2026-07论文中文摘要
FinCacheServe：面向可变企业文档的高性价比RAG服务的依赖一致答案复用
FinCacheServe: Dependency-Consistent Answer Reuse for Cost-Efficient RAG Serving over Mutable Enterprise Documents
arXiv 2607.26076 · 2026-07-30推荐系统中的可复现性：一项综述
Reproducibility in Recommender Systems: A Survey
arXiv 2607.26074 · 2026-07-30猜猜你要去哪里：高德地图中的生成式下一个兴趣点推荐
Guess Where You Go: Generative Next Point-of-Interest Recommendation in Amap
arXiv 2607.26073 · 2026-07-30IFCMemoryBench：评估基于大语言模型（LLM）的智能体在建筑信息模型（BIM）信息检索中的长期记忆能力
IFCMemoryBench: Evaluating Long-Term Memory of LLM-Based Agents in BIM Information Retrieval
arXiv 2607.26072 · 2026-07-30GuidedRAG：检索增强生成的语义引导
GuidedRAG: Semantic Steering of Retrieval-Augmented Generation
arXiv 2607.26071 · 2026-07-30SimpleWikiSearch：用于智能体搜索的简洁离线维基百科环境
SimpleWikiSearch: A Clean Offline Wikipedia Environment for Agentic Search
arXiv 2607.26070 · 2026-07-30AI安全优先级：全领域议程
AI Security Priorities: A Field-Wide Agenda
arXiv 2607.26069 · 2026-07-30人类效用因子：一种可计算的福利度量，将AI治理重新构建为约束优化问题
The Human Utility Factor: A Computable Welfare Metric That Reframes AI Governance as a Constrained Optimisation Problem
arXiv 2607.26068 · 2026-07-30易陷阱：为何大语言模型低估由误解驱动的难度
The Easy Trap: Why LLMs Underestimate Misconception-Driven Difficulty
arXiv 2607.26067 · 2026-07-30方法是否支持其主张？用于同行评审的论文内部验证
Do Methods Support the Claims? Intra-Paper Verification for Peer Review
arXiv 2607.26066 · 2026-07-30核岭回归与Hölder-Zygmund类的结合：极小极大最优性与正则化恰当性的失效
When Kernel Ridge Regression Meets the Hölder-Zygmund Class: Minimax Optimality and Failure of Properness
arXiv 2607.26065 · 2026-07-30AI智能体时代需要新的科学范式以维持可信科学
The Age of AI Agents Demands A New Scientific Paradigm To Sustain Trustworthy Science
arXiv 2607.26064 · 2026-07-30原型还是能力？用于建模学生数学能力的聚类方法
Archetypes or ability? Clustering for modelling student mathematical competence
arXiv 2607.26063 · 2026-07-30识别基于大语言模型（LLM）的聊天人工智能对智力障碍人群的隐含偏见
Identifying Implicit Bias in LLM-based Chat AI Toward People with Intellectual Disabilities
arXiv 2607.26062 · 2026-07-30Sim2Win：一种适用于足球的、与球队无关的、基于事件的赛前结果预测与战术画像系统
Sim2Win: A Team-Agnostic, Event-Based Pre-Match Outcome Prediction and Tactical Profiling System for Football
arXiv 2607.26061 · 2026-07-30基于客户数字孪生模拟的大规模聊天机器人验证
Large-Scale ChatBot Validation Through Customer Digital Twin Simulations
arXiv 2607.26060 · 2026-07-30深度强化学习中冻结的随机CNN特征提取器的涌现稀疏性
Emergent Sparsity in Frozen Random CNN Feature Extractors for Deep Reinforcement Learning
arXiv 2607.26059 · 2026-07-30MAC-Gyver：面向AI-RAN 6G系统的开放、可编程调度
MAC-Gyver: Open, Programmable, Scheduling for AI-RAN 6G Systems
arXiv 2607.26012 · 2026-07-30\textsc{IH - 基准测试}：用于大语言模型应用中指令层次鲁棒性的以冲突为中心的基准测试
IH-Benchmark: A Conflict-Centered Benchmark for Instruction-Hierarchy Robustness in LLM Applications
arXiv 2607.25987 · 2026-07-30SplatStream：用于自适应3D场景流的细粒度可扩展高斯点云绘制
SplatStream: Fine Granular Scalable Gaussian Splatting for Adaptive 3D Scene Streaming
arXiv 2607.25971 · 2026-07-30用于视频级矛盾心理和犹豫识别的交互流知识引导多模态推理
Knowledge-Guided Multimodal Reasoning over Interacting Streams for Video-Level Ambivalence and Hesitancy Recognition
arXiv 2607.25961 · 2026-07-30跨文本、表格和知识图谱检测知识不一致性
Detecting Knowledge Inconsistencies Across Text, Tables, and Knowledge Graphs
arXiv 2607.25959 · 2026-07-30Hermes：通过前缀共识实现低尾部延迟
Hermes: Low Tail-Latency Via Prefix Consensus
arXiv 2607.25916 · 2026-07-30交互式奖励智能体：通过环境状态验证进行图形用户界面任务评估
Interactive Reward Agent: GUI Task Evaluation via Environment-State Verification
arXiv 2607.25904 · 2026-07-30无共同量子过程的时空上的一致量子态
When Quantum States over Spacetime Have No Common Process
arXiv 2607.25899 · 2026-07-30AngelSpec：通过推测解码实现现实世界中的高性能推理
AngelSpec: Towards Real-World High Performance Inference with Speculative Decoding
arXiv 2607.25852 · 2026-07-30对具有随机缺失中介变量和结果数据的因果可解释元中介分析
Causally Interpretable Meta-Mediation Analysis With Missing At Random Mediator and Outcome Data
arXiv 2607.25822 · 2026-07-30在星际介质中检测 HNCOH+
Chemical modelling of interstellar MgS
arXiv 2607.25758 · 2026-07-30平衡度量与SYZ猜想
Balanced Metrics Know About SYZ
arXiv 2607.25733 · 2026-07-30工具并非孤岛：通过查询条件超边预测为语言模型智能体进行集合级工具检索
Tools Are Not Islands: Set-Level Tool Retrieval for LLM Agents via Query-Conditioned Hyperedge Prediction
arXiv 2607.25718 · 2026-07-30扭曲双层石墨烯中的阿哈罗诺夫 - 卡什尔相位
Aharonov-Casher phase in twisted bilayer graphene
arXiv 2607.25660 · 2026-07-30模型上下文协议应用的实证研究
An Empirical Study of Model Context Protocol Applications
arXiv 2607.25635 · 2026-07-30斯蒂费尔流形上分布的全拟合优度检验
Omnibus Goodness-of-Fit Testing for Distributions on Stiefel Manifolds
arXiv 2607.25618 · 2026-07-30超越自我认知：在语言模型中跨推理与检索传播不确定性
Beyond Self-Knowledge: Propagating Uncertainty Across Reasoning and Retrieval in LLMs
arXiv 2607.25600 · 2026-07-30向量空间中相交定理的射影奥雷度条件
Projective Ore-Degree Conditions for Intersection Theorems in Vector Spaces
arXiv 2607.25598 · 2026-07-30LAIA数据集：智能汽车的标记注意力
The LAIA Dataset: Labelled Attention for Intelligent Automobiles
arXiv 2607.25570 · 2026-07-30I2VShield：一种针对基于DiT的图像到视频模型的高效主动防御框架
I2VShield: An Efficient Proactive Defense Framework against DiT-based Image-to-Video Models
arXiv 2607.25522 · 2026-07-30具有可控误差权衡的作物损害评估安全感知级联推理
Safety-Aware Cascaded Inference for Crop Damage Assessment with Controlled Error Trade-offs
arXiv 2607.25468 · 2026-07-30基于系统误差量化的核磁偶极矩的从头算晶格计算
Ab initio lattice calculation of nuclear magnetic dipole moments with systematic error quantifications
arXiv 2607.25464 · 2026-07-30激光干涉空间天线（LISA）随机引力波背景推断中残余银河系双星前景：源功率集中与频谱简并
Residual Galactic binary foreground in LISA stochastic gravitational-wave background inference: source power concentration and spectral degeneracy
arXiv 2607.25349 · 2026-07-30时间距离联合嵌入预测架构：用于潜在世界模型预测控制的计划感知表示学习
Temporal-Distance JEPA: Plan-Aware Representation Learning for Latent World Model Predictive Control
arXiv 2607.25337 · 2026-07-30空间引导场理论：从梯度假设和电荷守恒到规范群结构
Spatial Guidance Field Theory: From Gradient Postulates and Charge Conservation to Gauge Group Structure
arXiv 2607.25336 · 2026-07-30每次我雇佣一位语言学家，推理成本就会降低：论语言规则作为有效的提示压缩器
Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors
arXiv 2607.25335 · 2026-07-30用于罕见事件比例的β-二项式元分析的惩罚似然推断
Penalized likelihood inference for beta-binomial meta-analysis of proportions of rare events
arXiv 2607.25320 · 2026-07-30用户提问，平台竞争：代理推荐市场如何形成
The User Asks, Platforms Compete: How Agentic Recommendation Markets Take Shape
arXiv 2607.25253 · 2026-07-30WHTMix：通过沃尔什-哈达玛令牌混合实现高效立体深度估计
WHTMix: Efficient Stereo Depth Estimation via Walsh-Hadamard Token Mixing
arXiv 2607.25234 · 2026-07-30基于智能体的人工智能助力大规模睡眠生理学研究
Agentic AI-enabled discovery across large-scale sleep physiology
arXiv 2607.25175 · 2026-07-30用于非关联莫尔-库仑模型岩土材料弹塑性分析的傅里叶特征物理信息神经网络
Fourier Feature Physics-Informed Neural Networks for Elasto-Plastic Analysis of Geomaterials with a Non-Associative Mohr-Coulomb Model
arXiv 2607.25150 · 2026-07-30通过稀疏自编码器实现可解释的隐藏规则游戏智能体
Interpretable GOHR Agents via Sparse Autoencoders
arXiv 2607.25132 · 2026-07-30通过含义识别和取消评估大语言模型中的交际信念更新
Evaluating Communicative Belief Updates in Large Language Models via Implicature Recognition and Cancellation
arXiv 2607.25094 · 2026-07-30
