2026-08论文中文摘要
从SQL到知识图谱：一种基于大语言模型的多智能体方法及数据模式改进
From SQL to Knowledge Graphs: An LLM-Driven Multi-Agent Approach with Data Schema Improvement
arXiv 2608.26117 · 2026-08-28期权隐含信号与崩盘风险：来自美国股票期权的可预测性及机器学习证据
Option-Implied Signals and Crash Risk: Predictability and Machine-Learning Evidence from U.S. Equity Options
arXiv 2608.26115 · 2026-08-28CIFQA：一种用于金融查询问答的确定性工具基多智能体大语言模型框架
CIFQA: A Deterministic Tool-Grounded Multi-Agent LLM Framework for Financial Query Answering
arXiv 2608.26114 · 2026-08-28PICasso：一种用于硅光子器件自主优化的AI辅助设计框架
PICasso: An AI-Enabled Design Framework for Autonomous Optimization of Silicon Photonic Devices
arXiv 2608.26113 · 2026-08-28用于电池 prognostics 与健康管理的大模型：综述与未来路线图
Large Models for Battery Prognostics and Health Management: A Review and Future Roadmap
arXiv 2608.26111 · 2026-08-28探索大语言模型（LLM）在高性能计算（HPC）编程中的作用：一项综述
Exploring the Role of LLMs in HPC Programming: A Survey
arXiv 2608.26110 · 2026-08-28独立大型语言模型（LLM）与预定义智能体管道用于解释ICU死亡率预测：基于eICU演示数据集的可行性研究
Standalone LLM and a Pre-specified Agentic Pipeline for Explaining ICU Mortality Predictions: a Feasibility Study on the eICU Demo Dataset
arXiv 2608.26109 · 2026-08-28用于安全加固的智能体AI containment架构
Agentic AI Containment Architecture for Security Hardening
arXiv 2608.26108 · 2026-08-28EduRiskX：用于早期学业风险预测的神经符号框架，结合F-Logic推理
EduRiskX: A Neuro-Symbolic Framework with F-Logic Reasoning for Early Academic Risk Prediction
arXiv 2608.26107 · 2026-08-28用于当日股票方向预测的统计金融基准：来自SPY的滚动向前验证证据
A Statistical-Finance Benchmark for Same-Day Directional Stock Prediction: Walk-Forward Evidence from SPY
arXiv 2608.26106 · 2026-08-28菱面石墨烯中的多分量磁畴壁
Multicomponent Magnetic Domain Walls in Rhombohedral Graphene
arXiv 2608.26104 · 2026-08-28Zero-WAM：基于人类视频的上下文世界-动作建模用于开放式任务泛化
Zero-WAM: In-Context World-Action Modeling from Human Videos for Open-Ended Task Generalization
arXiv 2608.26103 · 2026-08-28粘手性输运：贝里曲率对流体动力学涡旋的手性选择
Viscochiral Transport: Chiral Selection of Hydrodynamic Vortices by Berry Curvature
arXiv 2608.26102 · 2026-08-28用于大核卷积神经网络中移动高效逐点卷积的组共享低秩近似
Group-Shared Low-Rank Approximation for Mobile-Efficient Pointwise Convolutions in Large-Kernel CNNs
arXiv 2608.26069 · 2026-08-28$\u2113_p$子空间逼近的近最优强核心集
Nearly Optimal Strong Coresets for $\ell_p$ Subspace Approximation
arXiv 2608.26047 · 2026-08-28VISA：用于多模态指令遵循的智能体式自进化数据合成
VISA: Agentic Self-Evolving Data Synthesis for Multimodal Instruction Following
arXiv 2608.26013 · 2026-08-28DESCENT：面向机场地面运动预测的有向边场景编码
DESCENT: Directed Edge Scene Encoding for Airport Surface Movement Prediction
arXiv 2608.26002 · 2026-08-28揭示无训练大语言模型文本检测中的频谱机制
Unveiling Spectral Mechanisms in Training-Free LLM Text Detection
arXiv 2608.25944 · 2026-08-28双环面上定量刚性的有界平均运动之外的例子
Examples beyond Bounded Mean Motion for Quantitative Rigidity on the Two-Torus
arXiv 2608.25906 · 2026-08-28晚学习，早引导：用于公平人脸生成的时间步解耦语义引导
Learning Late, Guiding Early: Timestep-Decoupled Semantic Guidance for Fair Face Generation
arXiv 2608.25862 · 2026-08-28缺项斜积的正则性、定量偏差与非刚性
Regularity, quantitative deviation, and non-rigidity of a lacunary skew product
arXiv 2608.25821 · 2026-08-28一类随机微分方程的显式保域数值格式
Explicit domain preserving numerical schemes for a class of stochastic differential equations
arXiv 2608.25685 · 2026-08-28用QLoRA学习新事实：获取-保留前沿
Learning New Facts with QLoRA: An Acquisition-Retention Frontier
arXiv 2608.25677 · 2026-08-28采样token反向KL策略上蒸馏的token级分析
A Token-Level Analysis of Sampled-Token Reverse-KL On-Policy Distillation
arXiv 2608.25643 · 2026-08-28偶数维中从欧氏空间到球面的正定性传递的精确支撑反例
Exact-Support Counterexamples to Euclidean-to-Spherical Transfer of Positive Definiteness in Even Dimensions
arXiv 2608.25639 · 2026-08-28NGC 4438 活动星系核驱动气泡中的能量分配：从核区气泡到星系尺度外流
Energy Partition in AGN-driven Bubbles of NGC 4438: From Nuclear Bubbles to a Galaxy-scale Outflow
arXiv 2608.25552 · 2026-08-28ClueWeaver：面向紧凑型大语言模型的、针对文学长篇叙事的奖励引导双智能体证据推理框架
ClueWeaver: Reward-Guided Dual-Agent Evidence Reasoning for Compact LLMs on Literary Long Narratives
arXiv 2608.25531 · 2026-08-28TailorCoPilot：实现带版本控制状态跟踪的智能体式制版
TailorCoPilot: Enabling Agentic Pattern Making with Version-Controlled State Tracking
arXiv 2608.25462 · 2026-08-28核能级密度对r-过程稀土峰核合成的影响
Impact of Nuclear Level Density on $r$-Process Rare-Earth Peak Nucleosynthesis
arXiv 2608.25440 · 2026-08-28通过唤醒无线电重新思考无电池感知通信
Rethinking Battery-free Sensing Communication via Wake-up Radios
arXiv 2608.25292 · 2026-08-28改进的低开销通信高效字符串协调与编辑距离算法
Improved Low-Overhead Communication-Efficient String Reconciliation and Edit Distance
arXiv 2608.25179 · 2026-08-28细观力学统计模型将纤维网络中的诱导向列序与力学响应关联起来
Micromechanical statistical model links induced nematic order to mechanical response in fiber networks
arXiv 2608.25125 · 2026-08-28DataKernelBench：大语言模型能否在GPU上优化数据库查询？
DataKernelBench: Can LLMs Optimize Database Queries on GPUs?
arXiv 2608.25061 · 2026-08-28基础模型的无监督后训练：一项综述
Unsupervised Post-Training of Foundation Models: A Survey
arXiv 2608.24982 · 2026-08-28可检索但不可靠：检索增强生成中的攻击与防御研究综述
Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation
arXiv 2608.24977 · 2026-08-28反恐精英2（CS2）中用于账号一致性的同玩家验证
Account Consistency from Gameplay Traces: Same-Player Verification in Counter-Strike 2
arXiv 2608.24893 · 2026-08-28四变量五次型与七次型的Fröberg猜想
Fröberg's Conjecture for Quintics and Septics in Four Variables
arXiv 2608.24797 · 2026-08-28具有长程跃迁的两腿Creutz梯子模型中的动态量子相变
Dynamic quantum phase transitions in the two-leg Creutz ladder with long-range hopping
arXiv 2608.24514 · 2026-08-28反向传播飞秒脉冲的时空同步
Spatio-Temporal Synchronization of Counter-Propagating Femtosecond Pulses
arXiv 2608.24387 · 2026-08-28从状态到动作：用于可靠多轮工具使用的OODA-Tool
From State to Action: OODA-Tool for Reliable Multi-Turn Tool Use
arXiv 2608.24368 · 2026-08-28保留还是丢弃？用于紧凑视频表示的自适应分词器
Keep-or-Drop? Adaptive Tokenizer for Compact Video Representation
arXiv 2608.24293 · 2026-08-28典型吸积盘冕源4U 1822-37的首次X射线偏振测量
The First X-Ray Polarimetry of the Typical Accretion Disk Corona Source 4U 1822-37
arXiv 2608.24292 · 2026-08-28条件正定核与通用克里金法的样本压缩
Samplet compression for conditionally positive definite kernels and universal Kriging
arXiv 2608.24283 · 2026-08-28RePolicy：用于智能体安全保障中安全策略调用的强化学习方法
RePolicy: Reinforcement Learning for Safety-Policy Invocation in Agent Safeguards
arXiv 2608.24275 · 2026-08-28NVIDIA Cosmos-H-Dreams：面向外科机器人的实时生成式物理仿真
NVIDIA Cosmos-H-Dreams: Real-Time Generative Physics Simulation for Surgical Robotics
arXiv 2608.24199 · 2026-08-28FireRedAudio：一种具有解耦连续表示的通用音频语言模型，用于理解与生成
FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation
arXiv 2608.24168 · 2026-08-28基于故障代码驱动的测试用例合成与密集奖励塑造的鲁棒代码强化学习
Robust Code RL via Faulty-Code-Driven Test case Synthesis and Dense Reward Shaping
arXiv 2608.24135 · 2026-08-28少量共享随机比特足以实现常数轮的几乎稳定匹配
A Few Shared Random Bits Suffice for Constant-Round Almost Stable Matching
arXiv 2608.24102 · 2026-08-28RetrievalFormer：用于高效近似最近邻检索和冷物品推荐的双编码器Transformer
Keeping the Index Open: The Recommendation-Side Cost of Shared Search and Recommendation
arXiv 2608.24079 · 2026-08-28拓扑偏向的资源约束塑造Hindmarsh-Rose振子网络的同步路径
Topology-Biased Resource Constraints Shape Synchronization Pathways in Hindmarsh-Rose Oscillator Networks
arXiv 2608.23993 · 2026-08-28
