2026-07论文中文摘要
用于保护大语言模型生成的C代码的工具引导检索增强修复
Tool-Guided Retrieval-Augmented Repair for Securing LLM-Generated C Code
arXiv 2607.21641 · 2026-07-27电流体动力学对泄漏介电介质中液滴稳定性的影响
Influence of Electrohydrodynamic on Droplet Stability in Leaky Dielectric Media
arXiv 2607.21638 · 2026-07-27元强化学习的拟蒙特卡罗初始化
Quasi-Monte Carlo Initialization for Meta-Reinforcement Learning
arXiv 2607.21637 · 2026-07-27面向时间干预下个人语言模型代理的用户条件评估
Toward User-Conditioned Evaluation of Personal LLM Agents under Temporal Interventions
arXiv 2607.21635 · 2026-07-27MotifRole-Diff：用于掩码分子图扩散的风险最优角色感知损坏
MotifRole-Diff: Risk-Optimal Role-Aware Corruption for Masked Molecular Graph Diffusion
arXiv 2607.21634 · 2026-07-27一种基于共识的大语言模型相对偏好评估框架
A Consensus-Based Framework for Relative Preference Evaluation of Large Language Models
arXiv 2607.21632 · 2026-07-27Genesis：一个用于研究无适应度函数的开放式进化的实证平台
Genesis: An Empirical Platform for Studying Open-Ended Evolution Without Fitness Functions
arXiv 2607.21631 · 2026-07-27无适应度的自组织智能体演化：从约束驱动选择到发育编码的三个可证伪实验
Evolving Self-Organising Agents Without Fitness: Three Falsifiable Experiments from Constraint-Driven Selection to Developmental Encoding
arXiv 2607.21630 · 2026-07-27作为程序的散射振幅：理论与事件生成的自演化搜索
Scattering Amplitudes as Programs: Self-Evolving Search for Theory and Event Generation
arXiv 2607.21629 · 2026-07-27模块各司其职吗？复合语言模型系统中的角色漂移
Do Modules Stay in Their Lane? Role Drift in Compound LLM Systems
arXiv 2607.21627 · 2026-07-27离散动作空间是小模型连续控制中GRPO收敛的前提条件
Discrete Action Space as a Prerequisite for GRPO Convergence in Small-Model Continuous Control
arXiv 2607.21626 · 2026-07-27用于时间决策的轨迹感知检索代理
Trajectory-Aware Retrieval Agents for Temporal Decision- Making
arXiv 2607.21625 · 2026-07-27FBLayout：为在移动GPU上高效微调大语言模型优化内存布局
FBLayout: Optimizing Memory Layout for Efficient LLM Finetuning on Mobile GPUs
arXiv 2607.21624 · 2026-07-27云原生评估即服务：一种具有共形保证的可扩展人工智能监测微服务架构
Cloud-Native Evaluation-as-a-Service: A Microservices Architecture for Scalable AI Monitoring with Conformal Guarantees
arXiv 2607.21623 · 2026-07-27局部突触规则无需反向传播即可实现SIGReg梯度
Local Synaptic Rules Can Implement a SIGReg Gradient Without Backpropagation
arXiv 2607.21622 · 2026-07-27从概要到导向向量：用于个性化文本生成的全局稀疏先验和局部语义校准
From Profiles to Steering Vectors: Global Sparse Priors and Local Semantic Calibration for Personalized Text Generation
arXiv 2607.21620 · 2026-07-27对抗风格优化：通过基于GRPO的风格触发优化增强VLM越狱
Adversarial Style Optimization: Enhancing VLM Jailbreaks by GRPO-based Stylistic Triggers Optimization
arXiv 2607.21619 · 2026-07-27LeafData：一种用于数据迁移的智能系统
LeafData: An Agentic System for Data Migration
arXiv 2607.21618 · 2026-07-27迷失在上下文中：解决大语言模型中的上下文焦虑
Lost in Context: Addressing Context Anxiety in Large Language Models
arXiv 2607.21616 · 2026-07-27FrED：通过领域知识图谱基础进行外部数据影响估计
FrED: External Data Influence Estimation via Domain Knowledge Graph Grounding
arXiv 2607.21615 · 2026-07-27使用基于大语言模型的实体解析在混合格式居住数据中进行家庭移动检测
Household Movement Detection in Mixed-Format Occupancy Data Using LLM-Based Entity Resolution
arXiv 2607.21614 · 2026-07-27硬决策层：Transformer中确定推理的证据
The Hard Decision Layer: Evidence for Committed Inference in Transformers
arXiv 2607.21613 · 2026-07-27过程性知识不是低秩的：为什么LoRA无法内化多步骤过程
Procedural Knowledge Is Not Low-Rank: Why LoRA Fails to Internalize Multi-Step Procedures
arXiv 2607.21612 · 2026-07-27关于STEM领域图像描述技术的系统综述
A Systematic Survey on Image Description Techniques for STEM Domains
arXiv 2607.21611 · 2026-07-27SCOPE和SCION：用于文本模式归纳与融合的基准和可审计参考管道
SCOPE and SCION: A Benchmark and an Auditable Reference Pipeline for Schema Induction and Fusion from Text
arXiv 2607.21610 · 2026-07-27用于智能工作流合成的拓扑与执行耦合分层搜索
Coupled Hierarchical Search over Topology and Execution for Agentic Workflow Synthesis
arXiv 2607.21609 · 2026-07-27从义务到规范：关于在需求工程中验证欧盟人工智能法案要求的调查
From Obligation to Specification: A Survey on Validating EU AI Act Requirements in RE
arXiv 2607.21608 · 2026-07-27图神经网络中深度感知长距离传播的谱流证书
Spectral Flow Certificates for Depth-Aware Long-Range Propagation in Graph Neural Networks
arXiv 2607.21607 · 2026-07-27卫生职业教育中的自然语言处理：一项范围综述
Natural Language Processing in Health Professions Education: A Scoping Review
arXiv 2607.21605 · 2026-07-27AgentKVShift：用于智能体记忆系统的高效KV缓存重用
AgentKVShift: Efficient KV Cache Reuse for Agentic Memory Systems
arXiv 2607.21604 · 2026-07-27使用有序网络分析来分析中学生在协作式人工智能聊天机器人开发过程中的对话和行为
Analyzing Middle School Students' Dialogue and Behaviors during Collaborative AI Chatbot Development Using Ordered Network Analysis
arXiv 2607.21603 · 2026-07-27用于异构边缘设备上快速大语言模型筛选的可迁移延迟预测
Transferable Latency Prediction for Fast LLM Screening on Heterogeneous Edge Devices
arXiv 2607.21602 · 2026-07-27从帧级识别到事件级确认：公共空间手势交互的修复痕迹与运行时故障分析
From Frame-Level Recognition to Event-Level Confirmation: Repair Traces and Runtime Failure Analysis of Public-Space Gesture Interaction
arXiv 2607.21601 · 2026-07-27通过内部信息分解保护多模态人工智能
Securing Multimodal AI through Internal Information Decomposition
arXiv 2607.21600 · 2026-07-27解耦注意力融合：通过高效的键值缓存重用加速检索增强生成
Decoupled Attention Fusion: Accelerating RAG with Efficient KV Cache Reuse
arXiv 2607.21599 · 2026-07-27使用大语言模型的控制面板以澄清用户意图
Control panels to clarify user intent with Large Language Models
arXiv 2607.21598 · 2026-07-27幸福感感知中不公平和公平风险的来源
Sources of Inequity and Fairness Risks in Wellbeing Sensing
arXiv 2607.21527 · 2026-07-27无需云端的智能编码：在纵向数据准备任务中评估开放权重的大语言模型
Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks
arXiv 2607.21482 · 2026-07-27关系路径积分、有效作用量与量子框架协变性
Relational path integral, effective actions and quantum frame covariance in gravity
arXiv 2607.21463 · 2026-07-27通过错误定位进行测试时缩放
Test-Time Scaling via Error Localization
arXiv 2607.21453 · 2026-07-27来自手性耦合量子点的定向电信光子
Directional telecom photons from a chirally coupled quantum dot
arXiv 2607.21452 · 2026-07-27GrainGS：用于高效动态新视图合成的梯度解耦高斯点云渲染
GrainGS: Gradient-Decoupled Gaussian Splatting for Efficient Dynamic Novel View Synthesis
arXiv 2607.21448 · 2026-07-27近似反应概率函数：目标函数与训练数据采样
Approximating committor functions: Objective functions and training data sampling
arXiv 2607.21425 · 2026-07-27两相源与反应系数的斯蒂芬型问题
Two-phase source and reaction coefficient Stefan type problems
arXiv 2607.21388 · 2026-07-27模拟量子电路会改变卷积神经网络关注的内容吗？医学图像分类中的性能与可解释性比较
Do emulated quantum circuits change what CNNs look at? Performance and explainability comparison in medical image classification
arXiv 2607.21186 · 2026-07-27自旋介导对洛伦兹违反黑洞时空混沌边界违反的调制
Spin-mediated modulation of chaos bound violation in Lorentz-violating black hole spacetimes
arXiv 2607.21165 · 2026-07-27再转一圈，遗憾更少：基于遗憾的大语言模型澄清策略多轮基准测试
One More Turn, Less Regret: A Regret-Based Multi-Turn Benchmark for LLMs' Clarification Policies
arXiv 2607.21143 · 2026-07-27引导注意力：通过模仿学习实现面向OOD鲁棒机器人操纵的可解释和可校正视觉注意力
GuidedAttention: Interpretable and Correctable Visual Attention for OOD-Robust Robot Manipulation via Imitation Learning
arXiv 2607.21049 · 2026-07-27真空持续概率的边界补全
Boundary Completion of Vacuum Persistence Probability
arXiv 2607.20936 · 2026-07-27全息术中的量子姆潘巴效应
Quantum Mpemba effect in holography
arXiv 2607.20899 · 2026-07-27
