2026-08论文中文摘要
MOON：面向多任务学习的多目标正交归一化更新方法
MOON: Multi-Objective OrthoNormalized Updates for Multitask Learning
arXiv 2608.11749 · 2026-08-13让每一步都有价值：面向成像逆问题的时空信息分配
Making Every Step Count: Spatio-Temporal Information Allocation for Imaging Inverse Problems
arXiv 2608.11747 · 2026-08-13面向分布外泛化的、受Epiplexity引导的数据选择与生成
Epiplexity Guided Data Selection and Generation for Out-of-Distribution Generalization
arXiv 2608.11746 · 2026-08-13利用大语言模型（LLM）实现Python库的文档引导模糊测试
Harnessing LLMs for Document-Guided Fuzzing of Python Libraries
arXiv 2608.11744 · 2026-08-13具有小表示界的Artin代数
Artin algebras of small representation bound
arXiv 2608.11743 · 2026-08-13JieZi：用于古文字训诂的大规模专家审核数据集与基准
JieZi: A Large-Scale Expert-Audited Dataset and Benchmark for Ancient Chinese Character Exegesis
arXiv 2608.11741 · 2026-08-13G0.5：用于机器人推理与动作的单自回归流
G0.5: One Autoregressive Stream for Robot Reasoning and Action
arXiv 2608.11739 · 2026-08-13推进基于多模态大语言模型（MLLM）的无人机（UAV）图像理解与推理：基准测试及无训练多智能体系统
Advancing MLLM-based UAV Image Understanding and Reasoning: A Benchmark and a Training-Free Multi-Agent System
arXiv 2608.11738 · 2026-08-13Phoenix TTS：基于流匹配驱动的语音分词实现高保真合成与语音转换
Phoenix TTS: High-Fidelity Synthesis and Voice Conversion via Flow-Matching-Driven Speech Tokenization
arXiv 2608.11737 · 2026-08-13欧氏拉姆齐集的单点扩张
One-point extensions of Euclidean Ramsey sets
arXiv 2608.11736 · 2026-08-13定位并控制大语言模型中的隐式个性化
Locating and Controlling Implicit Personalization in Large Language Models
arXiv 2608.11735 · 2026-08-13平均击中时间与递推结构 II：循环图幂与正则图的笛卡尔积
Average hitting times and recurrence structures II: Cartesian products of powers of cycles and regular graphs
arXiv 2608.11734 · 2026-08-13基于塌陷生成的文本到图像扩散模型指纹识别
Fingerprinting Text-to-Image Diffusion Models via Collapsed Generation
arXiv 2608.11732 · 2026-08-13ContactIPM：一种利用结构的接触隐式轨迹优化内点求解器
ContactIPM: A Structure-Exploiting Interior-Point Solver for Contact-Implicit Trajectory Optimization
arXiv 2608.11731 · 2026-08-13针对后过滤访问控制的明文恢复
Plaintext Recovery Against Post-Filtering Access Control
arXiv 2608.11730 · 2026-08-13星际冰在碳质尘埃上的分子特异性扩散与脱附
Molecule-specific diffusion and desorption of interstellar ices on carbonaceous dust
arXiv 2608.11729 · 2026-08-13位形空间中有理多边形本征函数的量子混合
Quantum mixing for eigenfunctions of rational polygons in configuration space
arXiv 2608.11728 · 2026-08-13Harness-IF：评估编码智能体在不同指令层面的指令遵循能力
Harness-IF: Evaluating Instruction Following Across Instruction Surfaces in Coding Agents
arXiv 2608.11727 · 2026-08-13紧李群中的Kakeya集与维数压缩
Kakeya sets and dimension compression in compact Lie groups
arXiv 2608.11726 · 2026-08-13深度学习加速的高k HfO2电介质掺杂剂选择：Y、Si和Al的无序解析研究
Deep-Learning-Accelerated Dopant Selection for High-k HfO2 Dielectrics: A Disorder-Resolved Study of Y, Si and Al
arXiv 2608.11725 · 2026-08-13基于贝叶斯更新的概率分布比例类比
Proportional Analogies on Probability Distributions via Bayesian Updating
arXiv 2608.11724 · 2026-08-13每个三次图的2-细分都是反魔术图
Every 2-Subdivision of a Cubic Graph Is Antimagic
arXiv 2608.11723 · 2026-08-13具阻尼振荡的半线性波动方程的临界指数
The critical exponent for semilinear wave equations with damped oscillations
arXiv 2608.11722 · 2026-08-13分布式量子算法无法以概率1对环进行着色
Distributed Quantum Algorithms Cannot Color Cycles with Probability 1
arXiv 2608.11720 · 2026-08-13不要让CNOT门压倒解码器：为快速容错量子计算（FTQC）调度横向门
Do Not Let CNOTs Overwhelm the Decoder: Scheduling Transversal Gates for Fast FTQC
arXiv 2608.11719 · 2026-08-13硅光子芯片间的实验量子远程克隆
Experimental quantum telecloning across silicon photonic chips
arXiv 2608.11718 · 2026-08-13热化学非平衡效应对湍流边界层的影响
Thermochemical non-equilibrium effects on turbulent boundary layers
arXiv 2608.11717 · 2026-08-13思维链（Chain-of-Thought, CoT）展现了通往树的路径：实现分支复杂性
Chain-of-Thought Shows the Path to a Tree: Realizing Branching Complexity
arXiv 2608.11716 · 2026-08-13当API“说错”语言：重新审视多语言工具使用的后训练
When the API Speaks the Wrong Language: Revisiting Post-Training for Multilingual Tool Use
arXiv 2608.11715 · 2026-08-13量子比特系统中由量子通道诱导的Uhlmann相位几何
Quantum Channel-Induced Geometry of the Uhlmann Phase in Qubit Systems
arXiv 2608.11714 · 2026-08-13基于学习到的变量交互的高维多目标贝叶斯优化
High-dimensional Multi-objective Bayesian Optimization with Learned Variable Interactions
arXiv 2608.11713 · 2026-08-13如何击败先来先服务（FCFS）
How to Beat FCFS
arXiv 2608.11710 · 2026-08-13面向元宇宙入门环境的基于浏览器的手势驱动虚拟化身交互框架
A Browser-Based Gesture-Driven Avatar Interaction Framework for Metaverse Onboarding Environments
arXiv 2608.11708 · 2026-08-13(Cu,C)-1234中的厚度驱动超导体-绝缘体转变及(Cu,C)-1234/YBCO异质结构中邻近诱导的超导性恢复
Thickness-Driven Superconductor-Insulator Transition in (Cu,C)-1234 and Proximity-Induced Superconductivity Recovery in (Cu,C)-1234/YBCO Heterostructure
arXiv 2608.11707 · 2026-08-13通过径向 boost 形成陷获面
Trapped surface formation via radial boosts
arXiv 2608.11706 · 2026-08-13提升大语言模型的客观性：通过特质不变安全微调稳定LLM在不同特质下的安全行为
Making Your LLMs More Objective: Stabilizing LLM Safety Behavior Across Traits with Trait-Invariant Safety Tuning
arXiv 2608.11705 · 2026-08-13在壳振幅与黑洞微扰：精确的雷斯纳-诺德斯特龙（Reissner-Nordström, RN）混合
On-Shell Amplitudes and Black-Hole Perturbations: Exact Reissner-Nordström Mixing
arXiv 2608.11703 · 2026-08-13相位协变操作的桌面可逆性
Tabletop reversibility of phase-covariant operations
arXiv 2608.11702 · 2026-08-13Consolidator：学习跨上下文边界的持久路由记忆
Consolidator: Learning Persistent Routed Memory Across Context Boundaries
arXiv 2608.11701 · 2026-08-13非厄米准晶中的对数关联势场与局域化
Logarithmically Correlated Landscapes and Localization in Non-Hermitian Quasicrystals
arXiv 2608.11700 · 2026-08-13STAR：面向通用3D场景理解的空间拓扑感知路由框架
STAR: A Spatial-Topology Aware Routing Framework for Generalizable 3D Scene Understanding
arXiv 2608.11699 · 2026-08-13商业养殖环境下猪只点云的边界增强分割
Boundary-Enhanced Segmentation of Pig Point Clouds in Commercial Housing Environments
arXiv 2608.11697 · 2026-08-13经改变的宇宙有机物作为金星的紫外线吸收剂
Altered Cosmic Organics as Venus' Ultraviolet Absorbers
arXiv 2608.11696 · 2026-08-13d≥2 下的广义随机曲率流，以及带非线性扩散的随机 Allen-Cahn 方程的尖锐界面极限
Generalised stochastic curvature flow in $d \geq 2$, and sharp interface limit for the stochastic Allen-Cahn equation with nonlinear diffusion
arXiv 2608.11695 · 2026-08-13措辞效应：量化大语言模型基准测试性能中的双向漂移
The Wording Effect: Quantifying Two-Way Drift in LLM Benchmark Performance
arXiv 2608.11694 · 2026-08-13HUGIN：增强自主物流分拣的视觉-语言规划能力
HUGIN: Enhancing Vision-Language Planning for Autonomous Logistics Sorting
arXiv 2608.11692 · 2026-08-13LEMUR：基于视觉锚定推理重定向的潜在熵感知多模态遗忘
LEMUR: Latent Entropy-aware Multimodal Unlearning via Visual-anchored Reasoning Redirection
arXiv 2608.11691 · 2026-08-13漂移与依赖：基于重放的持续学习的分层信息论边界
Drift and Dependence: Layer-wise Information-Theoretic Bounds for Replay-Based Continual Learning
arXiv 2608.11690 · 2026-08-13用于非光滑凸优化问题的Wolfe型谱共轭梯度方法
A Wolfe-Type Spectral Conjugate Gradient Method for Nonsmooth Convex Optimization Problems
arXiv 2608.11689 · 2026-08-13APEX：面向内存高效型边缘MoE推理的自适应专家预取技术
APEX: Adaptive Expert Prefetching for Memory-Efficient Edge MoE Inference
arXiv 2608.11688 · 2026-08-13
