2026-08论文中文摘要
BBOWP-Bench：评估大型语言模型在黑盒优化文字问题上的表现
BBOWP-Bench: Evaluating LLMs on Black-Box Optimization Word Problems
arXiv 2608.02612 · 2026-08-05KernelBrain：面向智能体的GPU内核优化的粗到细、预算感知搜索
KernelBrain: Coarse-to-Fine, Budget-Aware Search for Agentic GPU Kernel Optimization
arXiv 2608.02611 · 2026-08-05Glob生态系统的实证分析
An Empirical Analysis of the Glob Ecosystem
arXiv 2608.02610 · 2026-08-05TabletCraft：通过双向阿卡德语神经机器翻译（NMT）与楔形文字渲染弥合4000年的文化鸿沟
TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering
arXiv 2608.02609 · 2026-08-05将智能与推理分离：边缘原生AI计算的标准
Separating Intelligence from Inference: A Standard for Edge-Native AI Computing
arXiv 2608.02608 · 2026-08-05从业务流程模型中提取ODRL策略：一种用于“通过提取实现合规性”的图遍历方法
Extracting ODRL Policies from Business Process Models: A Graph Traversal Approach to Compliance-by-Extraction
arXiv 2608.02607 · 2026-08-05多对对跖点迫使产生多对邻近点
Many Antipodal Pairs Force Many Neighboring Pairs
arXiv 2608.02605 · 2026-08-05ISEE：数据库字段的交互式语义丰富
ISEE: Interactive Semantic Enrichment for Database Fields
arXiv 2608.02604 · 2026-08-05ACEM：面向智能体软件工程的成本估算模型
ACEM: A Cost Estimation Model for Agentic Software Engineering
arXiv 2608.02582 · 2026-08-05用于单阶段哈达玛量化的两两独立抖动
Pairwise-Independent Dithering for Single-Stage Hadamard Quantization
arXiv 2608.02564 · 2026-08-05通过中点海森矩阵以 $2^{0.6039n}$ 的时间求解最短向量问题
Solving the Shortest Vector Problem in $2^{0.6039n}$ Time via Mid-point Hessian
arXiv 2608.02478 · 2026-08-05InfiniSplat：用于大基线单目视图合成的隐式高斯解码
InfiniSplat: Implicit Gaussian Decoding for Large-Baseline Monocular View Synthesis
arXiv 2608.02437 · 2026-08-05直接从第二绝热不变量优化仿星器全向性
Direct Optimization of Stellarator Omnigenity from the Second Adiabatic Invariant
arXiv 2608.02418 · 2026-08-05SkillTrace：遍历查询-技能图以构建可组合的大语言模型智能体
SkillTrace: Traversing a Query-Skill Graph for Composable LLM Agents
arXiv 2608.02356 · 2026-08-05ChainVLA：通过统一执行状态串联视觉-语言-动作查询以实现长 horizon 操纵
ChainVLA: Chaining Vision-Language-Action Queries through a Unified Execution State for Long-Horizon Manipulation
arXiv 2608.02326 · 2026-08-05用于提升PRIMA FIR-ESS千像素阵列可用像素良率的空间映射与电容修整技术进展
Spatial Mapping and Capacitor Trimming Developments to Improve Usable Pixel Yield in PRIMA FIRESS Kilo-Pixel Arrays
arXiv 2608.02273 · 2026-08-05用于自发裂变反应的带投影的精确含时生成坐标法
An exact time-dependent generator coordinate method with projection for spontaneous fission reactions
arXiv 2608.02220 · 2026-08-05利用具备甚长基线干涉测量（VLBI）探测到反向喷流的活动星系核（AGN）的变异性和自行测量宇宙学距离的改进方法
A refined method for measuring cosmological distances using variability and proper motions in AGN with VLBI-detected counter-jets
arXiv 2608.02202 · 2026-08-05解释延迟选择量子擦除器的标准量子力学处理方法
A Standard Quantum Mechanical Treatment to Rationalize the Delayed-Choice Quantum Eraser
arXiv 2608.02185 · 2026-08-05赋权有向图的同调移位理想
Homological shift ideals of weighted oriented graphs
arXiv 2608.02170 · 2026-08-05用于三重周期极小曲面约束生成的傅里叶潜扩散模型
Fourier-Latent Diffusion for Constrained Generation of Triply Periodic Minimal Surfaces
arXiv 2608.02151 · 2026-08-05基于渐进式经验演化的自我改进大语言模型
Self-Improving Large Language Models via Progressive Experience Evolution
arXiv 2608.02139 · 2026-08-05吸积诱导自旋加速：对吸积毫秒X射线脉冲星（AMXPs）质量约束的启示
Accretion-induced spin-up: Implications for mass constraints of AMXPs
arXiv 2608.02061 · 2026-08-05RSVideo：你的视觉-语言模型准备好应对遥感视频了吗？
RSVideo: Are Your Vision-Language Models Ready for Remote Sensing Videos?
arXiv 2608.02039 · 2026-08-05SwanTale：面向指令式与零样本任务的统一多说话人语音与音频生成模型
SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot Tasks
arXiv 2608.02023 · 2026-08-05在推理失效之前：智能体检索增强生成（RAG）中的证据前程序失效
Before Reasoning Can Fail: Pre-Evidence Procedural Failures in Agentic RAG
arXiv 2608.02011 · 2026-08-05P-MUSE：用于统一器乐合成与编辑的提示-MIDI可选模型
P-MUSE: Prompt-MIDI-Optional Model for Unified Instrumental Music Synthesis and Editing
arXiv 2608.01920 · 2026-08-05CRISP：用于训练高效深度搜索智能体的关键步骤感知
CRISP: Critical Step Perception for Training Efficient Deep Search Agents
arXiv 2608.01867 · 2026-08-05面向自回归超分辨率的可信粗尺度细节延续
Detail Continuation over a Trustworthy Coarse Scale for Autoregressive Super-Resolution
arXiv 2608.01823 · 2026-08-05LEAP：面向GPU内核生成的代码强化学习的自适应剪枝轻量环境反馈框架
LEAP: Lean Environment-Feedback via Adaptive Pruning for Code RL in GPU Kernel Generation
arXiv 2608.01804 · 2026-08-05梯度依赖半线性方程混合边界连接处的对数共振
Logarithmic Resonance at Mixed Boundary Junctions in Gradient-Dependent Semilinear Equationsv
arXiv 2608.01790 · 2026-08-05用于自动驾驶视觉语言模型可验证推理的未来轨迹延迟暴露
Deferred Exposure of Future Trajectories for Verifiable Reasoning in Autonomous Driving VLMs
arXiv 2608.01755 · 2026-08-05IDraw：基于数字绘画图像的艺术家身份验证
IDraw: Artist Verification from Digital Drawing Images
arXiv 2608.01737 · 2026-08-05来自洛伦兹环面的德西特全息图
de Sitter holography from a Lorentzian torus
arXiv 2608.01729 · 2026-08-05Katz-Long-Moody构造导出的Hecke代数表示
Hecke algebra representations from the Katz-Long-Moody construction
arXiv 2608.01727 · 2026-08-05当记忆成为权威：在记忆整合边界处对权威崩溃进行基准测试
When Memory Becomes Authority: Benchmarking Authority Collapse at the Memory Consolidation Boundary
arXiv 2608.01679 · 2026-08-05CellPrism：用于药物发现中探索AI驱动虚拟细胞的可视化分析系统
CellPrism: A Visual Analytics System for Exploring AI-Driven Virtual Cells in Drug Discovery
arXiv 2608.01669 · 2026-08-05LongCat 稀疏注意力：通过流感知分层跨层索引驯服闪电（Lightning）
LongCat Sparse Attention: Taming the Lightning via Streaming-aware Hierarchical Cross-Layer Indexing
arXiv 2608.01662 · 2026-08-05协变量偏移下岭回归的池化与集成方法对比
Pooling Versus Ensembling for Ridge Regression Under Covariate Shift
arXiv 2608.01641 · 2026-08-05超越形态的运动：从抽象运动表示引导跨类别运动迁移
Motion Beyond Morphology: Bootstrapping Cross-Category Motion Transfer from Abstract Motion Representations
arXiv 2608.01628 · 2026-08-05博尔扎曲面与克莱因四次曲线的谱行列式
Spectral determinants of the Bolza surface and the Klein quartic
arXiv 2608.01611 · 2026-08-05带有平均收益保证的交替时间时序逻辑
Alternating-Time Temporal Logic with Mean-Payoff Guarantees
arXiv 2608.01499 · 2026-08-05随机几何图中团渗流的格里菲斯相
Griffiths phase in clique percolation in random geometric graphs
arXiv 2608.01487 · 2026-08-05推理分歧之处：本地化多智能体辩论
Where Reasoning Diverges: Localized Multi-Agent Debate for Multi-Hop Question Answering
arXiv 2608.01463 · 2026-08-05后AGI经济学的一种新价值理论
A New Theory of Value for Post-AGI Economics
arXiv 2608.01432 · 2026-08-05以正确方式提问：用于复杂任务解决中提示词生成的多智能体对话系统
Asking Questions the Right Way: A Multi-Agent Conversational System for Prompt Formulation in Complex Task Resolution
arXiv 2608.01366 · 2026-08-05ACE-GraphRAG：面向分层GraphRAG的智能体上下文工程
ACE-GraphRAG: Agentic Context Engineering for Hierarchical GraphRAG
arXiv 2608.01269 · 2026-08-05重新研究针对多面体的分解不变条件梯度方法
Revisiting Decomposition-Invariant Conditional Gradient Methods for Polytopes
arXiv 2608.01243 · 2026-08-05智能膜：基于人工神经网络的片上高内涵原位监测屏障
Smart membrane: high content in situ monitoring barrier on chip with artificial neural network
arXiv 2608.01239 · 2026-08-05从森林到未来首都：利用PlanetScope影像追踪2021至2026年努桑塔拉首都（IKN）的土地覆盖变化
From Forest to Future Capital: Tracking Land Cover Change in Ibu Kota Nusantara (IKN) from 2021 to 2026 with PlanetScope Imagery
arXiv 2608.01230 · 2026-08-05
