2026-10论文中文摘要
绝热路径几何的界：基于能隙剖面的宽度类
Bounds on adiabatic path geometry from the width class of the gap profile
arXiv 2609.39286 · 2026-10-02RW-Flow：通过Wasserstein梯度流在紧致流形上的一步生成
RW-Flow: One-Step Generation on Compact Manifolds via Wasserstein Gradient Flows
arXiv 2609.39271 · 2026-10-02更信任评论家
Trust the Critic More
arXiv 2609.39247 · 2026-10-02爱因斯坦-幻影标量-高斯-博内理论中的非线性标量化与热力学
Nonlinear scalarization and thermodynamics in Einstein-Phantom scalar-Gauss-Bonnet theory
arXiv 2609.39236 · 2026-10-02QATFactory：面向大规模语言模型量化感知训练与蒸馏的通用、部署对齐框架
QATFactory: A Versatile, Deployment-Aligned Framework for Quantization-aware Training and Distillation of LLMs
arXiv 2609.39223 · 2026-10-02基于极限学习机的障碍函数综合
Fast and Sample Efficient Safety Verification via Extreme Learning Machine
arXiv 2609.39221 · 2026-10-02常步长SGD的锐利平稳高斯逼近
Sharp Stationary Gaussian Approximation for Constant-Stepsize SGD
arXiv 2609.39144 · 2026-10-02双变量自行车码与元校验：综合征修复、测量故障模糊性与逻辑障碍
Metachecks in Bivariate Bicycle Codes: Syndrome Distance, Measurement Faults, and Repair Limits
arXiv 2609.39121 · 2026-10-02Muon 中的行归一化之谜
The Row Normalization Puzzle in Muon
arXiv 2609.39114 · 2026-10-02通过错误发现率控制的视觉数据分割缓解大型视觉语言模型中的物体幻觉
Mitigating Object Hallucination in Large Vision-Language Models via False Discovery Controlled Visual Data Splitting
arXiv 2609.38979 · 2026-10-02私密币论证的紧致并行重复
Tight Post-Quantum Parallel Repetition for Private-Coin Arguments
arXiv 2609.38941 · 2026-10-02边界秩障碍与耗散平带制备中的测量辅助恢复
Boundary-rank obstructions and measurement-assisted recovery in dissipative flat-band preparation
arXiv 2609.38933 · 2026-10-02scTrilemma：单细胞表示学习中身份、不变性与保真度的平衡
scTrilemma: Balancing Identity, Invariance, and Fidelity in Single-Cell Representation Learning
arXiv 2609.38840 · 2026-10-02伪紧型度量图上的因子与调和态射
Divisors and harmonic morphisms on metric graphs of pseudocompact type
arXiv 2609.38821 · 2026-10-02黑板智能在全局约束问题上可超越自回归模型
Blackboard Intelligence Can Surpass Autoregressive on Globally Constrained Problems
arXiv 2609.38806 · 2026-10-02使用高阶调制相干态纠缠原子量子存储器
Entangling Atomic Quantum Memories Using High-Order Modulated Coherent States
arXiv 2609.38804 · 2026-10-02dattri-LLM：一个统一且高效的训练数据归因库，适用于LLM规模
dattri-LLM: A Unified and Efficient Library for Training Data Attribution at LLM Scale
arXiv 2609.38767 · 2026-10-02列表搜索的量子查询复杂度
Quantum Query Complexity for List Search
arXiv 2609.38736 · 2026-10-02打破巴别塔：用于长篇字幕翻译的自进化多智能体系统
Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle Translation
arXiv 2609.38660 · 2026-10-02具有多个最优臂的Bandits：极小极大遗憾与非适应性
Bandits with Multiple Optimal Arms: Minimax Regret and Non-Adaptivity
arXiv 2609.38659 · 2026-10-02探索基于生成式建模的论坛帖子检索
Exploring Forum Post Retrieval with Generative Modeling
arXiv 2609.38646 · 2026-10-02通过对探针进行训练实现对齐而不丧失可监控性
Alignment via Training Against Probes Without Losing Monitorability
arXiv 2609.38645 · 2026-10-02StreamDecisionBench：评估演化语言流上的生效决策
StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams
arXiv 2609.38612 · 2026-10-02基于高斯过程和多层网络的贝叶斯正例-未标注学习欺诈检测
Fraud Detection via Bayesian Positive-Unlabeled Learning with Gaussian Processes and Multilayer Networks
arXiv 2609.38586 · 2026-10-02可学习展平实现到多样骨骼的运动重定向
Retargeting Motions to Diverse Skeletons via Learnable Flattening
arXiv 2609.38578 · 2026-10-02关于c-多项式分解
On c-polynomial factorisations
arXiv 2609.38567 · 2026-10-02MOBA-VL：面向实时MOBA解说的事件定位多轮强化学习
MOBA-VL: Event-Localized Multi-Turn Reinforcement Learning for Real-Time MOBA Commentary
arXiv 2609.38428 · 2026-10-02Halluscoring 2026：关于大语言模型幻觉检测与答案验证的首个共享任务
Halluscoring 2026: The first shared task on llms hallucination detection and answer verification
arXiv 2609.38355 · 2026-10-02TAGGRAPH：用于智能体持久历史图检索的标签增强图
TAGGRAPH: Tag-Augmented Graphs for Graph Retrieval of Agent Persistent Histories
arXiv 2609.38353 · 2026-10-02EVOKE：在智能体中激发世界知识以实现可迁移决策
EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making
arXiv 2609.38334 · 2026-10-02使用Jackknife方法检验强透镜模型中星系团尺度暗物质晕与最亮星系团星系之间的偏移
Testing Offsets Between Cluster-Scale Halos and BCGs in Strong Lensing Models Using the Jackknife Method
arXiv 2609.38310 · 2026-10-02紧密脉冲星对通过引力波背景观测暗物质亚结构
Close Pulsar Pairs See Dark Matter Substructure through the Gravitational-Wave Background
arXiv 2609.38305 · 2026-10-02Zero2Repo：编码智能体能从零构建代码仓库吗？
Zero2Repo: Can Coding Agents Build Repositories from Scratch?
arXiv 2609.38269 · 2026-10-02SURE：构建可信人工智能的安全框架
SURE: Framework for Safety to Construct Trustworthy AI
arXiv 2609.38249 · 2026-10-02XENONnT $pp$主导太阳中微子约束$\nu_1$赝狄拉克劈裂
XENONnT $pp$-Dominated Solar Neutrinos Constrain the $ν_1$ Pseudo-Dirac Splitting
arXiv 2609.38233 · 2026-10-02优化、学习、精炼：螺旋软体机器人的全身抓取与拾取-投掷
Optimize, Learn, Refine: Whole-Body Grasping and Pick-and-Throw with a Spiral Soft Robot
arXiv 2609.38202 · 2026-10-02随机系统中物质拓扑相的分类
Classification of topological phases of matter in stochastic systems
arXiv 2609.38171 · 2026-10-02学习测试时AI4AI智能体框架设计中的元技能
Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI
arXiv 2609.38143 · 2026-10-02正截面曲率的奇异$8$-球面与三阶同伦$10$-球面
Positive sectional curvature on the exotic $8$-sphere and order-three homotopy $10$-spheres
arXiv 2609.38126 · 2026-10-02减速时空中平面对称流体的尖锐寿命结果
Sharp lifespan results for plane-symmetric fluids on decelerated spacetimes
arXiv 2609.38118 · 2026-10-02延迟物理系统的驱动因素与动力学的可辨识性保证
Identifiability Guarantees for Drivers and Dynamics of Delayed Physical Systems
arXiv 2609.37944 · 2026-10-02ByteTraX：通过优化阈值增强ByteTrack架构
ByteTraX: Enhancing the ByteTrack Architecture with Optimised Thresholding
arXiv 2609.37801 · 2026-10-02一种用于评估大语言模型回答中表达性临床推理的评分量规提案
A Proposed Rubric for Evaluating Expressed Clinical Reasoning in Large Language Model Responses
arXiv 2609.37788 · 2026-10-02Honeycomb：视频世界模型的恒定大小场景记忆表示
Honeycomb: Constant-Size Scene Memory Representation for Video World Models
arXiv 2609.37690 · 2026-10-02纹理空间材质扩散
Texture Space Material Diffusion
arXiv 2609.37654 · 2026-10-02BCS-BEC 交叉处的量子统计热机
Quantum Statistical Thermal Engine at the BCS-BEC crossover
arXiv 2609.37592 · 2026-10-02Fano 轨道簇允许自由 Campana 曲线
Fano orbifolds admit free Campana curves
arXiv 2609.37513 · 2026-10-02在策略状态空间中从互联网视频学习社交导航
Learning Social Navigation from Internet Videos in the Policy State Space
arXiv 2609.37476 · 2026-10-02码本引导的跨模态知识蒸馏用于结构异构特征
Codebook-Guided Cross-Modal Knowledge Distillation for Structurally Heterogeneous Features
arXiv 2609.37243 · 2026-10-02SafeVantage：面向可靠具身决策的视角感知记忆
SafeVantage: Vantage-Aware Memory for Reliable Embodied Decisions
arXiv 2609.36906 · 2026-10-02
