2026-08论文中文摘要
基于模块化大语言模型（LLM）的安全智能体的事后轨迹风险验证
Post-Hoc Trajectory-Risk Certification for Modular LLM-Based Security Agents
arXiv 2608.05199 · 2026-08-07波动率曲面的数学理论
The Mathematics of Volatility Surfaces
arXiv 2608.05198 · 2026-08-07MS-MLB：一个用于血液基质-MS分类的开源机器学习基准
MS-MLB: An Open Machine Learning Benchmark for Blood-Based MS Classification
arXiv 2608.05196 · 2026-08-07CLARA：通过结果分析澄清自然语言癌症基因组学查询中的语言歧义
CLARA: Clarification of Language Ambiguity through Result Analysis for Natural-Language Cancer Genomics Queries
arXiv 2608.05195 · 2026-08-07非阿基米德柯西-施瓦茨角-长度不等式与切比雪夫算术平均不等式
Non-Archimedean Cauchy-Schwarz Angle-Length and Chebyshev Arithmetic Mean Inequalities
arXiv 2608.05194 · 2026-08-07涉及中心二项式系数的无穷级数与有限级数及广义超几何函数的闭式
Infinite and finite series involving central binomial coefficients and closed forms of generalized hypergeometric functions
arXiv 2608.05193 · 2026-08-07关于前两大素因子平均值的刚性
Rigidity of Averages over the Two Largest Prime Factors
arXiv 2608.05191 · 2026-08-07桥与火炬问题的若干推广
Some Generalizations of the Bridge and Torch Problem
arXiv 2608.05190 · 2026-08-07格雷戈里函数及其完全格雷戈里变换
The Gregory function and its completed Gregory transform
arXiv 2608.05189 · 2026-08-07立场：是时候针对自一致性优化大型语言模型（LLMs）了
Position: It's Time to Optimize LLMs for Self-Consistency
arXiv 2608.05188 · 2026-08-07基于关联的胞腔复形组合几何
Incidence-based Combinatorial Geometry on Cell Complexes
arXiv 2608.05186 · 2026-08-076G中的媒体与通信：基础、关键技术及应用
Media Meets Communication in 6G: Fundamentals, Key Technologies, and Applications
arXiv 2608.05184 · 2026-08-07社交网站用户死亡的自动检测
Automatic Detection of Deaths from Social Networking Sites
arXiv 2608.05183 · 2026-08-07优先连接网络中基于似然的异常检测
Likelihood-based anomaly detection in preferential attachment networks
arXiv 2608.05182 · 2026-08-07超越人口统计学：建筑信息模型（BIM）参与度与AEC行业从业者工作满意度——一项机器学习试点研究
Beyond Demographics: BIM Engagement and Job Satisfaction Among AEC Professionals, A Machine Learning Pilot Study
arXiv 2608.05181 · 2026-08-07核决策基准：评估前沿大语言模型的核倾向
The Nuclear Decision-Making Benchmark: Evaluating Frontier LLMs on Nuclear Tendencies
arXiv 2608.05180 · 2026-08-07自主研究智能体：AI科学家与验证差距的综述
Autonomous Research Agents: A Survey of AI Scientists and the Verification Gap
arXiv 2608.05179 · 2026-08-07谁能获得权限？AI生成的学术守门场景中的全球区域与学术地位偏差
Who Gets Access? Global Region and Academic Status Bias in AI-Generated Academic Gatekeeping Scenarios
arXiv 2608.05178 · 2026-08-07利用AI生成的反馈提升批判性思维与写作能力
Using AI-Generated Feedback to Improve Critical Thinking and Writing Proficiency
arXiv 2608.05177 · 2026-08-07当工具能完成作业时教授人工智能入门课程：课程重新设计与学生权利法案
Teaching Intro AI When the Tools Can Do the Homework: A Course Redesign and a Student Bill of Rights
arXiv 2608.05175 · 2026-08-07人类代码中的艺术
Art in Humanity's Code
arXiv 2608.05174 · 2026-08-07关闭的窗口：政府可能如何失去约束先进人工智能的能力
The Closing Window: How Governments Could Lose Their Ability to Restrain Advanced AI
arXiv 2608.05173 · 2026-08-07估算工作任务所花费的时间
Estimating time spent on work tasks
arXiv 2608.05172 · 2026-08-07超越信息检索：生成式AI作为认知仲裁者以增强协作问题解决
Beyond Information Retrieval: Generative AI as an Epistemic Arbiter to Enhance Collaborative Problem-Solving
arXiv 2608.05171 · 2026-08-07DREAM：基于大语言模型的事件感知记忆图动态角色扮演
DREAM: LLM-based Dynamic Role-playing via Event-Aware Memory Graph
arXiv 2608.05170 · 2026-08-07啄木鸟蒸馏：弱模型诊断强模型中的推理错误
Woodpecker Distillation: Weak Models Diagnose Reasoning Bugs in Strong Models
arXiv 2608.05168 · 2026-08-07CNM-BERT：基于表意文字描述序列的汉字可插入式结构嵌入
CNM-BERT: A Drop-In Structural Embedding for Chinese Characters via Ideographic Description Sequences
arXiv 2608.05167 · 2026-08-07大语言模型中的条件认知偏差：有偏差的用户对话轮次如何调控上下文推理
Conditional Cognitive Biases in LLMs: How Biased User Turns Modulate In-Context Reasoning
arXiv 2608.05166 · 2026-08-07语言模型中的跨架构引导迁移：一项系统实证研究
Cross-Architecture Steering Transfer in Language Models: A Systematic Empirical Study
arXiv 2608.05164 · 2026-08-07英语源多语言检索增强生成（RAG）系统的隐私风险所在：五种查询语言下的阶段分解审计
Where Privacy Risk Lives in English-Source Multilingual RAG: A Stage-Decomposed Audit Across Five Query Languages
arXiv 2608.05163 · 2026-08-07PoolBench：面向仅解码器大语言模型概念表示评估的池化策略基准
PoolBench: A Benchmark for Pooling Strategies in Concept Representation Evaluation for Decoder-Only LLMs
arXiv 2608.05162 · 2026-08-07SemiAdapt-Instruct：通过潜在领域专用适配器实现可扩展的指令微调
SemiAdapt-Instruct: Extensible Instruction Tuning via Latent Domain-Specialised Adapters
arXiv 2608.05161 · 2026-08-07点火指数：测量语言模型中的全局工作空间动力学
The Ignition Index: Measuring Global Workspace Dynamics in Language Models
arXiv 2608.05160 · 2026-08-07智能体嵌套：现有企业应用集成与服务的新方法论
Agentic Nesting: A New Methodology for Existing Enterprise Application Integration and Services
arXiv 2608.05159 · 2026-08-07基于回路锚点的安全进化
Safe Evolution with Circuit Anchors
arXiv 2608.05158 · 2026-08-07大语言模型对双盲评审构成威胁
Large Language Models Threaten Double-blind Review
arXiv 2608.05157 · 2026-08-07支架介导的后训练：协同演化模型参数与程序性支架图
Scaffold-Mediated Post-Training: Co-Evolving Model Parameters and Procedural Scaffold Graphs
arXiv 2608.05156 · 2026-08-07超越情感：对比传统NLP与基于大语言模型的多维分析在政治新闻评估中的应用
Beyond Sentiment: Comparing Traditional NLP and LLM-Based Multi-Dimensional Analysis for Political News Evaluation
arXiv 2608.05155 · 2026-08-07普遍存在的缺陷与有条件的后果：面向多跳可追溯性的RAG的三重鲁棒性分析
Universal Pathologies, Conditional Consequences: A Triple-Robustness Analysis of RAG for Multi-Hop Traceability
arXiv 2608.05153 · 2026-08-07大型语言模型中思维链推理的平均场动力学
Mean-Field Dynamics of Chain-of-Thought Reasoning in Large Language Models
arXiv 2608.05152 · 2026-08-07面向工业因果推理的模拟器驱动大语言模型：用于污水处理决策支持的工具使用、结构化注入及工厂可迁移检索
Simulator-Grounded Large Language Models for Industrial Causal Reasoning: Tool-Use, Structured Injection, and Plant-Portable Retrieval for Wastewater Treatment Decision Support
arXiv 2608.05151 · 2026-08-07QCD轴子与中微子质量
The QCD Axion and Neutrino Masses
arXiv 2608.05150 · 2026-08-07损失函数看不到基，但Adam可以
The Loss Does Not See the Basis, but Adam Does
arXiv 2608.05136 · 2026-08-07OPD-V：结合模态平衡的视觉在线策略自蒸馏
OPD-V: Visual On-Policy Self-Distillation with Modality Balance
arXiv 2608.05131 · 2026-08-07IRIS：一种受视觉皮层启发的用于分析视觉Transformer中方向选择性的框架
IRIS: A Visual Cortex-Inspired Framework for Analyzing Orientation Selectivity in Vision Transformers
arXiv 2608.05122 · 2026-08-07基于潜在流匹配的多模态时空大气数据同化
Multimodal Spatiotemporal Atmospheric Data Assimilation with Latent Video Flow-matching
arXiv 2608.05103 · 2026-08-07Heston型随机动力学的弱形式恢复
Weak Form Recovery of Heston Type Stochastic Dynamics
arXiv 2608.05009 · 2026-08-07迈向多模态预训练的物理学：知识流、模态协同、早期统一与方法指南
Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes
arXiv 2608.05000 · 2026-08-07面向有效边界表示（B-Rep）生成：无需训练的线框异常检测与修复
Towards Valid B-Rep Generation: Training-Free Wireframe Anomaly Detection and Repair
arXiv 2608.04955 · 2026-08-07强化目标语言特征：基于SAE的多语言推理引导
Strengthening Target-Language Features: SAE-Based Steering for Multilingual Inference
arXiv 2608.04904 · 2026-08-07
