2026-10论文中文摘要
广义残差闭包：稳定性-可塑性兼容的通用学习动力学
Generalized Residual Closure: General Learning Dynamics for Stability-Plasticity Compatibility
arXiv 2609.38911 · 2026-10-01无挫折可停当局部哈密顿量问题在亚常数间隙下属于NP
Frustration Free Stoquastic Local Hamiltonian with Sub-Constant Gap is in NP
arXiv 2609.38910 · 2026-10-01使用对比遗忘集消除大型语言模型中的欺骗行为
Unlearning Deceptive Behaviors in LLMs with Contrastive Forget Sets
arXiv 2609.38909 · 2026-10-01CellMSA：用于单细胞表示学习的上下文建模
CellMSA: Context Modeling for Single-Cell Representation Learning
arXiv 2609.38908 · 2026-10-01通过实时课堂互动的上下文分析刻画提问模式与学生参与度
Characterizing Questioning Patterns and Student Engagement Through Contextual Analysis of Real-Time Classroom Interactions
arXiv 2609.38907 · 2026-10-01关于Calabi-Yau流形的双有理自同构的代数动力学与泛Coxeter群
On algebraic dynamics of birational automorphisms of Calabi-Yau manifolds and the universal Coxeter groups
arXiv 2609.38906 · 2026-10-01高斯随机变量的配分函数
On partition functions of Gaussian random variables
arXiv 2609.38904 · 2026-10-01DAMPER：面向平滑策略的回报优先梯度控制
DAMPER: Return-Prioritized Gradient Control for Smooth Policies
arXiv 2609.38903 · 2026-10-01摊销式数据借用与可交换性感知的神经后验估计
Amortized Data Borrowing with Exchangeability-Aware Neural Posterior Estimation
arXiv 2609.38902 · 2026-10-01可解释表征地标的认证近似
Certified Approximation for Interpretable Representer Landmarks
arXiv 2609.38901 · 2026-10-01MEMO：面向流式视频理解的多层级实体感知记忆
MEMO: Multi-Level Entity-Aware Memory for Streaming Video Understanding
arXiv 2609.38900 · 2026-10-01SceneJail：利用视频场景上下文越狱多模态大语言模型
SceneJail: Exploiting Video Scenario Context to Jailbreak Multimodal LLMs
arXiv 2609.38899 · 2026-10-01K2P：无标签知识到提示蒸馏
K2P: Label-Free Knowledge to Prompt Distillation
arXiv 2609.38898 · 2026-10-01FFASR：使用高保真模拟RIR的远场自动语音识别基准测试
FFASR: Benchmarking Far-Field Automatic Speech Recognition using High-Fidelity Simulated RIRs
arXiv 2609.38897 · 2026-10-01关于解析挠率渐近展开的系数
On the coefficients of the asymptotic of analytic torsion
arXiv 2609.38896 · 2026-10-01Unmerge：通过任务算术实现高效机器遗忘
Unmerge: Efficient Machine Unlearning via Task Arithmetic
arXiv 2609.38895 · 2026-10-01量子随机预言机模型中的量子时间锁谜题
Quantum Time-Lock Puzzles in the Quantum Random Oracle Model
arXiv 2609.38894 · 2026-10-01学习随机混合系统的连续神经表示
Learning Continuous Neural Representation of Stochastic Hybrid Systems
arXiv 2609.38893 · 2026-10-01反铁磁体中负g因子掺杂剂的光学研究：Nd:GdVO$_4$
Optical studies of negative g-factor dopants in an antiferromagnet: Nd:GdVO$_4$
arXiv 2609.38892 · 2026-10-01长时程智能体任务中的一致规划-执行
Consistent Plan-Act for Long-Horizon Agentic Tasks
arXiv 2609.38891 · 2026-10-01PRICE 动作块：具身强化学习的物理关系信用分配
PRICE the Action Chunks: Physical Relational Credit Assignment for Embodied Reinforcement Learning
arXiv 2609.38890 · 2026-10-01VERA：面向约束多智能体控制的可验证可行性表示与反事实信用
VERA: Verifiable Feasibility Representations with Counterfactual Credit for Constrained Multi-Agent Control
arXiv 2609.38889 · 2026-10-01随机 Landau--Lifshitz--Baryakhtar 方程在 $\mathbb{R}^d$ 中的适定性
Well-posedness of the Stochastic Landau--Lifshitz--Baryakhtar Equation in $\mathbb{R}^d$
arXiv 2609.38888 · 2026-10-01部分可观测机器人操作中技能级记忆的基准测试与增强
Benchmarking and Enhancing Skill-Level Memory for Partially Observable Robotic Manipulation
arXiv 2609.38886 · 2026-10-01用更少的令牌做更多的事：面向高效编码智能体的分层强化学习
Doing More with Less Tokens: Hierarchical Reinforcement Learning for Efficient Coding Agents
arXiv 2609.38885 · 2026-10-01正确答案，昂贵模型：基于LLM的优化建模中的效率差距
Right Answers, Costly Models: The Efficiency Gap in LLM-based Optimization Modeling
arXiv 2609.38884 · 2026-10-01迈向量子软件自动化：用于LLM引导进化的量子感知工具
Toward Quantum Software Automation: A Quantum-Aware Harness for LLM-Guided Evolution
arXiv 2609.38883 · 2026-10-01高斯玻色采样中函数计算的经典算法
Classical Algorithms for Function Computation in Gaussian Boson Sampling
arXiv 2609.38882 · 2026-10-01STRATA：面向实时策略游戏的角色对齐分层智能体的自学习
STRATA: Self-Learning Through Role-Aligned Tiered Agents for Real-Time Strategy Games
arXiv 2609.38881 · 2026-10-01异步TD学习在马尔可夫数据下的尖锐统计速率
Sharp Statistical Rates for Asynchronous TD Learning with Markovian Data
arXiv 2609.38880 · 2026-10-01学习蛋白质折叠能否泛化到更广泛的推理？
Does Learning Protein Folding Generalize to Broader Reasoning?
arXiv 2609.38879 · 2026-10-01音频令牌注意力在语言模型运行前即可预测
Audio Token Attention Is Predictable Before the Language Model Runs
arXiv 2609.38878 · 2026-10-01从视频中估计非线性标量动力学参数：不变量、标定与可辨识性
On Parameters of Nonlinear Scalar Dynamics from Video: Invariants, Calibration, and Identifiability
arXiv 2609.38877 · 2026-10-01Kromatic对称函数展开的新解释
New interpretations for Kromatic symmetric function expansions
arXiv 2609.38876 · 2026-10-01最优分配与曲面下体积
Optimal Allocation and Volume under Surface
arXiv 2609.38875 · 2026-10-01具有非线性发生率的SIRS流行病模型中产生5个极限环的余维5 Hopf分岔
A codimension-5 Hopf bifurcation yielding 5 limit cycles in a SIRS epidemic model with nonlinear incidence rate
arXiv 2609.38874 · 2026-10-01SoK：基于仿真的ARM Cortex-M固件动态分析研究的大规模实证研究（扩展版）
SoK: A Large-Scale Empirical Study of Emulation-Based Dynamic Analysis Research for ARM Cortex-M Firmware (Extended Version)
arXiv 2609.38872 · 2026-10-01面向单元中心近不可压缩弹性的核感知双网格预条件子
A Kernel-Aware Two-Grid Preconditioner for Cell-Centered Nearly Incompressible Elasticity
arXiv 2609.38871 · 2026-10-01等离激元异质二聚体的表面量子响应：揭示隐藏的非局域性
Surface quantum response of plasmonic heterodimers: revealing the hidden nonlocality
arXiv 2609.38870 · 2026-10-01仅利用接收信号功率测量的盲智能反射面波束成形
Blind IRS Beamforming Using Only Received Signal Power Measurements
arXiv 2609.38868 · 2026-10-01Talk2Agent：文本代理的语音接口基准评测
Talk2Agent: Benchmarking Voice Interfaces for Text Agents
arXiv 2609.38867 · 2026-10-01当上下文变化时：理解大语言模型中的更新失败
When Context Changes: Understanding Update Failures in LLMs
arXiv 2609.38866 · 2026-10-01关于 Nathanson 的一个与 $h$ 阶强极小渐近基相关的问题
On a problem of Nathanson related to strongly minimal asymptotic bases of order $h$
arXiv 2609.38865 · 2026-10-01AdaOcc：面向具身任务的自适应三维占据预测
AdaOcc: Adaptive 3D Occupancy Prediction for Embodied Tasks
arXiv 2609.38864 · 2026-10-01GeoNest：学习为圆形容器中的不规则背包问题选择故障感知邻域
GeoNest: Learning to Select Failure-Aware Neighborhoods for the Irregular Knapsack Problem in a Circular Container
arXiv 2609.38863 · 2026-10-01高效多模态规划：面向自动驾驶的奖励引导偏好优化
Efficient Multi-Modal Planning with Reward-Guided Preference Optimization for Autonomous Driving
arXiv 2609.38862 · 2026-10-01TRACE：面向目标感知的检索、归因证据与契约约束抽取，用于LitTraceQA
TRACE: Target-Aware Retrieval, Attributed Evidence, and Contract-Constrained Extraction for LitTraceQA
arXiv 2609.38861 · 2026-10-01有偏LLM评判者下主动偏好学习的最优设计
Optimal Design for Active Preference Learning with Biased LLM Judges
arXiv 2609.38860 · 2026-10-01面向智能反射面辅助稳健无线通信的盲干扰抑制
Blind Interference Suppression for IRS-Aided Robust Wireless Communications
arXiv 2609.38859 · 2026-10-01MATCH：武汉大学1米光学望远镜的性能表征与科学观测
MATCH: Performance Characterization and Scientific Observations of the Wuhan University 1-m Optical Telescope
arXiv 2609.38858 · 2026-10-01
