-08OTel：开放电信人工智能数据集、基准与模型
OTel: Open Telco AI Datasets, Benchmarks, and Models
arXiv 2610.07766 · 2026-10-08某些中空涡与气泡流之间的数学类比——存在性、唯一性及相同边界形状
A mathematical analogy between certain hollow vortex and bubble flows - existence, uniqueness, and identical boundary shapes
arXiv 2610.07709 · 2026-10-08p-adic Langlands 对应中的块分解
Block decompositions in the p-adic Langlands correspondence
arXiv 2610.07648 · 2026-10-08异构偏好下基于显式人格推断的大语言模型编排
Large Language Model Orchestration under Heterogeneous Preferences via Explicit Persona Inference
arXiv 2610.07587 · 2026-10-08PLUTO：一种用于交互式航天器交会轨迹设计的智能体AI工具
PLUTO: An Agentic AI Tool for Interactive Spacecraft Rendezvous Trajectory Design
arXiv 2610.07573 · 2026-10-08量子达尔文主义通过装置介导的环境记录
Quantum Darwinism through apparatus-mediated environmental recording
arXiv 2610.07568 · 2026-10-08解耦式多智能体编排
Decoupled Multi-Agent Orchestration
arXiv 2610.07556 · 2026-10-08WarpDRF：面向Warp级编程的数据竞争自由
WarpDRF: The Unwritten Contract of GPU Warp Programming
arXiv 2610.07541 · 2026-10-08强紧基数之上的广义连续统假设
The generalized continuum hypothesis above a strongly compact cardinal
arXiv 2610.07537 · 2026-10-08ProCut：用于自主电外科组织解剖的概率切割拓扑
ProCut: Probabilistic Cutting Topology for Autonomous Electrosurgical Tissue Dissection
arXiv 2610.07515 · 2026-10-08COMPASS：在语言模型中定位推理所在之处
COMPASS: Finding Where Reasoning Lives in Language Models
arXiv 2610.07469 · 2026-10-08保护性扰动必须经受住缩放：针对恶意编辑的尺度鲁棒图像免疫
Protective Perturbations Must Survive the Resize: Scale-Robust Image Immunization against Malicious Editing
arXiv 2610.07464 · 2026-10-08ElasticFit：通过VLM推理与生成式适配实现适配感知的3D物体插入
ElasticFit: Fit-Aware 3D Object Insertion via VLM Reasoning and Generative Adaptation
arXiv 2610.07460 · 2026-10-08BRAHMa：基于机器学习的棒状结构识别与勾画——面向星系棒定向包围盒检测的框架
BRAHMa: Bar Recognition And Hatching using Machine learning -- a framework for oriented bounding box detection of galactic bars
arXiv 2610.07410 · 2026-10-08DeepAJM：面向不规则采样数据的深度关联联合模型
DeepAJM: Deep Association Joint Model for Irregularly Sampled data
arXiv 2610.07388 · 2026-10-08重新分配伤害：文件转换与印度身份体系中跨性别包容的局限
Redistributing Harm: Document Transition and the Limits of Trans Inclusion in India's Identity Systems
arXiv 2610.07343 · 2026-10-08豪华停车位
Posh Parking Spaces
arXiv 2610.07280 · 2026-10-08Fréchet Inception Distance 的样本最优估计
Sample-Optimal Estimation of the Fréchet Inception Distance
arXiv 2610.07114 · 2026-10-08表示感知模块度与距离校准零模型
Representation-Aware Modularity with a Distance-Calibrated Null Model
arXiv 2610.07088 · 2026-10-08演示：视觉-语言模型引导的电磁数字孪生在线校准
Demo: Vision-Language Model-Guided Online Calibration of an Electromagnetic Digital Twin
arXiv 2610.07081 · 2026-10-08ImpactMat：用于逆冲击声音渲染的连续材料估计
ImpactMat: Continuous Material Estimation for Inverse Impact Sound Rendering
arXiv 2610.07061 · 2026-10-08有限向量空间中的极值子空间覆盖
Extremal subspace covers in finite vector spaces
arXiv 2610.07051 · 2026-10-08树打包与图论加法基猜想
Tree packing and the graphic Additive Base Conjecture
arXiv 2610.07034 · 2026-10-08STOCK-JEPA：股票市场中先验锚定的潜在修正表示学习
STOCK-JEPA: Prior-Anchored Latent Revision Representation Learning in Equity Markets
arXiv 2610.07006 · 2026-10-08APEX：面向LLM智能体执行边界的主动防护
APEX: Active Protection at Execution Boundaries for LLM Agents
arXiv 2610.06966 · 2026-10-08单层1H-NbS$_2$的关联低能模型：cRPA+DMFT研究
Correlated Low-Energy Model of Monolayer 1H-NbS$_2$: A cRPA+DMFT Study
arXiv 2610.06912 · 2026-10-08高级持续性威胁与移动目标防御之间随机博弈的动态低秩均衡计算
Dynamical low-rank equilibrium computation for stochastic games between advanced persistent threats and moving target defense
arXiv 2610.06885 · 2026-10-08建筑热负荷短期预测混合模型比较综述
Comparative review of hybrid forecasting models for short-term prediction of building thermal load
arXiv 2610.06881 · 2026-10-08极小顺从 ample 群胚的几乎有限—纯无限二分性
The Almost Finite--Purely Infinite Dichotomy for Minimal Amenable Ample Groupoids
arXiv 2610.06713 · 2026-10-08直流耦合电阻硅探测器（DC-RSD）中电极间结构的设计优化
Design optimization of the inter-electrode structures in DC-RSD
arXiv 2610.06676 · 2026-10-08LoGRA：利用低秩梯度草图扩展大语言模型强化学习
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches
arXiv 2610.06647 · 2026-10-08DGA-Muon：解耦几何对齐的自适应缩放用于Muon
DGA-Muon: Decoupled Geometry-Aligned Adaptive Scaling for Muon
arXiv 2610.06578 · 2026-10-08优化优化器：语言模型发现更快的分子弛豫方法
Optimizing the Optimizer: Language Models Discover Faster Molecular Relaxation Algorithms
arXiv 2610.06577 · 2026-10-08三次非线性薛定谔方程在 $\dot{H}^{\frac12}(\Bbb R^3)$ 中的无条件唯一性
Unconditional uniqueness for the cubic nonlinear Schrödinger equation in $\dot{H}^{\frac12}(\Bbb R^3)$
arXiv 2610.06501 · 2026-10-08自由能上偏差处的TAP态：伊辛自旋玻璃中的存在性、局域化与边缘稳定性
TAP states at upper deviations of the free energy: existence, localization and marginal stability in Ising spin glasses
arXiv 2610.06490 · 2026-10-08Sherrington-Kirkpatrick自由能的上偏差：五次律与约束Parisi测度的起始
Upper deviations of the Sherrington-Kirkpatrick free energy: the fifth-order law and the onset of constrained Parisi measures
arXiv 2610.06480 · 2026-10-08评估LiDAR数据源、预测变量分辨率与贝叶斯变更支持模型中的空间随机效应对森林清查的影响
Evaluating LiDAR Data Sources, Predictor Resolution, and Spatial Random Effects in Bayesian Change-of-Support Models for Forest Inventory
arXiv 2610.06476 · 2026-10-08RAISED：通过自蒸馏提升LLM智能体对提示注入的鲁棒性
RAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
arXiv 2610.06401 · 2026-10-08缩小缩放定律：资源受限大型语言模型中的参数效率与计算最优训练
Scaling Down the Scaling Laws: Parameter Efficiency and Compute-Optimal Training in Resource-Constrained Large Language Models
arXiv 2610.06387 · 2026-10-08约束Wasserstein梯度流用于平均场Langevin动力学
A constrained Wasserstein gradient flow for mean-field Langevin dynamics
arXiv 2610.06343 · 2026-10-08具有粗糙非局部扩散的动力学方程基本解的存在性与上界
Existence and upper bounds of fundamental solutions to kinetic equations with rough non-local diffusion
arXiv 2610.06214 · 2026-10-08
