2026-07论文中文摘要
Tokengeist：智能对话中的多轮归因追踪
Tokengeist: Multi-Turn Attribution Tracing in Agentic Conversations
arXiv 2607.22610 · 2026-07-28评估大语言模型作为动态系统的可解释控制器
Evaluating LLMs as Interpretable Controllers for Dynamical Systems
arXiv 2607.22609 · 2026-07-28可学习的预测不一定是可操作的：在线购买的适当修复维度
Learnable Predictions Need Not Be Actionable: Proper Repair Dimension for Online Buying
arXiv 2607.22608 · 2026-07-28临床试验管道揭示医疗保健领域人工智能的下一波浪潮：对8532项注册研究的多维分析
The Clinical Trial Pipeline Reveals the Next Wave of Artificial Intelligence in Healthcare: A Multidimensional Analysis of 8,532 Registered Studies
arXiv 2607.22607 · 2026-07-28大语言模型在医疗分诊中的社会经济推断：相同症状，不同邮政编码
Socioeconomic Inference in LLM Medical Triage: Same Symptoms, Different ZIP Code
arXiv 2607.22605 · 2026-07-28可持续生成式人工智能的谬误：欧盟数据中心环境监管的局限性及未来方向
The Fallacy of Sustainable Generative AI: Limitations in EU Environmental Regulation of Data Centres and Paths Forward
arXiv 2607.22604 · 2026-07-28DeepLook：通过前瞻进行更深入思考
DeepLook: Deeper Thinking with Lookahead
arXiv 2607.22602 · 2026-07-28视觉语言模型中的图表欺骗：从漏洞到缓解
Chart Deception in Vision-Language Models: From Vulnerability to Mitigation
arXiv 2607.22600 · 2026-07-28针对时间序列预测，向不确定成分差分扩散轨迹
Differencing the Diffusion Trajectory toward Uncertain Components for Time Series Forecasting
arXiv 2607.22599 · 2026-07-28面向维度建模课程的教学驱动型教师助手
A didactical-driven teacher assistant for a dimensional modeling course
arXiv 2607.22598 · 2026-07-28HyCE-RAG：用于可解释多跳问答的超图证据链检索增强生成
HyCE-RAG: Hypergraph Chain-of-Evidence Retrieval-Augmented Generation for Explainable Multi-hop Question Answering
arXiv 2607.22597 · 2026-07-28原子模拟的智能编排
An Agentic Orchestration of Atomistic Simulations
arXiv 2607.22596 · 2026-07-28xMIx：用于机械可解释性应用的高性能服务时间平台
xMIx: High-Performance Serving-Time Platform for Mechanistic Interpretability Apps
arXiv 2607.22595 · 2026-07-28达朗贝尔函数方程与正路径上的全局凸自由作用原理
d'Alembert's Functional Equation and a Globally Convex Free-Action Principle on Positive Paths
arXiv 2607.22594 · 2026-07-28非欧几里得带中的形状转变
Shape-Transition in Non-Euclidean Ribbons
arXiv 2607.22593 · 2026-07-28结构跨越尺度：用于检索增强生成的模式约束因果图
Structure Over Scale: Schema-Constrained Causal Graphs for RAG
arXiv 2607.22592 · 2026-07-28由大语言模型编排的未知环境中的词汇发现
Lexical discovery in unknown environments orchestrated by Large Language Models
arXiv 2607.22591 · 2026-07-28DRP-FLR：智能电网中用于灵活负荷调节的需求响应潜力的数据驱动评估
DRP-FLR: Data-Driven Assessment of Demand Response Potential for Flexible Load Regulation in Smart Grids
arXiv 2607.22590 · 2026-07-28二次型和扭曲最优传输中的菱形传输
Diamond transports in quadratic-form and distorted optimal transport
arXiv 2607.22589 · 2026-07-28ParBench：用于可靠评估大语言模型并行代码翻译的基准测试
ParBench: A Benchmark for Reliable Evaluation of LLM Parallel Code Translation
arXiv 2607.22588 · 2026-07-28TriSP：用于大语言模型的三信号结构化剪枝
TriSP: Tri-Signal Structured Pruning for Large Language Models
arXiv 2607.22587 · 2026-07-28MM-ShiftKV：用于多模态大语言模型的解码感知预填充阶段键值选择
MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models
arXiv 2607.22586 · 2026-07-28编码智能体中的支架效应：在编码智能体评估中利用选择作为隐藏变量
The Scaffold Effect in Coding Agents: Harness Choice as a Hidden Variable in Coding-Agent Evaluation
arXiv 2607.22585 · 2026-07-28用于检索增强生成的源感知重排：一种可靠性先验方法
Source-Aware Reranking for Retrieval-Augmented Generation: A Reliability Prior Approach
arXiv 2607.22584 · 2026-07-28用于带时间窗的时间依赖车辆路径问题（TDVRPTW）的分支定价与切割法
Branch \& Price \& Cut for the Time-Dependent Vehicle Routing Problem with Time Windows (TDVRPTW)
arXiv 2607.22582 · 2026-07-28带时间窗车辆路径问题的随机构造启发式算法：聚焦Regret-k
Randomized Constructive Heuristics for the VRPTW: A Focus on Regret-k
arXiv 2607.22581 · 2026-07-28具有一个快速服务器和两个相同慢速服务器的排队系统阈值策略的最优性
Optimality of a Threshold Policy for a Queueing System with One Fast Server and Two Identical Slow Servers
arXiv 2607.22580 · 2026-07-28通过修正的C类Ciric型压缩对非唯一不动点的统一方法
A unified approach to non-unique fixed points via modified C-class Ciric-type contractions
arXiv 2607.22579 · 2026-07-28HeraSys：通过细粒度端到端优化实现多个大语言模型工作流的协同服务
HeraSys: Collaborative Serving of Multiple LLM Workflows via Fine-Grained End-to-End Optimization
arXiv 2607.22578 · 2026-07-28大规模的cMoLLM：语言模型混合的水平扩展定律
cMoLLM at Scale: Horizontal Scaling Laws for Mixture-of-LLMs
arXiv 2607.22577 · 2026-07-28矩阵值函数的有理极小极大逼近：存在性、最优性与算法
Rational Minimax Approximations for Matrix-Valued Functions: Existence, Optimality and Algorithms
arXiv 2607.22576 · 2026-07-28时间上下文恢复驱动长上下文语言模型中的情景式顺序记忆
Temporal Context Reinstatement Drives Episodic-Like Order Memory in Long-Context Language Models
arXiv 2607.22575 · 2026-07-28证据过多，时间过少：通过多目标证据推理从文本到可操作的建议
Too much evidence, too little time: From text to actionable recommendations through multi-objective evidence reasoning
arXiv 2607.22574 · 2026-07-28PhononBench-MP40：用于声子稳定性的光谱分辨基准数据集
PhononBench-MP40: a spectrum-resolved benchmark dataset for phonon stability
arXiv 2607.22573 · 2026-07-28模式感知本地化（SAL）：用于Oracle NL2SQL的实时模式基础与幻觉验证
Schema-Aware Localisation (SAL): Live Schema Grounding and Hallucination Validation for Oracle NL2SQL
arXiv 2607.22572 · 2026-07-28SCAIR：用于企业知识图谱的模式条件代理迭代推理
SCAIR: Schema-Conditioned Agentic Iterative Reasoning for Enterprise Knowledge Graphs
arXiv 2607.22571 · 2026-07-28用于语言模型机制审计的参考特征图谱
Reference Feature Atlases for Mechanistic Auditing of Language Models
arXiv 2607.22570 · 2026-07-28软件工程流水线中编码代理的基于执行的安全测试
Execution-Grounded Security Testing for Coding Agents in Software Engineering Pipelines
arXiv 2607.22569 · 2026-07-28关键词很重要：揭示设备端大语言模型提示的能量敏感性
Keyword Matters: Unveiling the Energy Sensitivity of On-Device LLM Prompting
arXiv 2607.22568 · 2026-07-28MedLoCoMo：用于大语言模型的长上下文多会话医学对话基准
MedLoCoMo: A Long-Context Multi-Session Medical Dialogue Benchmark for Large Language Models
arXiv 2607.22566 · 2026-07-28DSTFView：基于双输入时空频率建模的多视图云边工作负载预测
DSTFView: Multi-View Cloud-Edge Workload Forecasting with Dual-Input Spatio-Temporal-Frequency Modeling
arXiv 2607.22565 · 2026-07-28使用多臂赌博机的卷积神经网络中的损失感知特征图剪枝
Loss-Aware Feature-Map Pruning in Convolutional Neural Networks Using Multi-Armed Bandits
arXiv 2607.22564 · 2026-07-28用于工业4.0智能体评估的合成场景生成
Synthetic Scenario Generation for Evaluation of Industry 4.0 Agents
arXiv 2607.22563 · 2026-07-28SF-AMS：大语言模型智能体中结构化记忆的策略性遗忘
SF-AMS: Strategic Forgetting for Structured Memory in LLM Agent
arXiv 2607.22562 · 2026-07-28通过贝叶斯优化实现高效多学科设计
Efficient multidisciplinary design via Bayesian optimization
arXiv 2607.22560 · 2026-07-28颗粒分数阶 Caputo-Katugampola 导数及其在模糊分数变分问题最优性条件中的应用
Granular fractional Caputo-Katugampola derivatives and their applications in optimality conditions for fuzzy fractional variational problems
arXiv 2607.22559 · 2026-07-28具有错误初始信息的主次线性二次平均场博弈：分布式误差估计与策略修正
Major-Minor LQ Mean Field Games with Erroneous Initial Information: Distributed Error Estimation and Strategy Modification
arXiv 2607.22558 · 2026-07-28具有实区间状态空间的不安分多臂老虎机的阈值可索引性：性能度量验证框架与长期平均分析
Threshold-indexability of restless bandits with real interval state spaces: a performance-metric verification framework and long-run average analysis
arXiv 2607.22557 · 2026-07-28MIITA：用于小语言模型持续学习的内存诱导推理时间自适应
MIITA: Memory-Induced Inference-Time Adaptation for Continual Learning with Small Language Models
arXiv 2607.22556 · 2026-07-28深度镜头诊断代理：代理工作流程设计使小型推理模型能够与前沿语言模型竞争
DeepLens Diagnosis Agent: Agentic Workflow Design Lets a Small Reasoning Model Compete with Frontier LLMs
arXiv 2607.22555 · 2026-07-28
