2026-10论文中文摘要
面向安全关键强化学习的统一贝尔曼算子
A Unified Bellman Operator for Safety-Critical Reinforcement Learning
arXiv 2610.12420 · 2026-10-09OneSearch-VL：面向图像与视频的统一多模态深度研究智能体
OneSearch-VL: Unified Multimodal Deep Research Agent for Image and Video
arXiv 2610.12419 · 2026-10-09宇宙拓扑学。第 IId 部分：透镜空间的本征模式与相关矩阵
Cosmic topology. Part IId. Eigenmodes and correlation matrices of lens spaces
arXiv 2610.12418 · 2026-10-09WOVEN：将视觉世界建模融入多模态大语言模型
WOVEN: Weaving Visual World Modeling into Multimodal LLMs
arXiv 2610.12417 · 2026-10-09MAMHOI：通过可供性分解场景感知的人-物交互
MAMHOI: Factorizing Scene-Aware Human-Object Interaction through Affordances
arXiv 2610.12416 · 2026-10-09ORCAGen：结合RAG引导的生成式AI编排上下文感知的恶意软件欺骗策略
ORCAGen: Orchestrating Context-Aware Malware Deception with RAG-Guided Generative AI
arXiv 2610.12415 · 2026-10-09N=4超杨-米尔斯理论大电荷极限下的能量-能量关联子
Energy-energy correlators in the large-charge limit of $\mathcal{N}=4$ super Yang-Mills theory
arXiv 2610.12414 · 2026-10-09基于高斯信道的带确定性识别码的承诺方案
Commitment over Gaussian channels with deterministic identification codes
arXiv 2610.12413 · 2026-10-09WorldCast：分布式多人世界模型
WorldCast: Distributed Multiplayer World Models
arXiv 2610.12412 · 2026-10-09GLIO2：一种用于鲁棒实时全局定位与建图的GPU并行紧耦合激光雷达-惯性-全球导航卫星系统（GNSS）系统
GLIO2: A GPU-Parallelized Tightly-Coupled LiDAR-Inertial-GNSS System for Robust and Real-Time Global Localization and Mapping
arXiv 2610.12411 · 2026-10-09基于价值表征的对齐泛化预测
Predicting Alignment Generalization with Value Representations
arXiv 2610.12410 · 2026-10-09探寻“有害弃权”：一项对AI安全基准的心理测量学审计
Searching for "Harmful Refusal": A Psychometric Audit of an AI Safety Benchmark
arXiv 2610.12409 · 2026-10-09基于 Lund 聚类树的紧凑可查询树张量网络喷注标记器
A compact, interrogable Tree Tensor Network Jet Tagger on Lund declustering trees
arXiv 2610.12408 · 2026-10-09LeWAM：一种基于扩散引导模型预测控制（MPC）的JEPA世界动作模型
LeWAM: A JEPA World Action Model with Diffusion-Steering-Based MPC
arXiv 2610.12407 · 2026-10-09光镊阵列的局域偏振控制
Local polarization control of optical tweezer arrays
arXiv 2610.12406 · 2026-10-09OctoSense：构建开源无线感知的统一生态系统
OctoSense: Building a Unified Ecosystem for Open-Source Wireless Sensing
arXiv 2610.12405 · 2026-10-09面向协作机器人运动生成的物理信息碰撞学习框架
A Physics-Informed Collision Learning Framework for Collaborative Robot Motion Generation
arXiv 2610.12404 · 2026-10-09ViSkill：用不断演进的视觉原生技能增强视觉语言模型（VLM）智能体
ViSkill: Reinforcing VLM Agents with Evolving Visual-Native Skills
arXiv 2610.12403 · 2026-10-09SpaceCast-Bench：评估视觉语言模型中的预测空间推理能力
SpaceCast-Bench: Evaluating Predictive Spatial Reasoning in Vision-Language Models
arXiv 2610.12402 · 2026-10-09基于全球-区域对齐的公里级天气预报学习
Learning Kilometer-Scale Weather Prediction with Global-Regional Alignment
arXiv 2610.12401 · 2026-10-09先测量，后缓解：逻辑量子电路的虚拟误差抵消
Measure Now, Mitigate Later: Virtual Error Cancellation for Logical Quantum Circuits
arXiv 2610.12400 · 2026-10-09SpaceFlow：可局部控制的3D生成
SpaceFlow: Locally Controllable 3D Generation
arXiv 2610.12399 · 2026-10-09ACORN I. 宇宙学模拟中来自星团的大质量黑洞种子与潮汐瓦解事件
ACORN I. Massive Black Hole Seeding and Tidal Disruption Events from Star Clusters in Cosmological Simulations
arXiv 2610.12398 · 2026-10-09基于源侧训练动态的OOD退化前瞻性预测
Prospective Prediction of OOD Degradation from Source-Side Training Dynamics
arXiv 2610.12397 · 2026-10-09用于拓扑感知金融决策的Mixup条形码
Mixup Barcodes for Topology-Aware Financial Decision Making
arXiv 2610.12396 · 2026-10-09一种神经特征映射方法：基于流映射的拉格朗日物理信息神经网络（PINNs），用于输运主导问题
A neural characteristic mapping method: Lagrangian PINNs based on flow maps for transport-dominated problems
arXiv 2610.12395 · 2026-10-09流入驱动的星系演化——II：恒星形成星系标度关系中的层级结构
Inflow-driven galaxy evolution - II: A hierarchy in the scaling relations of star-forming galaxies
arXiv 2610.12394 · 2026-10-09HRIL：通过高阶张量建模学习多模态协同
HRIL: Learning Multimodal Synergy via Higher-Order Tensor Modeling
arXiv 2610.12393 · 2026-10-09单个多立方体的平移平铺三维空间是不可判定的
Tiling 3D by Translates of a Single Polycube is Undecidable
arXiv 2610.12392 · 2026-10-09GeoReform：面向多模态几何问题求解的反思式形式化演化框架
GeoReform: Reflective Formalization Evolution for Multimodal Geometry Problem Solving
arXiv 2610.12391 · 2026-10-09用于预测特征的长文本：基于可执行程序搜索的大语言模型引导分块特征工程
Long Text to Predictive Features: LLM-Guided Blockwise Feature Engineering via Executable Program Search
arXiv 2610.12390 · 2026-10-09高度相对论真空气泡碰撞产生的引力波的成对代理模型
A Pairwise Surrogate for Gravitational-Wave Spectra from Highly Relativistic Vacuum Bubble Collisions
arXiv 2610.12389 · 2026-10-09GenIA：基于测试时输入对齐的生成式重建
GenIA: Generative Reconstruction with Test-Time Input Alignment
arXiv 2610.12388 · 2026-10-09二维等温可压缩Navier-Stokes方程的端点紧性与有限能量弱解
Endpoint Compactness and Finite-Energy Weak Solutions for the 2D Isothermal Compressible Navier-Stokes
arXiv 2610.12387 · 2026-10-09ARC：机器人基础模型的推理方案
ARC: A Reasoning Recipe for Robot Foundation Models
arXiv 2610.12386 · 2026-10-09(1+1)维$\boldsymbol{\rm CP^1}$模型中的柏拉图自对偶孤子
Platonic self-dual solitons in a (1+1)-dimensional $\mathbb{CP}^1$ model
arXiv 2610.12385 · 2026-10-09曲线模空间上的Betti自守范畴的局部常性
Local Constancy of the Betti Automorphic Category over the Moduli of Curves
arXiv 2610.12384 · 2026-10-09微分$T_2$-对偶与带联络的主3-丛的张景
Differential $T_2$-Duality and Spans of Principal 3-Bundles with Connections
arXiv 2610.12383 · 2026-10-09WorldAlign：用于世界一致性视频生成的解耦4D奖励
WorldAlign: Decoupled 4D Reward for World-Consistent Video Generation
arXiv 2610.12382 · 2026-10-09受控条件下逻辑回归模型的数据生成
Data generation for the logistic regression model under controlled conditions
arXiv 2610.12381 · 2026-10-09面向高空平台站（HAPS）的感知干扰下行功率控制：兼顾服务质量（QoS）与最大最小公平性
Interference-Aware Downlink Power Control for High-Altitude Platform Stations with QoS and Max-Min Fairness
arXiv 2610.12380 · 2026-10-09Marformer：一种用于预测缺失数据分布的Transformer模型
Marformer: A Transformer for Predicting Missing Data Distributions
arXiv 2610.12379 · 2026-10-09具有完美完备性的4对1游戏的困难性
On the Hardness of $4$-to-$1$ Games with Perfect Completeness
arXiv 2610.12378 · 2026-10-09clone-censor-weight（CCW）的形式化与一致性
Formalization and consistency of clone-censor-weight
arXiv 2610.12377 · 2026-10-09潜在核心分词器：有意义地进行压缩
Latent Core Tokenizer: Compress, but Meaningfully
arXiv 2610.12376 · 2026-10-09OnTrack：基于流结构感知最优传输的LLM智能体轨迹实时监控与干预
OnTrack: Real-Time Monitoring and Intervention in LLM Agent Trajectories via Streaming Structure-Aware Optimal Transport
arXiv 2610.12375 · 2026-10-09AgentGarten：用于智能体演化的代码世界
AgentGarten: Code Worlds for Evolving Agents
arXiv 2610.12374 · 2026-10-09味丰富SMEFT中的LFV：六维跑动与八维混合
LFV in flavourful SMEFT: Dimension-Six Running versus Dimension-Eight Mixing
arXiv 2610.12373 · 2026-10-09高维旋转时空：真空背景与黑洞
Swirling spacetimes in higher dimensions: Vacuum backgrounds and black holes
arXiv 2610.12372 · 2026-10-09FeatureZ：基于逐点界与星形分类的特征保留型压缩通用框架
FeatureZ: A General Framework for Feature-Preserving Compression via Pointwise Bounds and Star Classification
arXiv 2610.12371 · 2026-10-09
