2026-09论文中文摘要
几乎所有代理的公平份额分配
Fair Share Allocations for Almost All Agents
arXiv 2609.33234 · 2026-09-29INSPIRE：面向开放研究问题的科学文献检索基准
Inspire: Benchmarking Scientific Literature Search for Open Research Problems
arXiv 2609.33233 · 2026-09-29MTLiquid：利用液态神经网络实现轻量级医疗监测系统中的高效多任务学习
MTLiquid: Enabling Efficient Multi-Task Learning using Liquid Neural Networks for Lightweight Healthcare Monitoring Systems
arXiv 2609.33232 · 2026-09-29具有相同 $S$、$T$ 和 $W$ 的非等价模张量范畴
Inequivalent modular tensor categories with identical $S$, $T$ and $W$
arXiv 2609.33231 · 2026-09-29AevaScenes：用于远距离感知的FMCW激光雷达数据集与基准
AevaScenes: An FMCW LiDAR Dataset and Benchmark for Long-Range Perception
arXiv 2609.33230 · 2026-09-29面内最优量子比特操作，控制保真度超过99%
In-plane optimal qubit operation with control fidelities exceeding 99 %
arXiv 2609.33229 · 2026-09-29磁通泵浦SQUID谐振器中基于占据态选择的光子对相互作用
Occupation-selective photon-pair interactions in a flux-pumped SQUID resonator
arXiv 2609.33228 · 2026-09-29索引调制辅助相位旋转AFDM用于高效集成传感与通信：框架、优化与性能分析
Index Modulation Aided Phase-Rotated AFDM for Efficient Integrated Sensing and Communication: Framework, Optimization and Performance Analysis
arXiv 2609.33227 · 2026-09-29超越记忆构建：重新思考基于LLM的对话代理的记忆访问
Beyond Memory Construction: Rethinking Memory Access for LLM-based Conversational Agents
arXiv 2609.33226 · 2026-09-29网络拥塞与协调博弈中纳什均衡的复杂性
The Complexity of Nash Equilibrium in Network Congestion and Coordination Games
arXiv 2609.33225 · 2026-09-29PackServe：面向大规模智能体LLM服务的SLO感知请求调度
PackServe: SLO-Aware Request Scheduling for Agentic LLM Serving at Scale
arXiv 2609.33224 · 2026-09-29MaNGA中的棒状星系：恒星质量依赖性与黑洞增长
Barred Galaxies in MaNGA: Stellar Mass Dependence and Black Hole Growth
arXiv 2609.33222 · 2026-09-29RMB：奖励模型提升缓解奖励黑客问题
RMB: Reward Model Boosting Mitigates Reward Hacking
arXiv 2609.33221 · 2026-09-29模型何时承认自己错了？强化学习下失败披露的不稳定性
When Do Models Admit They Are Wrong? Failure Disclosure Is Unstable Under Reinforcement Learning
arXiv 2609.33220 · 2026-09-29带线性约束的张量核范数最小化
Tensor nuclear norm minimization with linear constraints
arXiv 2609.33219 · 2026-09-29Scope-WM：用于高效视觉世界模型的作用域计算
Scope-WM: Scoped Computation for Efficient Visual World Models
arXiv 2609.33218 · 2026-09-29RepFlow：互惠监督提升流模型中的生成与表示
RepFlow: Reciprocal Supervision Improves Generation and Representation in Flow Models
arXiv 2609.33217 · 2026-09-29利用对称边界积分方程加速透射散射中的分层低秩压缩
Hierarchical low-rank compression exploiting symmetric boundary integral equations in transmission scattering
arXiv 2609.33216 · 2026-09-29关于Komlos差异问题的Guo-Fang-Lu算法
On the Guo-Fang-Lu Algorithm for Komlos Discrepancy
arXiv 2609.33215 · 2026-09-29从混合信息来源中学习
Learning from a Mixture of Information Sources
arXiv 2609.33214 · 2026-09-29CHANG-ES XLII：侧向星系中的宇宙射线电子输运与空间分辨的射电-恒星形成率关系
CHANG-ES XLII: Cosmic-Ray Electron Transport and the Spatially Resolved Radio-SFR Relation in Edge-on Galaxies
arXiv 2609.33211 · 2026-09-29背景梯度塑造流匹配中的记忆化
Background Gradients Shape Memorization in Flow Matching
arXiv 2609.33210 · 2026-09-29超越校准：类型化决策模型的概率是否遵循概率公理？
Beyond Calibration: Do a Typed-Decision Model's Probabilities Obey the Probability Axioms?
arXiv 2609.33209 · 2026-09-29WorldAgent: 验证引导的智能体物理世界构建
WorldAgent: Verification-Guided Agentic Physical World Construction
arXiv 2609.33208 · 2026-09-29MorphAtt：一种用于脉冲视觉Transformer中高效多头注意力处理的神经形态加速器
MorphAtt: A Neuromorphic Accelerator for Efficient Multi-Head Attention Processing in Spiking Vision Transformers
arXiv 2609.33207 · 2026-09-29大规模非线性最优控制的自适应并行非精确牛顿方法
An Adaptive, Parallel, and Inexact Newton Method for Large-scale Nonlinear Optimal Control
arXiv 2609.33206 · 2026-09-29SMORE：面向时间相关偏微分方程的稳定性提升的网格无关模型降阶方法
SMORE: Stability-Promoting Mesh-Agnostic Model Reduction for Time-Dependent PDEs
arXiv 2609.33205 · 2026-09-29将语音语言模型转变为多语言听众
Turning Speech Language Models into Multilingual Listeners
arXiv 2609.33204 · 2026-09-29结构化残差连接对扩散变换器的重要性
Structured Residual Connectivity Matters for Diffusion Transformers
arXiv 2609.33203 · 2026-09-29ReAL：通过共享前瞻的分段推进加速流匹配
ReAL: Accelerating Flow Matching through Segment Advancement with Shared Lookahead
arXiv 2609.33202 · 2026-09-29一类Hermite型平流方程离散化的稳定性壁垒
The stability barrier of a class of Hermite-type discretizations of advection equations
arXiv 2609.33201 · 2026-09-29双黑洞参数推断中自旋先验假设的检验及其天体物理意义
Testing Spin Prior Assumptions of Binary Black Hole Parameter Inference and Their Astrophysical Implications
arXiv 2609.33199 · 2026-09-29利用XENONnT仅电离数据探测太阳变色龙粒子
Probing Solar Chameleons with XENONnT Ionization-Only Data
arXiv 2609.33198 · 2026-09-29TAO-DA：迈向自主操作——一种用于协调操作的双臂视觉-语言-动作模型
TAO-DA: Towards Autonomous Operation--A Dual-Arm Vision-Language-Action Model for Coordinated Manipulation
arXiv 2609.33197 · 2026-09-29基准测试可靠吗？通过样本级能力边界进行结构诊断
Are Benchmarks Reliable? Toward Structural Diagnosis via Sample-Level Capability Boundaries
arXiv 2609.33196 · 2026-09-29ReVR：多模态虚假新闻检测的双路径概念推理
ReVR: Dual-Path Concept Reasoning for Multimodal Fake News Detection
arXiv 2609.33195 · 2026-09-29正交见证控制用于通过Sigmoid谱重塑的Muon优化
Orthogonal Witness Control for Muon Optimization via Sigmoid Spectral Reshaping
arXiv 2609.33194 · 2026-09-29表观压缩，真实稳定性：学习量子波函数的固有维度
Apparent Compression, Real Stability: The Intrinsic Dimension of Learning a Quantum Wavefunction
arXiv 2609.33193 · 2026-09-29AG-CoT：Clifford电路上LLM程序合成的验证算法轨迹
AG-CoT: Verified Algorithmic Traces for LLM Program Synthesis on Clifford Circuits
arXiv 2609.33192 · 2026-09-29五列递归线Zarankiewicz数的十二行种子与单行扩展
A Twelve-Row Seed and a One-Row Extension for Five-Column Recursive-Line Zarankiewicz Numbers
arXiv 2609.33191 · 2026-09-29FocusDrive：基于视觉焦点的自动驾驶推理
FocusDrive: Reasoning with Visual Focus for Autonomous Driving
arXiv 2609.33190 · 2026-09-29当评估规则编写训练标签：测量NAVSIM中人类参考宽容度
When an Evaluation Rule Writes Training Labels: Measuring Human-Reference Forgiveness in NAVSIM
arXiv 2609.33189 · 2026-09-29量子场纠缠中的纳什均衡
Nash Equilibrium from Quantum-Field Entanglement
arXiv 2609.33188 · 2026-09-29一个乘积不等式及其在交叉相交族中的应用
A product inequality and its application to cross-intersecting families
arXiv 2609.33187 · 2026-09-29基于混合贝尔曼残差与自适应评论家表示的离线策略评估
Offline Policy Evaluation via Mixed Bellman Residuals and Adaptive Critic Representations
arXiv 2609.33186 · 2026-09-29奇特征下零 $p$-秩曲线的自同构大群
Large automorphism groups of curves of zero $p$-rank in odd characteristic
arXiv 2609.33185 · 2026-09-29面向长上下文LLM服务的高资源效率投机解码
Resource-Efficient Speculative Decoding for Long-Context LLM Serving
arXiv 2609.33184 · 2026-09-29识别Transcoders中用于时间敏感事实回忆的时间特征
Identifying Temporal Features within Transcoders for Time Sensitive Factual Recall
arXiv 2609.33183 · 2026-09-29解锁大语言模型中的潜在个性化能力
Unlocking Latent Personalization in LLMs
arXiv 2609.33182 · 2026-09-29评分错误的问题：受限选项评估中的读取失败
Scoring the Wrong Question: Readout Failures in Constrained-Option Evaluation
arXiv 2609.33179 · 2026-09-29
