2026-09论文中文摘要
非线性广义最小残差法（NGMRES）在压缩和非压缩迭代中的收敛性分析及加速证明
NGMRES convergence analysis and proof of acceleration for contractive and noncontractive iterations
arXiv 2607.14268 · 2026-09-07保护能力幻觉：当大语言模型声称具备不存在的能力时
Protective Capacity Hallucination: When Large Language Models Claim Nonexistent Capabilities
arXiv 2607.13596 · 2026-09-07使用对称注意力的水反射检测
Water Reflection Detection Using Symmetric Attention
arXiv 2607.10749 · 2026-09-07海上贸易中断带来的宏观经济风险
Macroeconomic Risks from Maritime Trade Disruptions
arXiv 2607.09951 · 2026-09-07EvoCUA-1.5：用于多轮计算机使用智能体的在线强化学习
EvoCUA-1.5: Online Reinforcement Learning for Multi-turn Computer-Use Agents
arXiv 2607.09773 · 2026-09-07非饱和土壤水的统计物理学：动力学理论与非对易孔隙水动力学
The Statistical physics of unsaturated soil water: kinetic theory and non commutative pore water dynamics
arXiv 2607.09416 · 2026-09-07Chevalley - Herbrand公式与p - 主虚二次域的Z_p - 扩张
Chevalley-Herbrand formulas and Zp -extensions of a p-principal imaginary quadratic field
arXiv 2607.09258 · 2026-09-07使用Scotty合成诊断通过垂直位移和极向角扫描进行湍流谱的DBS测量比较
Comparison of DBS measurements of turbulence spectra by vertical-displacement and poloidal-angle scans using the Scotty synthetic diagnostic
arXiv 2607.09219 · 2026-09-07从推理中估计不确定性：对大语言模型中多语言和跨语言MCQA性能的大规模研究
Estimating Uncertainty from Reasoning: A Large-Scale Study of Multi- and Crosslingual MCQA Performance in LLMs
arXiv 2607.06327 · 2026-09-07截断移位杨代数的范畴$\mathcal{O}$与双无限Bott - Samelson簇
Category $\mathcal{O}$ for truncated shifted Yangians and the bi-infinite Bott-Samelson variety
arXiv 2607.04480 · 2026-09-07连续酉值映射的量子柯尔莫哥洛夫 - 阿诺德表示定理
Quantum Kolmogorov--Arnold representation theorem for continuous unitary-valued maps
arXiv 2607.03187 · 2026-09-07基于静态原子总线的容错量子计算
Fault-tolerant quantum computation with static atomic buses
arXiv 2607.02804 · 2026-09-07来自Peccei-Quinn对称性的单圈质子衰变
One-loop proton decay from Peccei-Quinn symmetry
arXiv 2607.02026 · 2026-09-07利用FAST对3I/ATLAS进行周期性射电技术特征搜索
Periodic Radio Technosignature Search toward 3I/ATLAS with FAST
arXiv 2607.01666 · 2026-09-07有限Sidon集的向量值平滑
An Improved Upper Bound for Finite Sidon Sets via Vector-Valued Smoothing
arXiv 2607.01169 · 2026-09-07迈向鲁棒的驾驶感知：用于自监督单目深度估计的灵活尺度驱动系列
Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation
arXiv 2607.00736 · 2026-09-07使用输运模型在中心快度区识别强子的偶极流
Dipolar flow of identified hadrons at mid-rapidity using transport models
arXiv 2607.00703 · 2026-09-07基于印刷字体的单目距离估计用于高级驾驶辅助系统
Calibrated Multichannel Monocular Ranging From Standardized License Plates With Metrology-Exact Validation
arXiv 2607.00319 · 2026-09-07代理间金融：面向自主AI代理的区块链支付与信任基础设施
Agent-to-Agent Finance: Blockchain Payments and Trust Infrastructure for Autonomous AI Agents
arXiv 2607.00245 · 2026-09-07X射线双星系统高频准周期振荡与S2星轨道对量子Oppenheimer--Snyder黑洞的联合约束
Joint Constraints on Quantum Oppenheimer--Snyder Black Holes from High-Frequency Quasi-Periodic Oscillations in X-ray Binary Systems and the S2 Star Orbit
arXiv 2609.16018 · 2026-09-05时间自蒸馏：无监督学习视频中的视觉状态跟踪
Temporal Self-Distillation: Learning Visual State Tracking in Videos Without Supervision
arXiv 2609.04203 · 2026-09-04TokenMatch：基于曲率引导分词的3D网格对应Transformer
TokenMatch: 3D Mesh Correspondence Transformer with Curvature-Guided Tokenisation
arXiv 2609.04202 · 2026-09-04Scal3R：学习高效的多相对位姿查询以实现可扩展的在线三维重建
Scal3R: Learning Efficient Multi-Relative Pose Query for Scalable Online 3D Reconstruction
arXiv 2609.04201 · 2026-09-04Principia：面向视频模型的关系物理测试基准
Principia: Relational Physics Tests for Video Models
arXiv 2609.04200 · 2026-09-04通过训练进行编译：将自然语言规范转换为局部神经函数
Compile by Training: Turning Natural-Language Specifications into Local Neural Functions
arXiv 2609.04199 · 2026-09-04清洁工程，不稳定测量：共享端点上黑盒大语言模型评判器的预注册可靠性失效
Clean Engineering, Unstable Measurement: A Preregistered Reliability Failure of Black-Box LLM Observers on Shared Endpoints
arXiv 2609.04198 · 2026-09-04ESPO：基于诊断、多样化与稳定化的错误结构化提示优化
ESPO: Error-Structured Prompt Optimization via Diagnose, Diversify, and Stabilize
arXiv 2609.04197 · 2026-09-04Puffin-World：基于原生3D世界状态扩展统一多模态模型
Puffin-World: Scaling a Unified Multimodal Model with Native 3D World States
arXiv 2609.04196 · 2026-09-04可解释性≠可理解性：对比思维链推理中的主观判断重要性与实际重要性
Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning
arXiv 2609.04194 · 2026-09-04GIFT：面向机器人操作的基于动作导向结构监督的引导式中间特征训练
GIFT: Guided Intermediate Feature Training via Action-Oriented Structural Supervision for Robotic Manipulation
arXiv 2609.04193 · 2026-09-04欧几里得（Euclid）筹备：Λ冷暗物质（ΛCDM）与非标准宇宙学中暗物质晕轮廓的形状
Euclid preparation. The shape of halo profiles in $Λ$CDM and non-standard cosmologies
arXiv 2609.04192 · 2026-09-04虫洞诱导引力轴子质量隙的婴儿宇宙态依赖
Baby-universe state dependence of the wormhole-induced gravi-axion mass gap
arXiv 2609.04191 · 2026-09-04一个编辑器，多种编辑：用于多样化视频编辑的统一无训练框架
One Editor, Many Edits: A Unified Training-Free Framework for Diverse Video Editing
arXiv 2609.04190 · 2026-09-04任意子超导体的波函数
Wavefunctions for Anyon Superconductors
arXiv 2609.04187 · 2026-09-04LZ 2026 excess 的大气中微子上散射解释
Atmospheric neutrino up-scattering explanation of LZ 2026 excess
arXiv 2609.04185 · 2026-09-04以小模型成本实现前沿质量的声明式UI生成
Toward Frontier-Quality Declarative UI Generation at Small-Model Cost
arXiv 2609.04184 · 2026-09-04先看后合成：面向弱监督密集视频字幕的VLM引导式过渡事件发现
Seeing Before Synthesizing: VLM-Guided Transition Event Discovery for Weakly-Supervised Dense Video Captioning
arXiv 2609.04183 · 2026-09-04量子通信和贝尔非定域性需要无限的经典通信才能模拟
Quantum communication and Bell nonlocality require infinite classical communication to simulate
arXiv 2609.04182 · 2026-09-04来自LUX-ZEPLIN高能事件的季节暗物质
Seasonal dark matter from the LUX-ZEPLIN high-energy event
arXiv 2609.04181 · 2026-09-04预训练期间的知识获取？辅助视角让大语言模型学习得更好
Knowledge Acquisition During Pre-training? Large Language Models Learn Better With Auxiliary Views
arXiv 2609.04180 · 2026-09-04晶体学Coxeter群的标准表示
Standard Representations of Crystallographic Coxeter Groups
arXiv 2609.04179 · 2026-09-04一种计算可行的因果概率解释框架
A Computationally Feasible Framework for Causal Probabilistic Explanation
arXiv 2609.04177 · 2026-09-04卡塔兰常数是无理数
Catalan's constant is irrational
arXiv 2609.04176 · 2026-09-04利用高能边带对抗LZ实验事件的希格斯微子解释
Confronting the Higgsino Interpretation of the LZ Event with the High-Energy Sideband
arXiv 2609.04175 · 2026-09-04基于3D基础模型场景表示的零样本新视角深度合成
Zero-Shot Novel Depth Synthesis Using 3D Foundation Models Scene Representations
arXiv 2609.04174 · 2026-09-04大型语言模型的在线策略蒸馏再思考 II：一个训练示例
Rethinking On-Policy Distillation of Large Language Models II: One Training Example
arXiv 2609.04172 · 2026-09-04氮化硅微谐振器中的双光子产生
Twin-photon generation in a silicon nitride microresonator
arXiv 2609.04171 · 2026-09-04自主研究群体中涌现的作弊与举报行为案例研究
A Case Study on Emergent Cheating and Whistleblowing in Autonomous Research Swarms
arXiv 2609.04170 · 2026-09-04Para-Pipe：在片上系统（SoC）上利用机器学习计算图的分层算子并行性
Para-Pipe: Exploiting Hierarchical Operator Parallelism of ML Computational Graphs on SoCs
arXiv 2609.04168 · 2026-09-04SWE-Gate：软件工程智能体仅通过功能测试是不够的
SWE-Gate: Passing Functional Tests Is Not Enough for Software Engineering Agents
arXiv 2609.04167 · 2026-09-04
