2026-10论文中文摘要
EVISKILL：在可重放证据中实现技能演化
EVISKILL: Grounding Skill Evolution in Replayable Evidence
arXiv 2610.05030 · 2026-10-06SPACE-CLIPv2：从冻结的CLIP解码局部几何用于单目深度估计
SPACE-CLIPv2: Decoding Local Geometry from Frozen CLIP for Monocular Depth Estimation
arXiv 2610.05029 · 2026-10-06基于贝叶斯马尔可夫链蒙特卡洛的城市主干道异质交通同步速度预测模型
Bayesian Markov Chain Monte Carlo-Based Simultaneous Speed Prediction Model for Heterogeneous Traffic on Urban Primary Roads
arXiv 2610.05027 · 2026-10-06GeoBridge-VLA：用于视觉-语言-动作模型的几何感知残差适配
GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models
arXiv 2610.05026 · 2026-10-06通过预测不确定性触发通用推理的双系统视觉-语言-动作模型
Triggering Generalist Reasoning via Predictive Uncertainty for Dual-System VLA
arXiv 2610.05025 · 2026-10-06LightVLN：具有紧凑记忆与历史引导局部聚合的高效空中视觉语言导航
LightVLN: Efficient Aerial Vision-and-Language Navigation with Compact Memory and History-Guided Local Aggregation
arXiv 2610.05024 · 2026-10-06看向你所说之处：用于视觉推理的自接地注意力
Look Where You Say You're Looking: Self-Grounded Attention for Visual Reasoning
arXiv 2610.05023 · 2026-10-06扩展黑洞热力学中带辅助规范场的热力学体积
Thermodynamic Volumes of Extended Black Hole Thermodynamics with Auxiliary Gauge Fields
arXiv 2610.05022 · 2026-10-06两块奇数分拆函数的二元结构及 $pod(n)$ 的 theta 加权同余式
Dyadic Structure of the Two-Block Odd Partition Function and Theta-Weighted Congruences for $pod(n)$
arXiv 2610.05020 · 2026-10-06基于环境反向散射的ISAC中的联合目标-标签估计
Joint Target-Tag Estimation in Ambient Backscatter-Based ISAC
arXiv 2610.05019 · 2026-10-06$\mathfrak{g}_2$ 的交换概形
The commuting scheme of $\mathfrak{g}_2$
arXiv 2610.05018 · 2026-10-06基于拓扑表示的神经网络概念的操作抽象
Operational Abstractions of Neural Network Concepts via Topological Representations
arXiv 2610.05017 · 2026-10-06一种采用28纳米CMOS工艺、通过编程多级状态进行偏移校准的FeFET电压-时间转换器
A FeFET Voltage-to-Time Converter with Offset Trim by Programmed Multilevel State in 28-nm CMOS
arXiv 2610.05016 · 2026-10-06距离集的勒贝格测度与$L^2$球膨胀
Lebesgue measure of distance sets and $L^2$ ball inflation
arXiv 2610.05015 · 2026-10-06MetaKernelBench：衡量超越代码的GPU内核知识迁移
MetaKernelBench: Measuring GPU Kernel Knowledge Transfer Beyond Code
arXiv 2610.05014 · 2026-10-06EMBER-Bench：长时程具身任务中跨事件因果记忆的基准测试
EMBER-Bench: Benchmarking Cross-Event Causal Memory in Long-Horizon Embodied Tasks
arXiv 2610.05013 · 2026-10-06将量子汇编反编译为结构化程序
Decompiling Quantum Assembly into Structured Programs
arXiv 2610.05012 · 2026-10-06PortraitAes：意图条件化的结构化人像美学评估
PortraitAes: Intent-Conditioned Structured Portrait Aesthetics Assessment
arXiv 2610.05010 · 2026-10-06重新思考智能体根因分析的工具设计：一项受控实证研究
Rethinking Tool Design for Agentic RCA: A Controlled Empirical Study
arXiv 2610.05009 · 2026-10-06异步移动机器人的最小-最大均匀圆编队
Min-Max Uniform Circle Formation by Asynchronous Mobile Robots
arXiv 2610.05008 · 2026-10-06迈向轻量化空中5G gNB：在商用ARM平台上可复现的OAI测试平台
Toward Lightweight Aerial 5G gNBs: Reproducible OAI Testbed on Commodity ARM Platforms
arXiv 2610.05007 · 2026-10-06在协变量密度有界条件下期望条件协方差的最小最大估计
Minimax estimation of the expected conditional covariance under bounds on the covariate density
arXiv 2610.05006 · 2026-10-06任意相位CZ门及跨空间分离超导节点间的远程读取
Arbitrary-phase CZ gate and remote readout across spatially separated superconducting nodes
arXiv 2610.05005 · 2026-10-06物理增强的图变换器用于贴片天线正演与逆向设计
Physics-Augmented Graph Transformers for Patch-Antenna Forward and Inverse Design
arXiv 2610.05004 · 2026-10-06角案例有多角？用于高速公路场景生成的百分位控制
How corner is a corner case? Percentile control for highway scenario generation
arXiv 2610.05003 · 2026-10-06多位铁电NAND用于高通量大规模数据库搜索
Multi-bit Ferroelectric-NAND for High-throughput Massive Database Search
arXiv 2610.05002 · 2026-10-06基于大语言模型的测试生成：信息源、生成策略与质量证据
LLM-Based Test Generation: Information Sources, Generation Strategies, and Quality Evidence
arXiv 2610.05001 · 2026-10-06VisualErase：用于文本到图像扩散模型中稳健概念擦除的双分支视觉轨迹重定向
VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models
arXiv 2610.05000 · 2026-10-06超越过参数化：利用主动查询可证明地学习输入凸多层多项式网络
Beyond Overparameterization: Provable Learning of Input-Convex Multi-Layer Polynomial Networks with Active Queries
arXiv 2610.04999 · 2026-10-06EmoRSS：缓解大型语言模型中情绪引发的过度拒绝
EmoRSS: Mitigating Emotion-Induced Over-Refusal in Large Language Models
arXiv 2610.04998 · 2026-10-06悲观极小极大学习用于单侧覆盖下的公私信息博弈
Pessimistic Minimax Learning for Public-Private Information Games under Unilateral Coverage
arXiv 2610.04997 · 2026-10-06时间反转在哈密顿量性质测试中的威力
The power of time reversal in Hamiltonian property testing
arXiv 2610.04996 · 2026-10-06通过预共享纠缠盲远程实现任意两量子比特酉操作
Blind Remote Implementation of Arbitrary Two-Qubit Unitary Operation via Pre-Shared Entanglement
arXiv 2610.04995 · 2026-10-06tau不变量的均匀正性
Uniform positivity of the tau invariant
arXiv 2610.04994 · 2026-10-06余数集合的误差项估计
Error Term Estimates for Sets of Remainders
arXiv 2610.04993 · 2026-10-06ContractLens：用于恶意智能合约检测的潜在安全知识
ContractLens: Latent Security Knowledge for Malicious Smart Contract Detection
arXiv 2610.04992 · 2026-10-06CIPO：用于自适应工具粒度选择的反事实想象策略优化
CIPO: Counterfactual Imagination Policy Optimization for Adaptive Tool Granularity Selection
arXiv 2610.04991 · 2026-10-06在支撑位买入：多偏斜布朗运动下的入场问题
Buying at Support: The Entry Problem under Multi-Skew Brownian Motion
arXiv 2610.04990 · 2026-10-06图何时是协方差？协方差不确定性下的贝叶斯残差传播
When Is a Graph a Covariance? Bayesian Residual Propagation under Covariance Uncertainty
arXiv 2610.04989 · 2026-10-06量子凸体体积估计的改进上下界
Improved Upper and Lower Bounds for Quantum Convex-Body Volume Estimation
arXiv 2610.04987 · 2026-10-06候选 O I 7774 Å 吸收：来自 BD+05 4868 Ab 的逃逸岩石蒸气外流
Candidate O I 7774 Å Absorption from an Escaping Rock-Vapor Outflow of BD+05 4868 Ab
arXiv 2610.04986 · 2026-10-06Jev的隐藏风险：安全、隐私与双重用途的实证研究
Hidden Risks of Jev: An Empirical Study of Security, Privacy, and Dual Use
arXiv 2610.04985 · 2026-10-06非线性诱导的手性对称一维系统中的拓扑相变：实空间处理
Nonlinearity-induced Topological Phase Transitions in One-dimensional Systems with Chiral Symmetry: A Real-Space Treatment
arXiv 2610.04984 · 2026-10-06黑洞周围磁流体动力学的时空微分同胚方法：膨胀协变性、线性化与无毛极限
A Spacetime-Diffeomorphic Approach to Magnetohydrodynamics around Black Holes: Dilation Covariance, Linearization,and the No-Hair Limit
arXiv 2610.04983 · 2026-10-06控制政策与经济体制的非线性财政乘数
Nonlinear fiscal multiplier controlling for policy and economic regimes
arXiv 2610.04982 · 2026-10-06TeleGen：通过运行时遥测改进基于LLM的Web应用生成
TeleGen: Improving LLM-Based Web Application Generation via Runtime Telemetry
arXiv 2610.04981 · 2026-10-06面向文本到图像扩散模型的几何感知偏好优化
Geometry-Aware Preference Optimization for Text-to-Image Diffusion Models
arXiv 2610.04980 · 2026-10-06基于反射卷积核范数最小化的鲁棒张量补全
Robust Tensor Completion via Reflective Convolution Nuclear Norm Minimization
arXiv 2610.04979 · 2026-10-06后训练将修复什么？逐问题增益在独立强化学习运行间共享，且现有检查点比先验信号更能预测它们
What Will Post-Training Fix? Per-Problem Gains Are Shared Across Independent RL Runs, and Existing Checkpoints Predict Them Better Than A Priori Signals
arXiv 2610.04978 · 2026-10-06大型语言模型是否理解序列结构？一项关于推理与生成的受控研究
Do LLMs Understand Sequential Structure? A Controlled Study of Inference and Generation
arXiv 2610.04977 · 2026-10-06
