2026-08论文中文摘要
预测具有经济意义的比特币走势：采用利润优化阈值的多尺度TCN
Forecasting Economically Significant Bitcoin Moves: A Multi-Scale TCN with Profit-Optimized Thresholds
arXiv 2608.26174 · 2026-08-28ClassVision：AI驱动的课堂考勤系统
ClassVision: AI-Powered Classroom Attendance System
arXiv 2608.26173 · 2026-08-28DRL：用于模式图扩展下事务安全的企业级NL2SQL的确定性关系中间件层
DRL: A Deterministic Relational Middleware Layer for Transaction-Safe Enterprise NL2SQL Under Schema-Graph Scaling
arXiv 2608.26172 · 2026-08-28预注册外部评估在三个转录组基础模型中产生一致的部分重复类别
Pre-Registered External Evaluation Yields a Consistent Partial-Replication Category across Three Transcriptomic Foundation Models
arXiv 2608.26170 · 2026-08-28基于多尝试编程轨迹的感知修订的提交成功预测
Revision-Aware Success Prediction from Multi-Attempt Programming Trajectories
arXiv 2608.26169 · 2026-08-28大语言模型中的幻觉：基于生命周期的原因、检测、缓解与预防调查
Hallucinations in LLMs: A Lifecycle-Based Survey of Causes, Detection, Mitigation, and Prevention
arXiv 2608.26168 · 2026-08-28拒绝不等同于鲁棒性：在经证实无信息的临床疼痛语音转录本上审计大型语言模型的自信编造
Refusal Is Not Robustness: Auditing Confident Fabrication in Large Language Models on a Provably Uninformative Clinical Pain Speech Transcript
arXiv 2608.26167 · 2026-08-28以用户为中心的思维链推理提升大语言模型可解释性
Improving LLM Interpretability with User-Centric Chain-of-Thought Reasoning
arXiv 2608.26166 · 2026-08-28使用多编码器（Poly-Encoders）实现计算高效的自动化创造力评估
Using Poly-Encoders for Computationally Efficient Automated Creativity Assessment
arXiv 2608.26165 · 2026-08-28以任务为中心的本体与确定性领域规则：AI辅助化学问题求解的可验证核心
A Task-Centric Ontology and Deterministic Domain Rules as a Verifiable Core for AI-Assisted Chemistry Problem Solving
arXiv 2608.26164 · 2026-08-28从声音到症状：面向对话式医疗智能体的实时呼吸信号理解
From Sound to Symptom: Real-Time Respiratory Signal Understanding for Conversational Healthcare Agents
arXiv 2608.26163 · 2026-08-28用于心理健康支持的安全门控多模态AI后端：Anian中的分层状态表示、保守风险融合与受控生成
A Safety-Gated Multimodal AI Backend for Mental-Health Support: Hierarchical State Representation, Conservative Risk Fusion, and Controlled Generation in Anian
arXiv 2608.26162 · 2026-08-28基于双种子比较的互去偏方法用于大语言模型的概率采样
Mutual Debiasing via Dual-Seed Comparison for Probabilistic Sampling in Large Language Models
arXiv 2608.26161 · 2026-08-28面向边-雾-云连续体的基于SAREF的分布式AI工作流本体
SAREF-based Ontology for Distributed AI Workflows across the Edge-Fog-Cloud Continuum
arXiv 2608.26160 · 2026-08-28加密货币市场中基于刻度与分钟的信息条的频率控制比较
A Frequency-Controlled Comparison of Tick- and Minute-Based Information Bars for Cryptocurrency Markets
arXiv 2608.26158 · 2026-08-28GROUND：通过受监管的语义定义减少基于大语言模型的企业分析中的幻觉现象
GROUND: Reducing Hallucinations in LLM-Based Enterprise Analytics Through Governed Semantic Definitions
arXiv 2608.26157 · 2026-08-28零售智能中的选择偏差校正
Selection Bias Correction in Retail Intelligence
arXiv 2608.26156 · 2026-08-28VFA：通过无视觉适配赋能多语言多模态大语言模型
VFA: Empowering Multilingual MLLMs via Vision-Free Adaptation
arXiv 2608.26155 · 2026-08-28评估面向癌症患者的AI生成摘要
Evaluating AI Generated Summaries for Cancer Patients
arXiv 2608.26154 · 2026-08-28EEG-to-Report：用于在临床脑电图上训练语言模型的标注与特征-文本框架
EEG-to-Report: An Annotation and Feature-Text Framework for Training Language Models on Clinical EEG
arXiv 2608.26153 · 2026-08-28面向电信客户流失预测的可解释人工智能：一种CRM集成框架
Explainable Artificial Intelligence for Customer Churn Prediction in Telecommunications: A Framework for CRM Integration
arXiv 2608.26151 · 2026-08-28利用大型语言模型开展疾病传播模型的系统文献综述
Leveraging Large Language Models for Systematic Literature Review of Disease Spread Models
arXiv 2608.26150 · 2026-08-28用于五维多表分析的方法与概念框架：复杂数据复用的统一方法
Methodological and Conceptual Framework for 5D Multi-Table Analysis: A Unified Approach for Complex Data Reuse
arXiv 2608.26149 · 2026-08-28面向可解释的抑郁检测：将声学特征与DSM-5指标关联
Towards Interpretable Depression Detection: Linking Acoustic Features to DSM-5 Indicators
arXiv 2608.26148 · 2026-08-28CARE：面向医学大语言模型的因果对齐推理探索
CARE: Causally-Aligned Reasoning Exploration for Medical Large Language Models
arXiv 2608.26147 · 2026-08-28Vagdhenu：一个知晓格律（Vrutta）的梵文颂歌（Shloka）转吟唱（TTS）系统
Vagdhenu: A Vrutta (Meter) Aware Shloka-to-Chant (TTS) System for Sanskrit
arXiv 2608.26146 · 2026-08-28用于学术工作流的大语言模型：对使用大语言模型短、长上下文窗口生成的文献综述的评估
LLMs for Academic Workflows: An Evaluation of Literature Reviews Generated with Short and Long Context Windows of LLMs
arXiv 2608.26145 · 2026-08-28为何当前可解释人工智能（XAI）对于阿拉伯语自然语言处理（NLP）而言仍不足：可解释性差距的批判性综述
Why Current XAI Is Not Enough for Arabic NLP: A Critical Survey of the Explainability Gap
arXiv 2608.26144 · 2026-08-28超越准确率：对用于表情包仇恨言论检测的视觉-语言模型的定性分析
Beyond Accuracy: A Qualitative Analysis of Vision-Language Models for Hate Speech Detection in Memes
arXiv 2608.26143 · 2026-08-28位置即你所需：一种用于基于多模态大语言模型的指代表达分割的免费午餐式令牌压缩策略
Position Is All You Need: A Free Lunch Token Compression Strategy for MLLM-based Referring Expression Segmentation
arXiv 2608.26142 · 2026-08-28AdaThinking-E：用于自适应思考的单token熵调控
AdaThinking-E: One-Token Entropy Regulation for Adaptive Thinking
arXiv 2608.26141 · 2026-08-28句法与语义：Transformer如何学习深层依赖关系
Syntax vs. Semantics: How Transformers Learn Deep Dependencies
arXiv 2608.26139 · 2026-08-28心理健康自然语言处理中的跨平台泛化失效：对社交媒体上Transformer模型的五维度公平性审计
Cross-Platform Generalisation Failure in Mental Health Natural Language Processing: A Five-Axis Fairness Audit of Transformer Models on Social Media
arXiv 2608.26138 · 2026-08-28可解释、经公平评估且超越单人类天花板的L2口语自动评估模型——以及为何停顿编码不改变LLM流利度评分
Interpretable, Fairly Evaluated Automated L2 Speaking Assessment that Beats the Single-Human Ceiling and Why Pause Encoding Does Not Change LLM Fluency Scores
arXiv 2608.26137 · 2026-08-28奖励感知稀疏自编码器与解决方案完备性混淆
Reward-Informed Sparse Autoencoders and the Solution-Completeness Confound
arXiv 2608.26136 · 2026-08-28用于评估荣誉候选人的数据科学方法
Data Science Approaches to Evaluating Honours Candidates
arXiv 2608.26135 · 2026-08-28精度-效率悖论：量化设备上能量预测中的净能量损失
The Accuracy-Efficiency Paradox Quantifying Net Energy Loss in on-Device Energy Forecasting
arXiv 2608.26134 · 2026-08-28Agent Seer：基于规范理解的场景合成
Agent Seer: Synthesizing Scenarios from Specification Understanding
arXiv 2608.26133 · 2026-08-28面向带属性标签图学习的SLM条件分层关系路由
SLM-Conditioned Hierarchical Relation Routing for Labeled Property Graph Learning
arXiv 2608.26132 · 2026-08-28在真实对话语境中评估语言模型
Evaluating Language Models in Realistic Conversational Contexts
arXiv 2608.26131 · 2026-08-28智能体不进行分页：面向大语言模型工具响应的首块选择
Agents Don't Paginate: First-Chunk Selection for LLM Tool Responses
arXiv 2608.26130 · 2026-08-28FIRSTPASS：基于真实编辑结果的多领域多轮同行评审数据集
FIRSTPASS: A Multi-Domain, Multi-Round Peer Review Dataset Grounded in Real Editorial Outcomes
arXiv 2608.26129 · 2026-08-28基于图的金融波动率动态建模
Graph-Based Modeling of Financial Volatility Dynamics
arXiv 2608.26127 · 2026-08-28TelecomGPT-R1：面向电信栈的统一开源推理器
TelecomGPT-R1: A Unified Open-Source Reasoner for the Telecom Stack
arXiv 2608.26126 · 2026-08-28多语言仇恨言论检测的训练时可解释性：对齐模型推理与人类理由
Training-Time Explainability for Multilingual Hate Speech Detection: Aligning Model Reasoning with Human Rationales
arXiv 2608.26125 · 2026-08-28从自然语言政策到可执行决策：一种可解释的大语言模型框架
Natural-Language Policies to Executable Decisions: An Interpretable Large Language Model Framework
arXiv 2608.26124 · 2026-08-28从电价到利润：面向电池储能系统（BESS）交易的多维概率预测
From electricity prices to profits: multidimensional probabilistic forecasting for BESS trading
arXiv 2608.26122 · 2026-08-28模型能否免费捕捉自身的幻觉？：无标签的怀疑信号在弃权（不执行）任务中可与带标签数据集媲美
Can a Model Catch Its Own Hallucinations for Free?: Label-Free Doubt Signals Hold Their Own Against a Labelled Dataset for Abstention
arXiv 2608.26121 · 2026-08-28通过采样引导和扩展大语言模型的方法
Recipes for Steering and Scaling LLMs via Sampling
arXiv 2608.26120 · 2026-08-28DeflectBench：评估大语言模型中修辞谬误生成的基准
DeflectBench: A Benchmark for Evaluating Rhetorical Fallacy Generation in LLMs
arXiv 2608.26119 · 2026-08-28
