2026-10论文中文摘要
Merian巡天：亚银河系质量星系的明亮卫星统计普查
The Merian Survey: A Statistical Census of Bright Satellites of Sub-Milky Way-Mass Galaxies
arXiv 2610.03981 · 2026-10-06FADE：基于帧感知扩散Transformer的多概念擦除用于视频遗忘
FADE: Frame-Aware Diffusion-Transformer-based Multi-Concept Erasure for Video Unlearning
arXiv 2610.03980 · 2026-10-06量子普适性的基础谜题：从维格纳的朋友到黑洞
Foundational puzzles on quantum universality: From Wigner's friend to black holes
arXiv 2610.03979 · 2026-10-06学习潜在蛋白质语言用于自回归生成
Learning Latent Protein Languages for Autoregressive Generation
arXiv 2610.03978 · 2026-10-06活性手性介质中驱动粒子的输运、霍尔效应与集体动力学
Transport, Hall Effect, and Collective Dynamics of Driven Particles in Active Chiral Media
arXiv 2610.03977 · 2026-10-06$\u00a0\mathrm{GL}(2)$\u00a0上尖点形式的特殊\u00a0$L$\u00a0值在完全分裂素数幂扭下的非零性：无需类数假设
Nonvanishing of special $L$-values of cusp forms on $\mathrm{GL}(2)$ with totally split prime power twists: without the class number hypothesis
arXiv 2610.03976 · 2026-10-06超大规模人工智能数据中心噪声排放：社区健康研究门口的一个特征化不足的暴露问题
Noise Emissions from Hyperscale AI Data Centers: A Poorly Characterized Exposure at the Doorstep of Community Health Research
arXiv 2610.03975 · 2026-10-06最近的、最亮的碎片盘宿主恒星：具有饱和2MASS测光恒星的最新光球模型与尘埃性质
The Nearest, Brightest Debris Disk Host Stars: Updated Photosphere Models and Dust Properties for Stars with Saturated 2MASS Photometry
arXiv 2610.03974 · 2026-10-06等变生成扩散学习并泛化非晶氧化物的结构系综
Equivariant generative diffusion learns and generalizes the structural ensemble of amorphous oxides
arXiv 2610.03973 · 2026-10-06概率算法在Ising机器中的应用：从优化到生成式人工智能
Probabilistic Algorithms for Ising Machines from Optimization to Generative AI
arXiv 2610.03972 · 2026-10-06基于LLM的HLS修复的可执行基准：设计复杂度与修复欠约束
An Executable Benchmark for LLM-Based HLS Repair:Design Complexity and Repair Underconstraint
arXiv 2610.03971 · 2026-10-06SCRM：空间网络风险管理的一个可操作框架
SCRM: An Actionable Framework for Space Cyber Risk Management
arXiv 2610.03970 · 2026-10-06编码智能体如何优化软件并报告性能验证？开源拉取请求的大规模实证研究
How Do Coding Agents Optimize Software and Report Performance Validation? A Large-Scale Empirical Study of Open-Source Pull Requests
arXiv 2610.03969 · 2026-10-06度量流形与具有曲率界的度量空间的可定向性
Orientability of metric manifolds and metric spaces with curvature bounds
arXiv 2610.03968 · 2026-10-06基于SIC解码技术的RIS辅助多向通信系统
RIS Aided Multi-Way Communication Systems based on SIC Decoding Techniques
arXiv 2610.03967 · 2026-10-06ROAR：统一异构AI驱动研究系统中的运行数据
ROAR: Unifying Runs across Heterogeneous AI-Driven Research Systems
arXiv 2610.03966 · 2026-10-06里德伯介导的可调带宽微波-光转换
Rydberg-mediated microwave-to-optical transduction with tunable bandwidth
arXiv 2610.03965 · 2026-10-06IRS辅助混合RF/VLC系统中基于协作多智能体深度强化学习的网络适配
Network Adaptation in IRS-Aided Hybrid RF/VLC Systems Using Cooperative Multi-Agent DRL
arXiv 2610.03964 · 2026-10-06确定钙需求的临床-代谢方法：葡萄糖轴、PTH-维生素D信号与骨矿物质密度的相互作用
A Clinical--Metabolic Approach to Determining Calcium Requirements: Interplay of the Glucose Axis, PTH--Vitamin D Signalling and Bone Mineral Density
arXiv 2610.03963 · 2026-10-06BAIBAICHUCHU 在 NTCIR-19 FinArg-3 任务中的表现：何时能从投资者文本中预测最大可能利润？
BAIBAICHUCHU at the NTCIR-19 FinArg-3 Task: When Is Maximum Possible Profit Predictable from Investor Text?
arXiv 2610.03962 · 2026-10-06基于波数分割复用的非视距全息MIMO中不可分离散射分析
Analysis of Non-Separable Scattering in WDM-Based NLoS Holographic MIMO
arXiv 2610.03961 · 2026-10-06定位与检测：审计指令微调模型中的任务级投毒
Localize-and-Detect: Auditing Task-Level Poisoning in Instruction-Tuned Models
arXiv 2610.03960 · 2026-10-06LatentQuant：在NVFP4 VAE量化下保持面向策略的潜在契约
LatentQuant: Preserving the Policy-Facing Latent Contract under NVFP4 VAE Quantization
arXiv 2610.03959 · 2026-10-06FICO：面向语料库级电子表格问答的“先查找后计算”方法
FICO: Find-Then-Compute for Corpus-Level Spreadsheet Question Answering
arXiv 2610.03958 · 2026-10-06关于叶状最优传输
On Foliated Optimal Transport
arXiv 2610.03957 · 2026-10-06基于潜在分数的贝叶斯克拉美-罗界估计用于高维成像系统
Latent Score-Based Bayesian Cramér-Rao Bound Estimation for High-Dimensional Imaging Systems
arXiv 2610.03956 · 2026-10-06自适应协同服务：现代推理引擎上的大语言模型水印技术
Adaptive Co-Serving LLM Watermarking on Modern Inference Engines
arXiv 2610.03955 · 2026-10-06在重子数破坏重标量框架下重新诠释ATLAS双喷注与超对称搜索
Reinterpreting ATLAS Dijet and Supersymmetry Searches in A Baryon-Number-Violating Heavy Scalar Framework
arXiv 2610.03954 · 2026-10-06从软件工程视角刻画开源视频游戏
Characterizing Open-Source Video Games from a Software Engineering Perspective
arXiv 2610.03953 · 2026-10-06一步曲率探针遗漏拟合算子：持续学习中的保留容量与终端零空间校正
One-Step Curvature Probes Miss the Fitting Operator: Retained Capacity and Terminal Null-Space Correction for Continual Learning
arXiv 2610.03952 · 2026-10-06演化LLM生成特征用于可解释分类
Evolving LLM-Generated Features for Interpretable Classification
arXiv 2610.03951 · 2026-10-06一族三维超共形场论的有效中心荷精确公式
Exact effective central charges for a family of 3d superconformal field theories
arXiv 2610.03950 · 2026-10-06最大独立集问题中Pauli关联编码的拟设基准测试
Benchmarking Ansatze for Pauli Correlation Encoding in the Maximum Independent Set Problem
arXiv 2610.03949 · 2026-10-06检索增强的大语言模型自动驾驶决策：以中国哲学智慧为引导
Retrieval-Augmented Large Language Model Decision-Making for Autonomous Driving Guided by Chinese Philosophical Wisdom
arXiv 2610.03948 · 2026-10-06用Transformer合成物理公式
Synthesizing Physics Formulae with Transformers
arXiv 2610.03947 · 2026-10-06重子印记对暗物质晕的影响：利用CAMELS刻画浓度-质量完整概率分布
Baryonic Imprints on DM Halos: characterizing the full concentration-mass probability distribution with CAMELS
arXiv 2610.03946 · 2026-10-06糖萼驱动的微囊泡形成动力学
Dynamics of microvesicle formation driven by glycocalyx
arXiv 2610.03944 · 2026-10-06黑洞自旋与轴对称发射几何的偏振测量
Polarimetry of Black Hole Spin and Axisymmetric Emission Geometry
arXiv 2610.03943 · 2026-10-06动态时间步预测在类模糊图像恢复任务的逆热扩散中的应用
Dynamic Time Step Prediction in Inverse Heat Dissipation for Blur-Like Image Restoration Tasks
arXiv 2610.03942 · 2026-10-06球对称树上的伊辛模型：分支间隙、共存与边界影响
Branching Gaps, Coexistence, and Boundary Influence for the Ising Model on Spherically Symmetric Trees
arXiv 2610.03941 · 2026-10-06遗忘的一步：优化器历史与答案质量的损失
A Step Towards Forgetting: Optimiser History and the Loss of Answer Mass
arXiv 2610.03940 · 2026-10-06TreeWalker：分组树集成推理的偏特化求值
TreeWalker: Partial Evaluation for Grouped Tree-Ensemble Inference
arXiv 2610.03939 · 2026-10-06多模态大语言模型在智能体化使用工具时无法拒绝有害请求
MLLMs Fail to Refuse when Using Tools Agentically
arXiv 2610.03938 · 2026-10-06聚合关系数据中的社区检测：可辨识性与谱恢复
Community Detection from Aggregated Relational Data: Identifiability and Spectral Recovery
arXiv 2610.03937 · 2026-10-06具有涌现伊辛转变的吸收态模型
An Absorbing State Model with an Emergent Ising Transition
arXiv 2610.03936 · 2026-10-06通用决策模型：超越Jev的基准测试与洞见
General Decision Models: Benchmarking and Insights Beyond Jev
arXiv 2610.03935 · 2026-10-06SGAnalog：来自开源硅片流片的端到端电路基准
SGAnalog: An End-to-End Circuit Benchmark from Open-Source Silicon Tapeouts
arXiv 2610.03934 · 2026-10-06一种用于细纤维热特性表征的新型迭代拟合方法
A Novel Iterative-Fitting Approach for the Thermal Characterization of Thin Fibers
arXiv 2610.03933 · 2026-10-06治理、映射、度量、吸收：公共部门参与式人工智能治理中的可读性陷阱
Govern, Map, Measure, Absorb: The Legibility Trap in Public Sector Participatory AI Governance
arXiv 2610.03932 · 2026-10-06基于稀疏校准的核方法步态相位与速度估计的可穿戴IMU个性化
Sparse Calibration-Based Personalization of Kernel-Based Gait Phase and Speed Estimation Using Wearable IMUs
arXiv 2610.03931 · 2026-10-06
