2026-08论文中文摘要

• 面向现实能源预测：一种时延增强的分数阶供需模型

  Toward Realistic Energy Forecasting: A Delay-Enhanced Fractional-Order Supply-Demand Model

  arXiv 2608.18121 · 2026-08-20

• 可交易伊藤特征：一种用于动态对冲的无模型、可解释框架

  Tradable Itô Signatures: A Model-Free, Interpretable Framework for Dynamic Hedging

  arXiv 2608.18120 · 2026-08-20

• 社会倾向的转变，稳定的亲社会特质：人类人格的全球年龄-时期-队列分析

  Shifting Social Dispositions, Stable Prosocial Traits: A Global Age-Period-Cohort Analysis of Human Personality

  arXiv 2608.18119 · 2026-08-20

• 基于生成式障碍证书的非线性系统形式化安全验证

  Formal Safety Verification for Nonlinear Systems with Generative Barrier Certificate

  arXiv 2608.18118 · 2026-08-20

• 立场：AI排行榜对全球南方的服务不足——来自印度的案例研究

  Position: AI Leaderboards Are Underserving the Global South: A Case Study from India

  arXiv 2608.18117 · 2026-08-20

• 你即你所提示的：农业食品视觉语言模型中的提示质量、领域偏移与不确定性

  You Are What You Prompt: Prompt Quality, Domain Shift, and Uncertainty in Agrifood Vision-Language Models

  arXiv 2608.18116 · 2026-08-20

• 面向 token 级幻觉检测的多信号时序融合

  Temporal Multi-Signal Fusion for Token-Level Hallucination Detection

  arXiv 2608.18115 · 2026-08-20

• 从非侵入式脑记录中准确解码自然句子

  Accurate Decoding of Natural Sentences from Non-Invasive Brain Recordings

  arXiv 2608.18114 · 2026-08-20

• 系统性风险平均场模型中的最优损失分配

  Optimal Loss Allocation in a Mean-Field Model of Systemic Risk

  arXiv 2608.18113 · 2026-08-20

• 对一个猜想的Schatten范数不等式的反例

  Counterexamples to a conjectured Schatten norm inequality

  arXiv 2608.18112 · 2026-08-20

• 求解不等于绘图：奥林匹克几何图解推理基准

  Solving Is Not Drawing: A Benchmark for Diagrammatic Reasoning in Olympiad Geometry

  arXiv 2608.18111 · 2026-08-20

• 智能体AI的涌现：关于演进、背景、工作原理、应用、采用因素及未来研究方向的综述

  Emergence of Agentic AI: A Review on Evolution, Background, Working Principles, Applications, Adoption Factors, and Future Research Directions

  arXiv 2608.18110 · 2026-08-20

• 叙事熵（Sn）的可操作化：双场景注册试点报告与预验证方案

  Operationalizing Narrative Entropy (Sn): A Two-Scene Registered Pilot Report and Pre-Validation Protocol

  arXiv 2608.18109 · 2026-08-20

• 相同事实，不同更新：推理设置塑造医疗分配场景下大语言模型的行为

  Same Facts, Different Updates: Inference Setup Shapes LLM Behavior in Medical Allocation

  arXiv 2608.18108 · 2026-08-20

• 大型语言模型中作为地理偏见的机构声望：来自三项带自助法置信区间的析因实验的证据

  Institutional Prestige as Geographic Bias in Large Language Models: Evidence from Three Factorial Experiments with Bootstrap Confidence Intervals

  arXiv 2608.18107 · 2026-08-20

• 言语化过度自信的不同方面：一项可解释性研究

  Different Facets of Verbalised Overconfidence: an Interpretability Study

  arXiv 2608.18106 · 2026-08-20

• StocksTalk：一种面向网络数据结构化查询生成的语音赋能对话智能体

  StocksTalk: A Voice-Enabled Conversational Agent for Structured Query Generation over Web Data

  arXiv 2608.18105 · 2026-08-20

• 自进化智能体作为动态图变换：一项综述与新视角

  Self-Evolving Agents as Dynamic Graph Transformation: A Survey and New Perspective

  arXiv 2608.18104 · 2026-08-20

• DeepTCM1.0：基于通用大语言模型的用于解析中药复方机制的多专家智能体

  DeepTCM1.0: A Multi-Expert AI Agent for Deciphering Mechanisms of Chinese Herbal Formulae Based on General Large Language Models

  arXiv 2608.18103 · 2026-08-20

• 面向机器生成文本鲁棒水印检测的感知稳定性特征设计

  Stability-Aware Feature Design for Robust Watermark Detection in Machine-Generated Text

  arXiv 2608.18102 · 2026-08-20

• BERTilda：基于相似度与流的时序图实现可解释的主题生命周期跟踪及分裂/合并检测

  BERTilda: Explainable Topic Lifecycle Tracking with Split/Merge Detection via Similarity-and-Flow Temporal Graphs

  arXiv 2608.18101 · 2026-08-20

• 计算东方主义：使用中东文化敏感性评分（MECSS）测量大语言模型中的结构性话语偏见

  Computational Orientalism: Measuring Structural Discourse Bias in Large Language Models Using the Middle East Cultural Sensitivity Score (MECSS)

  arXiv 2608.18100 · 2026-08-20

• FinSkillBench：评估用于投资管理的智能体AI与领域技能

  FinSkillBench: Evaluating AI Agents and Domain Skills for Investment Management

  arXiv 2608.18099 · 2026-08-20

• 分数衰减键值缓存：面向对话系统提升推理相关性的所有权感知内存管理

  Fractional Decay KV-Cache: Ownership-Aware Memory Management for Improved Inference Relevancy in Dialog Systems

  arXiv 2608.18098 · 2026-08-20

• FrenchNews-7：跨出版商法国新闻编辑部分类基准

  FrenchNews-7: Benchmarking Cross-Publisher French News Editorial Desk Classification

  arXiv 2608.18097 · 2026-08-20

• MAVEN：基于紧凑对齐评估器的多模态内容宏观社会价值评估框架

  MAVEN: A Macro-Societal Value Evaluation Framework of Multimodal Content with Compact Aligned Evaluators

  arXiv 2608.18096 · 2026-08-20

• 语言模型与视觉-语言模型中的后门学习

  Backdoor Learning in Language Models and Vision-Language Models

  arXiv 2608.18095 · 2026-08-20

• NE-BERT：适用于9种印度东北部语言的多语言语言模型

  NE-BERT: A Multilingual Language Model for Nine Northeast Indian Languages

  arXiv 2608.18094 · 2026-08-20

• 通过拒绝别名缓解删除（Abliteration）攻击

  Abliteration Mitigation via Refusal Aliases

  arXiv 2608.18093 · 2026-08-20

• 立场：多智能体系统应优先考虑并发控制

  Position: Multi-Agent Systems Should Prioritize Concurrency Control

  arXiv 2608.18092 · 2026-08-20

• 九个情绪质心：跨四种模态迁移的无标签效价轴

  Nine Emotion Centroids: A Label-Free Valence Axis That Transfers Across Four Modalities

  arXiv 2608.18090 · 2026-08-20

• 低资源非洲语言的潜在空间拒绝锚定：无需重新训练的机制安全恢复

  Latent Space Refusal Anchoring for Low-Resource African Languages: Mechanistic Safety Recovery Without Retraining

  arXiv 2608.18089 · 2026-08-20

• 一种用于基于飞行日志的无人机螺旋桨健康监测的变形人工年龄评分决策支持原型

  A Metamorphic Artificial Age Score Decision-Support Prototype for Flight-Log-Based Drone Propeller Health Monitoring

  arXiv 2608.18088 · 2026-08-20

• 立场：当前模型卡片不足以支持开放权重基础模型的下游治理

  Position: Current Model Cards Are Insufficient for Downstream Governance of Open-Weight Foundation Models

  arXiv 2608.18086 · 2026-08-20

• 面向任务型对话的角色引导大语言模型智能体

  Persona-Guided LLM Agents for Task-Oriented Dialogue

  arXiv 2608.18085 · 2026-08-20

• 定位：行为系统需要行为测试

  Position: Behavioral Systems Require Behavioral Tests

  arXiv 2608.18081 · 2026-08-20

• 心理健康领域的大语言模型：应用、创新与伦理挑战的系统综述

  Large Language Models in Mental Health: A Systematic Review of Applications, Innovations, and Ethical Challenges

  arXiv 2608.18080 · 2026-08-20

• Position：通过转移复杂度分析游戏世界

  Position: Profiling Game Worlds by Transition Complexity

  arXiv 2608.18079 · 2026-08-20

• 立场：AI推理智能体间的合谋风险证明其在制定市场决策时需满足认证要求

  Position: Collusion Risks Among AI Reasoning Agents Justify Certification Requirements for Making Market Decisions

  arXiv 2608.18078 · 2026-08-20

• 用于无监督动态对比增强MRI重建的基元表示学习

  Primitive Representation Learning for Unsupervised Dynamic Contrast Enhanced MRI Reconstruction

  arXiv 2608.18055 · 2026-08-20

• 混合范数Brascamp-Lieb不等式

  Mixed-norm Brascamp-Lieb inequalities

  arXiv 2608.17952 · 2026-08-20

• 超越器械运动：面向手术技能评估的组织张力识别

  Beyond Instrument Motion: Recognizing Tissue Tension Toward Surgical Skill Assessment

  arXiv 2608.17935 · 2026-08-20

• 元素不同性查询的平均情况最优编码与高效最坏情况索引

  Average-Case Optimal Encodings and Efficient Worst-Case Indices for Element Distinctness Queries

  arXiv 2608.17907 · 2026-08-20

• 关于立方图中(1,2)-控制的两个猜想的反例

  Counterexamples to two conjectures on (1, 2)-domination in cubic graphs

  arXiv 2608.17851 · 2026-08-20

• 共形爱因斯坦反自对偶空间及其推广

  Conformally Einstein anti-self-dual spaces and their generalisations

  arXiv 2608.17847 · 2026-08-20

• 利用低成本衍射光学元件实现稳健的轨道角动量传递

  Robust Orbital Angular Momentum Transfer Using Low-Cost Diffractive Optics

  arXiv 2608.17816 · 2026-08-20

• 尺度重要：跨物种三维植物器官分割的自适应粒度选择

  Scale Matters: Adaptive Granularity Selection for Cross-Species 3D Plant Organ Segmentation

  arXiv 2608.17803 · 2026-08-20

• 最大相交问题的k维版本

  A $k$-Dimensional Version of the Largest Intersection Problem

  arXiv 2608.17771 · 2026-08-20

• D²ACCI：一种用于保留证据的智能体记忆的双循环诊断协议

  D$^2$ACCI: A Dual-Loop Diagnostic Protocol for Evidence-Preserving Agent Memory

  arXiv 2608.17756 · 2026-08-20

• 爆炸式DecPOMDPs的奇特案例：通过策略计数控制这一“火势”

  The Curious Case of Exploding DecPOMDPs: Containing the Fire through Policy Counting

  arXiv 2608.17749 · 2026-08-20