2026-09论文中文摘要
LoRo-Mark：可证明无损且鲁棒的智能体水印
LoRo-Mark:Provably Lossless and Robust Agent Watermarking
arXiv 2609.34080 · 2026-09-29GenoMorph：基于通路锚定的基因组疾病推理与自适应潜在计算
GenoMorph: Pathway-Grounded Genomic Disease Reasoning via Adaptive Latent Computation
arXiv 2609.34079 · 2026-09-29WhiteCon：通过白化变换和双重一致性实现半监督域适应回归
WhiteCon: Semi-Supervised Domain Adaptation Regression Through Whitening Transform and Dual Consistency
arXiv 2609.34078 · 2026-09-29基于观测器的模型预测控制在EXL-50U球形托卡马克等磁通调节中的应用
Observer-Based Model Predictive Control for Isoflux Regulation in the EXL-50U Spherical Tokamak
arXiv 2609.34076 · 2026-09-29信息的双刃剑：学校选择中揭示与隐藏的抽签
The Double-Edged Sword of Information: Revealed versus Hidden Lotteries in School Choice
arXiv 2609.34074 · 2026-09-29dS/CFT 全息包括费米子场
dS/CFT Holography including Fermionic Fields
arXiv 2609.34073 · 2026-09-29PhysFieldBench：多模态模型能否理解物理场？
PhysFieldBench: Can Multimodal Models Understand Physical Fields?
arXiv 2609.34072 · 2026-09-29SNaP：针对含噪逆问题的一步后验采样
SNaP: One-Step Posterior Sampling for Noisy Inverse Problems
arXiv 2609.34071 · 2026-09-29FLARE：基于局部轴角表示的流匹配方法用于随机微磁演化
FLARE: Flow Matching with Local Axis-Angle Representations for Stochastic Micromagnetic Evolution
arXiv 2609.34070 · 2026-09-29面向证书驱动的软件移植：一种用于科学程序优化的自改进智能体框架
Towards Certificate-Driven Software Porting: A Self-Improving Agentic Harness for Scientific Program Optimization
arXiv 2609.34069 · 2026-09-29树的乘积与交换部分映射的Borel复杂度
Borel complexity for product of trees and commuting partial maps
arXiv 2609.34068 · 2026-09-29关于DAG可交换性与广义圈积的注记
Remarks on DAG exchangeability and generalized wreath products
arXiv 2609.34067 · 2026-09-29具有粗糙噪声的随机波动方程的双参数局部线性化
Bi-parameter local linearization for the stochastic wave equation with rough noise
arXiv 2609.34066 · 2026-09-29谁获得 Token，它携带什么？大型语言模型中不平等的名称支持与概念访问
Who Gets a Token, and What Does It Carry? Unequal Name Support and Concept Access in Large Language Models
arXiv 2609.34065 · 2026-09-29学习具有稳定优化的LLM智能体的扰动鲁棒策略
Learning Perturbation Robust Policies for LLM Agents with Stable Optimization
arXiv 2609.34064 · 2026-09-29局部重建增益作为残差补全中最终保真度代理的反例
Counterexamples to Local Reconstruction Gain as a Proxy for Final Fidelity in Residual Completion
arXiv 2609.34063 · 2026-09-29Scanvas：在生成式设计空间中探索与开发协同机会
Scanvas: Discovering and Developing Synergistic Opportunities in Generative Design Spaces
arXiv 2609.34062 · 2026-09-29视觉-语言-动作模型的分位数头
Quantile Head for Vision-Language-Action Models
arXiv 2609.34061 · 2026-09-29极化码的并行连续消除扰动增强解码：基于离线方差设计
Parallel Successive Cancellation Perturbation-Enhanced Decoding of Polar Codes via Offline Variance Design
arXiv 2609.34059 · 2026-09-29世界模型能学习全局理解吗？
Do World Models Learn Global Understanding?
arXiv 2609.34058 · 2026-09-29关于非阿贝尔张量平方 $G \otimes G$、非阿贝尔外平方 $G \wedge G$ 与弱交换构造 $\X(G)$ 的剩余有限性
On the residual finiteness of the non-abelian tensor square $G \otimes G$, the non-abelian exterior square $G \wedge G$ and the weak commutativity construction $\X(G)$
arXiv 2609.34057 · 2026-09-29通过价值移植引导语言模型目标
Steering Language Model Goals with Value Transplant
arXiv 2609.34056 · 2026-09-29Skala机器学习交换相关泛函在CP2K中的原生实现：用于分子和凝聚相计算的统一单中心重构
Native implementation of the machine-learned Skala exchange-correlation functional in CP2K: Unified one-centre reconstruction for molecular and condensed-phase calculations
arXiv 2609.34055 · 2026-09-29图式 Hecke 范畴中的显式 $H_3$ 关系
Explicit $H_3$ relation in the diagrammatic Hecke category
arXiv 2609.34053 · 2026-09-29面向神经音频编解码器的基于稀疏自编码器的可解释框架：面向年龄、性别和口音引导的探索
Towards Interpretable Framework for Neural Audio Codecs via Sparse Autoencoders: Exploration toward Age, Gender, and Accent Steering
arXiv 2609.34052 · 2026-09-29聚合物信息学超级智能路线图
A roadmap for polymer informatics super-intelligence
arXiv 2609.34051 · 2026-09-29带反馈的因果发现中的统计代价
The Statistical Cost of Causal Discovery with Feedback
arXiv 2609.34050 · 2026-09-29跳出框框思考：滑动窗口KV推理中信息的保留与传递
Thinking Outside the Box: Retention and Transmission of Information in Sliding-Window KV Inference
arXiv 2609.34049 · 2026-09-29链路中断下DAG工作流的调度修复
Schedule Repair for DAG Workflows under Link Disruptions
arXiv 2609.34048 · 2026-09-29ARCH-B：建筑表征、理解与层级基准
ARCH-B: Architectural Representation, Comprehension and Hierarchy Benchmark
arXiv 2609.34047 · 2026-09-29StructSim：通过结构表示大规模度量想法相似性
StructSim: Measuring Idea Similarity at Scale Through Structural Representation
arXiv 2609.34046 · 2026-09-29Kafila：在可信的异构商用机器集合上服务大型语言模型
Kafila: Serving Large Language Models on a Trusted Set of Heterogeneous Commodity Machines
arXiv 2609.34045 · 2026-09-29SCOPD：面向高效视觉语言模型的稀疏上下文在线策略自蒸馏
SCOPD: Sparse-Context On-Policy Self-Distillation for Efficient Vision-Language Models
arXiv 2609.34044 · 2026-09-29非负矩阵分解的奇异性及其在贝叶斯推断中的应用
Singularities of Non-negative Matrix Factorization and their application to Bayesian inference
arXiv 2609.34043 · 2026-09-29Xiao 亏格二纤维化在次数 3、4 和 5 下的单值性
The monodromy of Xiao's genus-two fibrations in degrees 3, 4 and 5
arXiv 2609.34042 · 2026-09-29约束强化学习中的视觉-语言信号：无需预判的安全增益
Vision--Language Signals in Constrained RL: Safety Gains Without Anticipation
arXiv 2609.34041 · 2026-09-29横模耦合光机谐振器中的多模振荡及其通往混沌的路径
Multimode oscillation and its route to chaos in transverse-mode-coupled optomechanical resonators
arXiv 2609.34040 · 2026-09-29大型语言模型用于结构化临床数据分析：双智能体接地与验证
Large Language Models for Structured Clinical Data Analysis: Dual-Agent Grounding and Validation
arXiv 2609.34039 · 2026-09-29锥形光纤法布里-珀罗腔用于产生近可见、相位锁定和宽带Kerr频率梳
Tapered Fibre Fabry-Pérot Cavities for the Generation of Near-Visible, Phase-locked, and Broadband Kerr Frequency Combs
arXiv 2609.34037 · 2026-09-29UOPD：多轮智能体在线策略蒸馏中的不确定性感知干预
UOPD: Uncertainty-Aware Intervention for On-Policy Distillation of Multi-Turn Agents
arXiv 2609.34036 · 2026-09-29基于状态空间模型的三维点跟踪
3D Point Tracking with State Space Models
arXiv 2609.34035 · 2026-09-29ADPTNet：用于序列建模的自适应与规定时间尺度的非线性SSM
ADPTNet: Adaptive with Prescriptive Timescales Non-Linear SSM for Sequence Modelling
arXiv 2609.34034 · 2026-09-29忠实激活言语化：减少大语言模型表示解释中的幻觉
Faithful Activation Verbalization: Reducing Hallucinations in LLM Representation Interpretation
arXiv 2609.34033 · 2026-09-29Re:Cognize -- 开放集漫画角色再识别
Re:Cognize -- Open-Set Comic Character Re-Identification
arXiv 2609.34032 · 2026-09-29后验状态与潜在欺骗：隐马尔可夫模型中的变分贝叶斯推断在金融交易序列欺诈检测中的应用
Posterior Regimes and Latent Deception: Variational Bayesian Inference in Hidden Markov Models for Sequential Fraud Detection in Financial Transactions
arXiv 2609.34031 · 2026-09-29揭示音频分类器中的捷径学习：通过发现时间解释中的重复概念
Uncovering shortcut learning in audio classifiers by discovering recurring concepts in temporal explanations
arXiv 2609.34030 · 2026-09-29基于分数的正压准地转湍流随机降阶模型
Score-based stochastic reduced-order models of barotropic quasi-geostrophic turbulence
arXiv 2609.34028 · 2026-09-29PerceptFence：屏幕共享AI助手的基于内容的调解架构与确定性覆盖
PerceptFence: Content-Mediation Architecture and Deterministic Coverage for Screen-Share AI Assistants
arXiv 2609.34027 · 2026-09-29结构自适应树场积分器
Structure-Adaptive Tree Field Integrators
arXiv 2609.34025 · 2026-09-29圆盘图的强边着色：一个6-近似算法与改进的单位圆盘界
Strong Edge Colouring of Disk Graphs: A 6-Approximation and an Improved Unit-Disk Bound
arXiv 2609.34023 · 2026-09-29
