2026-07论文中文摘要
审计大语言模型智能体动作选择中的溯源敏感性
Auditing Provenance Sensitivity in LLM Agent Action Selection
arXiv 2607.20827 · 2026-07-24再探回音壁模式谐振器的非线性与量子光学
Nonlinear and Quantum Optics with Whispering Gallery Resonators Revisited
arXiv 2607.20826 · 2026-07-24模型理论与代数几何中的群块
Group Chunks in Model Theory and Algebraic Geometry
arXiv 2607.20824 · 2026-07-24随机爆发网络的涨落不可能性结果
Fluctuation impossibility results for stochastic burst networks
arXiv 2607.20823 · 2026-07-24重尾预测变量的球余差筛选
Ball-Codifference Screening for Heavy-Tailed Predictors
arXiv 2607.20821 · 2026-07-24基于频谱图的连续记录 IBR 波形中事件的联合检测、定位和分类
Spectrogram-Based Joint Detection, Localization, and Classification of Events in Continuously Recorded IBR Waveforms
arXiv 2607.20817 · 2026-07-24用指南引导的多模态语言模型增强可解释的心脏诊断
Enhancing Explainable Cardiac Diagnosis with Guide-Grounded Multimodal LLMs
arXiv 2607.20814 · 2026-07-24事件镇问题的超饱和构造
Constructions for supersaturation of eventown problems
arXiv 2607.20812 · 2026-07-24神经网络训练可处理性的新复杂性理论前沿
New Complexity-Theoretic Frontiers of Tractability for Neural Network Training
arXiv 2607.20811 · 2026-07-24铁电半导体中用于储层计算的光生电流和短期记忆
Photo-induced currents and short-term memory for reservoir computing in a ferroelectric semiconductor
arXiv 2607.20810 · 2026-07-24三维聚焦能量临界薛定谔方程圆柱阈值动力学的条件散射准则
Conditional scattering criteria for cylindrical threshold dynamics of the three-dimensional focusing energy-critical Schrödinger equation
arXiv 2607.20809 · 2026-07-24关于闭双曲曲面的长度集
On the length sets of closed hyperbolic surfaces
arXiv 2607.20808 · 2026-07-24轻量级大语言模型剖析
Profiling Lightweight Large Language Models
arXiv 2607.20806 · 2026-07-24自正交码的新准则与构造
New Criteria and Constructions for Self-Orthogonal Codes
arXiv 2607.20805 · 2026-07-24有噪声随机电路采样的硬度与复杂度转变
Hardness and Complexity Transition of Noisy Random Circuit Sampling
arXiv 2607.20804 · 2026-07-24Talking Ink：用于有效信道建模和探测器基准测试的基于流的多分子分子通信测试平台
Talking Ink: A Flow-Based Multi-Molecule Molecular Communication Testbed for Effective Channel Modeling and Detector Benchmarking
arXiv 2607.20802 · 2026-07-24精确地进行不精确概率编程：通过分级单子、BDD和半环参数推理的可信集（函数式珍珠）
Imprecise Probabilistic Programming, Precisely: Credal Sets via Graded Monads, BDDs, and Semiring-Parametric Inference (Functional Pearl)
arXiv 2607.20801 · 2026-07-24基于同质性-简约性权衡的外部聚类验证
External Clustering Validation by the Homogeneity-Parsimony Trade-off
arXiv 2607.20799 · 2026-07-24函数演算生成的框架
Frames from Functional Calculus
arXiv 2607.20798 · 2026-07-24一种使用SPHEREx分光光度法和DESI遗产调查成像的恒星-星系分离多模态方法
A Multimodal Approach to Star--Galaxy Separation using SPHEREx Spectrophotometry and DESI Legacy Survey Imaging
arXiv 2607.20797 · 2026-07-24人工智能系统能有创造力吗？来自艺术与工程学的批判性视角
Can an AI System Be Creative? A Critical Perspective from Art and Engineering
arXiv 2607.20796 · 2026-07-24NbN纳米环中的几何超导二极管效应
Geometric Superconducting Diode Effect in an NbN Nanoring
arXiv 2607.20794 · 2026-07-24超轻暗物质重组后磁生成中的数值反作用与有限时间能量转移
Numerical Backreaction and Finite-Time Energy Transfer in Post-Recombination Magnetogenesis from Ultralight Dark Matter
arXiv 2607.20793 · 2026-07-24回忆录：模型在思考时应该写入其内存吗？
Memoir: Should a Model Write to Its Memory While It Thinks?
arXiv 2607.20792 · 2026-07-24虚拟现实中的眼部验证
Ocular Verification for Virtual Reality
arXiv 2607.20790 · 2026-07-243D-GIMP：当3D高斯图像修复与PatchMatch相遇
3D-GIMP: When 3D Gaussian Inpainting Meets PatchMatch
arXiv 2607.20789 · 2026-07-24使用声子插值法对缺陷的电子 - 声子谱函数进行第一性原理计算
First-principles calculation of electron-phonon spectral functions for defects using phonon interpolation
arXiv 2607.20788 · 2026-07-24合成少数群体数据是冗余的或无效的：一种数据依赖的有效性理论和去偏检验
Synthetic minority data is redundant or invalid: a data-dependent validity theory and a de-biased test
arXiv 2607.20787 · 2026-07-24通过发光角度限制实现光伏能量转换的非平衡发射分布
Emergent Non-Equilibrium Emission Profiles via Luminescent Angle Restriction for Photovoltaic Energy Conversion
arXiv 2607.20783 · 2026-07-24一种将多相偏微分方程肿瘤模型耦合的贝叶斯优化框架，以有效设计联合治疗方案
A Bayesian-optimization framework coupling a multiphase PDE tumor model to efficiently design combination therapy schedules
arXiv 2607.20782 · 2026-07-24旗空间、拟阵舒尔代数与斯坦伯格表示
Flag Space, Matroidal Schur Algebras and the Steinberg Representation
arXiv 2607.20779 · 2026-07-24迈向针对大气化学进行微调的人工智能基础模型的机理可解释性
Toward Mechanistic Interpretability of an AI Foundation Model Fine-Tuned for Atmospheric Chemistry
arXiv 2607.20778 · 2026-07-24大复结构边界附近的通量真空
Flux Vacua Near the Boundary of Large Complex Structure
arXiv 2607.20777 · 2026-07-24高阶泊松方程带定量余项的逐点均值公式
Pointwise mean-value formulas with quantitative remainder for higher-order Poisson equations
arXiv 2607.20776 · 2026-07-24Flint：一种语义驱动的数据可视化中间语言
Flint: A Semantics-Driven Data Visualization Intermediate Language
arXiv 2607.20775 · 2026-07-24高能轻同位素实验（HELIX）仪器
The High-Energy Light Isotope eXperiment (HELIX) Instrument
arXiv 2607.20774 · 2026-07-24HARP：人机人工智能研究平台
HARP: The Human--AI Research Platform
arXiv 2607.20773 · 2026-07-24作为光自适应人工虹膜的数字可编程光致变色水凝胶隐形眼镜
Digitally Programmable Photochromic Hydrogel Contact Lenses as Light-Adaptive Artificial Irises
arXiv 2607.20770 · 2026-07-24半摊销参数优化中的内存-计算权衡
Memory-Computation Tradeoffs in Semi Amortized Parametric Optimization
arXiv 2607.20769 · 2026-07-24多样性指标真的在衡量多样性吗？对大语言模型集成中多数投票增益的能力控制审计
Are Diversity Metrics Measuring Diversity? A Capability-Controlled Audit of Majority-Vote Gain in LLM Ensembles
arXiv 2607.20768 · 2026-07-24Rushes：用于多元对齐的人类偏好数据集
Rushes: A Human Preference Dataset for Pluralistic Alignment
arXiv 2607.20767 · 2026-07-24高斯分子形状重叠的重要性抽样估计：精确并集体积和置信度受限的虚拟筛选
Importance-Sampling Estimation of Gaussian Molecular Shape Overlap: Exact Union Volumes and Confidence-Bounded Virtual Screening
arXiv 2607.20766 · 2026-07-24长度为333的勒让德对的乘数障碍
Multiplier obstructions for Legendre pairs of length 333
arXiv 2607.20765 · 2026-07-24ArbiGraph：用于评估上下文管理的任意可扩展可验证任务图
ArbiGraph: Arbitrarily Scalable Verifiable Task Graphs for Evaluating Context Management
arXiv 2607.20764 · 2026-07-24关于一个含雅各布斯塔尔数和斐波那契数的丢番图方程
On a Diophantine Equation with Jacobsthal and Fibonacci Numbers
arXiv 2607.20763 · 2026-07-24量化自动做市商路由中的次优性
Quantifying Sub-Optimality in Routing for Automated Market Makers
arXiv 2607.20762 · 2026-07-24核子的自旋结构：综述
Spin Structure of the Nucleon: Overview
arXiv 2607.20761 · 2026-07-24IssueTrojanBench：针对恶意问题请求对人工智能编码代理进行基准测试
IssueTrojanBench: Benchmarking AI Coding Agents Against Malicious Issue Requests
arXiv 2607.20759 · 2026-07-24通过平衡融合抑制模拟逻辑旋转门中的误差
Suppressing errors in analog logical rotation gates via balanced fusion
arXiv 2607.20756 · 2026-07-24一种用于达西问题与热方程耦合的虚拟单元法
A virtual element method for Darcy's problem coupled with a heat equation
arXiv 2607.20755 · 2026-07-24
