2026-09论文中文摘要
语言模型中的深层与浅层偏见
Deep and shallow biases in language models
arXiv 2609.09901 · 2026-09-10关于旋转双分量$b$-族浅水波系统：推导与波浪破碎分析
On the rotation-two-component $b$-family shallow-water system: derivation and wave-breaking analysis
arXiv 2609.09900 · 2026-09-10对自身陌生：语言模型关于自身的表述是泛化的
Strangers to Themselves: What Language Models Say About Themselves Is Generic
arXiv 2609.09899 · 2026-09-10面向自然语言到PDDL问题生成的接地评估与修复
Grounded Evaluation and Repair for NL-to-PDDL Problem Generation
arXiv 2609.09898 · 2026-09-10C$^{18}$O发射示踪的MMS 2/OMC-3周围致密气体结构
The Dense Gas Structures Around MMS 2/OMC-3 Traced by C$^{18}$O Emission
arXiv 2609.09897 · 2026-09-10重心细分和匹配细分的零因子图的度量基
Metric Bases of Barycentric and Matching Subdivisions of Zero-Divisor Graphs
arXiv 2609.09896 · 2026-09-10我们能信任视频幻觉检测器吗？VidHalLoc用于评估评估者
Can We Trust Video Hallucination Detectors? VidHalLoc for Evaluating the Evaluators
arXiv 2609.09895 · 2026-09-10不定权 Minkowski 平均曲率 Neumann 问题的正解：多重性与渐近行为
Positive Solutions for an Indefinite-Weight Minkowski Mean Curvature Neumann Problem: Multiplicity and Asymptotic Behaviour
arXiv 2609.09894 · 2026-09-10JWST-SUPER I：来自MIRI对HD 106315 c观测的受辐照暖海王星大气新见解
JWST-SUPER I: New insights into irradiated warm Neptunes atmospheres from MIRI observations of HD 106315 c
arXiv 2609.09893 · 2026-09-10离散与连续时间感染过程的比较不等式
Comparison inequalities for discrete- and continuous-time infection processes
arXiv 2609.09892 · 2026-09-10ProMeta：跨E3连接酶的少样本PROTAC靶向降解预测
ProMeta: Few-shot PROTAC-targeted degradation prediction across E3 ligases
arXiv 2609.09891 · 2026-09-10StreetDiff：基于结构提示的跨视角一致多视角稳定扩散街道场景生成
StreetDiff: Multi-view Street Scenes Generation via Cross-view Consistent Multi-view Stable Diffusion with Structure Prompts
arXiv 2609.09890 · 2026-09-10利用韩语语音识别中的细粒度错误纠正提升咨询服务
Leveraging Fine-grained Error Correction in Korean Speech Recognition for Consultation Services
arXiv 2609.09889 · 2026-09-10带借贷约束的消费对利率的响应：一种分析方法
The Response of Consumption to Interest Rates with Borrowing Constraints: An Analytical Approach
arXiv 2609.09888 · 2026-09-10被告陈述何时重要？LLM模拟陪审员中的偏见与说服研究
When Does Defendant Statement Matter? A Study of Bias and Persuasion in LLM-Simulated Jurors
arXiv 2609.09887 · 2026-09-10首次在690GHz实现甚长基线干涉测量条纹检测
First Very Long Baseline Interferometry Fringe Detection at 690GHz
arXiv 2609.09886 · 2026-09-10用于无人机搭载可重构智能表面辅助动态D2D通信的决策Transformer
Decision Transformer for UAV-Mounted RIS-Assisted Dynamic D2D Communications
arXiv 2609.09885 · 2026-09-10基于潜桥匹配的反照率估计
Albedo Estimation via Latent Bridge Matching
arXiv 2609.09884 · 2026-09-10行为语言模型中的评分式与生成式读出：一种引出格式的实证研究
Scored vs. Generated Readouts in Behavioral Language Models: An Empirical Study of Elicitation Format
arXiv 2609.09882 · 2026-09-10CLFTv2：基于层次化特征金字塔的高效相机-激光雷达语义分割融合
CLFTv2: Efficient Camera-LiDAR Fusion for Semantic Segmentation via Hierarchical Feature Pyramids
arXiv 2609.09881 · 2026-09-10极端贫金属红巨星BD$-18^{\circ}$5550的恒星参数与表面核合成的理论研究——第一部分：$\alpha$元素增强与内部混合的影响
A Theoretical Study of the Stellar Parameters and Surface Nucleosynthesis of the Extremely Metal-Poor RGB Giant BD$-18^{\circ}$5550 - Part I: Effects of $α$-enhancement and Internal Mixing
arXiv 2609.09880 · 2026-09-10真实AI对话中的购买建议与可观察的买家反应
Purchase Advice and Observable Buyer Responses in Real AI Conversations
arXiv 2609.09878 · 2026-09-10两个后公共包层双星系统的研究
Studies of Two Post Common Envelope Binary Stars
arXiv 2609.09877 · 2026-09-10从像素到层次序列：用于视觉-语言二值变化检测的四叉树掩码编码
From Pixels to Hierarchical Sequences: Quadtree Mask Encoding for Vision-Language Binary Change Detection
arXiv 2609.09876 · 2026-09-10AgentAudit：一个用于AI智能体全生命周期可信度评估的开放、可扩展框架
AgentAudit: An Open, Extensible Framework for Full-Lifecycle Trust Evaluation of AI Agents
arXiv 2609.09875 · 2026-09-10量化体内上皮组织中细胞形状和密度的波动
Quantifying cell shape and density fluctuations in epithelial tissue in vivo
arXiv 2609.09874 · 2026-09-10诱导与非诱导覆盖数之间的有界性与分离性
Boundedness and Separation Between Induced and Non-Induced Covering Numbers
arXiv 2609.09873 · 2026-09-10倾斜绝对值函数的初始化依赖BFGS试验率
Initialization-dependent BFGS trial rates for tilted absolute values
arXiv 2609.09872 · 2026-09-10étale Chow 群的 Oka 原理
The Oka principle for étale Chow groups
arXiv 2609.09871 · 2026-09-10相变材料中嵌入的通信波段量子点纳米线的光致波长调谐
Photo-induced Wavelength-tuning of Telecom-band Quantum Dot Nanowires Embedded in a Phase Change Material
arXiv 2609.09870 · 2026-09-10具有int-amplified自同态的紧致Kähler簇的正性
Positivity of compact Kähler varieties admitting an int-amplified endomorphism
arXiv 2609.09869 · 2026-09-10氢能共址可再生能源海水淡化厂的最优调度
Optimal Dispatch of a Hydrogen-Colocated Renewable-Powered Desalination Plant
arXiv 2609.09868 · 2026-09-10极端低位LLM中量化移动的上下文效用
Contextual Utility of Quantization Moves in Extreme Low-Bit LLMs
arXiv 2609.09867 · 2026-09-10UniStream：面向48 kHz因果流式音频编码的多专家残差向量量化
UniStream: Multi-Expert Residual Vector Quantization for 48 kHz Causal Streaming Audio Coding
arXiv 2609.09866 · 2026-09-10保持评估公平：通过成员推断攻击检测代码生成基准中的数据泄漏
Keep Evaluation Fair: Detecting Data Leakage in Code Generation Benchmarks via Membership Inference Attacks
arXiv 2609.09865 · 2026-09-10情感计算的关系范式转变：情感共振、活力情感与声音交互场
Shifting Relational Paradigms for Affective Computing: Affective Resonance, Vitality Affects, and Vocal Interaction Fields
arXiv 2609.09864 · 2026-09-10预训练和蒸馏比架构家族对无标签单细胞分类更重要
Pretraining and Distillation Matter More Than Architecture Family for Label-Free Single-Cell Classification
arXiv 2609.09863 · 2026-09-10通用范数与de Rham表示
Universal norms and de Rham representations
arXiv 2609.09862 · 2026-09-10平衡k-shot采样下的精确简并：对LLM嵌入小样本判别分析的影响
Exact Degeneracy Under Balanced k-Shot Sampling:Consequences for Small-Sample Discriminant Analysis on LLM Embeddings
arXiv 2609.09860 · 2026-09-10经济如何在空间中收缩：人口下降下的就业凹性与向上整合
How an Economy Shrinks in Space: Concavity-on-Jobs and Upward Consolidation under Demographic Decline
arXiv 2609.09859 · 2026-09-10MPFA：黎曼双目标优化的帕累托前沿逼近方法
MPFA: A Pareto Front Approximation Method for Riemannian Bi-objective Optimization
arXiv 2609.09857 · 2026-09-10有了 Thermomix，你就失去了烹饪能力：生成式人工智能在教育中应用的厨房机器类比
With a Thermomix You Lose the Ability to Cook: A Kitchen Machine Analogy for Applications of Generative AI in Education
arXiv 2609.09856 · 2026-09-10概率作为模型预测的统一视角
A Unifying Perspective on Probabilities as Model Predictions
arXiv 2609.09855 · 2026-09-10低比特量化何时能保持向量搜索的决策？
When Does Low-Bit Quantization Preserve the Decisions of Vector Search?
arXiv 2609.09854 · 2026-09-10Eon基准的Era：一个具有精确真实标签的生成式企业资产，用于基准测试LLM智能体
The Era by Eon Benchmark: A Generated Enterprise Estate with Exact Ground Truth for Benchmarking LLM Agents
arXiv 2609.09853 · 2026-09-10$S^3$-Bench：评估作为科学语音助手的语音交互模型
$S^3$-Bench: Evaluating Speech Interaction Models as Scientific Voice Assistants
arXiv 2609.09852 · 2026-09-10属于所有最小标识码的顶点
On the Vertices That Belong to All Minimum Identifying Codes
arXiv 2609.09851 · 2026-09-10量子加速何时能在端到端中幸存？
When Does a Quantum Speedup Survive End-to-End?
arXiv 2609.09850 · 2026-09-10AI代理能否检测并修复网络实验中的工件漂移？
Can AI Agents Detect and Repair Artifact Drift in Network Experiments?
arXiv 2609.09849 · 2026-09-10差分隐私合成文本的子组隶属推断审计
Subgroup Membership Inference Audits of Differentially Private Synthetic Text
arXiv 2609.09848 · 2026-09-10
