2026-10论文中文摘要
构建智能体/证明器接口：面向Rocq和Lean成本高效定理证明的进化式工具设计
Growing an Agent/Prover Interface: Evolutionary Tool Design for Cost-Efficient Theorem Proving in Rocq and Lean
arXiv 2609.39544 · 2026-10-09有限斜率 p 进 L 函数：$\u200b\u200b\mathrm{GL}_{3}$
Finite-slope p-adic L-functions for $\mathrm{GL}_{3}$
arXiv 2609.39539 · 2026-10-09公平变量选择
Fair Variable Selection
arXiv 2609.39485 · 2026-10-099-扶手椅石墨烯纳米带阵列中的表面依赖声子动力学
Surface-Dependent Phonon Dynamics in 9-Armchair Graphene Nanoribbon Arrays
arXiv 2609.39278 · 2026-10-09超越记忆中的世界：面向演化世界的持久导航的预测性4D信念
Beyond the Remembered World: Predictive 4D Belief for Persistent Navigation in Evolving Worlds
arXiv 2609.39166 · 2026-10-09太赫兹强度测绘仪用动力学电感探测器的千像素性能
Kilopixel Performance of the Kinetic Inductance Detectors for the Terahertz Intensity Mapper
arXiv 2609.38769 · 2026-10-09这里的世界是立体的：学习动态空间对应关系以进行沉浸式联合视频-音频生成
Hear the World in Stereo: Learning Dynamic Spatial Correspondence for Immersive Joint Video-Audio Generation
arXiv 2609.38748 · 2026-10-09约鲁巴语中声调轮廓的标记
Marking Contour Tones in Yorùbá: A Typographic and Computational Proposal
arXiv 2609.38627 · 2026-10-09一种无参数零阶方法：协方差矩阵自适应与有效维度
Parameter-Free Zeroth-Order Optimization with Ellipsoidal Sampling
arXiv 2609.38561 · 2026-10-09Diffusion-2BC：自动驾驶离线行为克隆的混合扩散与回归训练
Diffusion-2BC: Hybrid Diffusion and Regression Training for Offline Behavior Cloning in Autonomous Driving
arXiv 2609.38472 · 2026-10-09自动化条件下保持人工验证能力
Maintaining Human Verification Capacity under Automation
arXiv 2609.38459 · 2026-10-09上下文机器人学习简化：一种面向操作任务的民主化配方
In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks
arXiv 2609.38173 · 2026-10-09目标感知的序贯推断：合并与分层任意有效设计
Target-Aware Sequential Inference: Pooled versus Stratified Anytime-Valid Designs
arXiv 2609.37873 · 2026-10-09CompOrca：指令微调数据的语料级合规标注
CompOrca: Corpus-Scale Compliance Labelling of Instruction-Tuning Data
arXiv 2609.37807 · 2026-10-09每个可数群都嵌入到一个 $\mathrm{FP}_n$ 型群中
Every countable group embeds in a group of type $\mathrm{FP}_n$
arXiv 2609.37562 · 2026-10-09利用二级结构信息实现核酸结构精确预测的OFoldNA
Leveraging secondary-structure information for accurate nucleic acid structure prediction with OFoldNA
arXiv 2609.37414 · 2026-10-09从侦察到响应：现代企业攻击中的定量风险参数化与博弈论遏制
From Reconnaissance to Response: Quantitative Risk Parameterization and Game Theoretic Containment in Modern Enterprise Attack
arXiv 2609.36485 · 2026-10-09基于算法微分的meta广义梯度近似的密度泛函微扰理论
Density functional perturbation theory of meta-generalized gradient approximations using algorithmic differentiation
arXiv 2609.35572 · 2026-10-09基于兴趣点监测的顺风感知网约车市场在不确定出行需求下的鲁棒识别
Robust identification of drive-by sensing ride-hailing market with Points of Interest monitoring under uncertain ride demand
arXiv 2609.35220 · 2026-10-09来自量子引力凝聚体的动力学暗能量
Dynamical dark energy from quantum gravity condensates
arXiv 2609.34917 · 2026-10-09von Neumann 子代数之间的距离：自旋模型、交换平方与自由群因子
Distances Between von Neumann Subalgebras: Spin Models, Commuting Squares, and Free Group Factors
arXiv 2609.34856 · 2026-10-09FlowState：将执行状态作为长时程LLM智能体的记忆
FlowState: Execution State as Memory for Long-Horizon LLM Agents
arXiv 2609.34565 · 2026-10-09Polish 模的 Kaplansky 分解
Kaplansky decompositions of Polish modules
arXiv 2609.34505 · 2026-10-09WorldWeave：为视频生成构建持续增长的持久几何世界
WorldWeave: Growing Persistent Geometric Worlds for Video Generation
arXiv 2609.34221 · 2026-10-09通过潜在空间中的约束优化实现高效推理
Efficient Reasoning via Constrained Optimization in Latent Space
arXiv 2609.34181 · 2026-10-09Opera：面向长时程编码智能体的口头批评框架
Opera: A Verbal Critic Framework for Long-horizon Coding Agents
arXiv 2609.33987 · 2026-10-09重希格斯ino暗物质与低再加热温度：基于LZ高反冲事件
Heavy Higgsino Dark Matter With Low Reheating in View of the LZ High-Recoil Event
arXiv 2609.33869 · 2026-10-09自相互作用超辐射云产生的随机引力波背景
Stochastic gravitational-wave background from self-interacting superradiant clouds
arXiv 2609.33596 · 2026-10-09MA-JEPA：面向多智能体强化学习的联合嵌入世界模型
MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement Learning
arXiv 2609.33563 · 2026-10-09SMAT：简单高效的合并感知训练
SMAT: Simple and Efficient Merge-Aware Training
arXiv 2609.33437 · 2026-10-09何时驱逐，而非保留什么：基于草稿引导的驱逐用于免训练KV缓存压缩
When to Evict, Not What to Keep: Draft-Guided Eviction for Training-Free KV-Cache Compression
arXiv 2609.33334 · 2026-10-09链环的卫星与不变量
Satellites and invariants of links
arXiv 2609.33309 · 2026-10-09具有某些调和符号的超循环Bergman-Toeplitz算子
Hypercyclic Bergman-Toeplitz operators with some harmonic symbols
arXiv 2609.32650 · 2026-10-09FA-Bench：在干净与噪声条件下进行词级和音素级强制对齐及ASR时间戳的基准测试
FA-Bench: A Benchmark for Phone- and Word-Level Timestamp Accuracy in Forced Alignment and ASR on Clean and Noisy Speech
arXiv 2609.32396 · 2026-10-09PlurVA-LLM-2026 共享任务赛道1：通过多语言微调和阈值校准实现大语言模型中的多元价值对齐
PlurVA-LLM-2026 Shared Task Track-1: Pluralistic Value Alignment in LLMs via Multilingual Fine-Tuning and Threshold Calibration
arXiv 2609.32382 · 2026-10-09READ-Bench：面向时间序列诊断的历史实例检索基准
READ-Bench: Benchmarking Historical Instance Retrieval for Time-Series Diagnosis
arXiv 2609.32123 · 2026-10-09使用语言模型建模句子理解过程中语境和指代的影响
Using LMs to Model the Effects of Context and Coreference during Sentence Comprehension
arXiv 2609.32119 · 2026-10-09CeGaGe中通过增强关联实现的手性Weyl-Kondo半金属态
Chiral Weyl-Kondo semimetallic state through enhanced correlation in CeGaGe
arXiv 2609.32012 · 2026-10-09侧耦合光子晶体微腔中的离散品质因子调控：倏逝Bloch隧穿与有限元胞修正
Discrete quality-factor control in a side-coupled photonic crystal microcavity: evanescent Bloch tunnelling and the finite-cell correction
arXiv 2609.31952 · 2026-10-09并行与分布式费米子模拟：基于动态编码的方法
Parallel and Distributed Fermionic Simulation via Dynamic Encoding
arXiv 2609.31502 · 2026-10-09完成的配对掩盖了封顶失败：选择性上下文投影的 ReVerPi 案例研究
Evaluating Budgeted Context Projection with Unexecuted Companion Runs
arXiv 2609.31381 · 2026-10-09广义调和测度：同步逼近与正调和表示
Generalized Harmonic Measures: Synchronized Approximation, Representation and Quantitative Measure Recovery
arXiv 2609.31279 · 2026-10-09AdS/CFT 平坦极限中的希尔伯特空间 I. 有质量粒子
Hilbert space in the flat limit of AdS/CFT. I. Massive particles
arXiv 2609.31055 · 2026-10-09CCRV-Bench：基于约束的视觉语言模型因果推理评估
CCRV-Bench: Constraint-Based Evaluation of Causal Reasoning in Vision-Language Models
arXiv 2609.30979 · 2026-10-09RecToolBench：模糊用户意图下的推荐专用工具编排基准
RecToolBench: Benchmarking Recommendation-Specific Tool Orchestration under Fuzzy User Intent
arXiv 2609.30717 · 2026-10-09期望阈值与分数期望阈值之间的对数星比较
A Log-Star Comparison Between Expectation Threshold and Fractional Expectation Threshold
arXiv 2609.30680 · 2026-10-09木板覆盖的面积稳定性
Volume stability of plank covers
arXiv 2609.30409 · 2026-10-09SlideLab：以受众为中心的科技幻灯片生成与评估
SlideLab: Audience-Centered Scientific Slide Generation and Evaluation
arXiv 2609.30294 · 2026-10-09JevOut：自然上下文可使决策模型翻转
JevOut: Natural Context Can Flip Decision Models
arXiv 2609.30243 · 2026-10-09ExplorationBench：在可验证的外星世界中衡量AI系统的探索能力
ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds
arXiv 2609.30199 · 2026-10-09
