2026-09论文中文摘要
MaLiang-Harness：通往图像与视频生成的可编程路径
MaLiang-Harness: A Programmable Path to Image and Video Generation
arXiv 2609.34309 · 2026-09-29格图上后向平移的$\mathcal F$-传递性
$\mathcal F$-Transitivity of Backward Shifts on Lattice Graphs
arXiv 2609.34308 · 2026-09-29关于 $\mathrm{GL}_n(\mathrm{D})$ 上具有广义线性周期的表示
On Representations of $\mathrm{GL}_n(\mathrm{D})$ admitting a generalized linear period
arXiv 2609.34307 · 2026-09-29SPRINT: 通过平均概率速度实现单步生成式推荐
SPRINT: Single-Step Generative Recommendation via Average Probability Velocity
arXiv 2609.34306 · 2026-09-29利用去卷积改进地面成像的透镜建模
Improving Lens Modelling from Ground-Based Imaging with Deconvolution
arXiv 2609.34305 · 2026-09-29EXO-200 面向AI/ML应用的公共数据发布
EXO-200 Public Data Release for AI/ML Applications
arXiv 2609.34304 · 2026-09-29连续优化的渐近下界
Asymptotic Lower Bounds for Continuous Optimization
arXiv 2609.34303 · 2026-09-29一个序列，多种解码：CAGenMol-2 将药物设计重构为掩码分子推断
One Sequence, Many Decodings: CAGenMol-2 Recasts Drug Design as Masked Molecular Inference
arXiv 2609.34301 · 2026-09-29当世界模型说谎：错误想象下的自适应安全分析
When World Models Lie: Adaptive Safety Analysis Under Wrong Imaginations
arXiv 2609.34300 · 2026-09-29PSM：基于难度精确统计匹配的数据集蒸馏
PSM: Dataset Distillation Based on Precise Statistical Matching by Difficulty
arXiv 2609.34299 · 2026-09-29核散射相移：量子处理器上的因子化几何时间 RODEO
Nuclear scattering phase shifts with factorized geometric-time RODEO on a quantum processor
arXiv 2609.34298 · 2026-09-29TLC-DiT：任务对齐的局部视觉条件用于鲁棒的多任务机器人操作
TLC-DiT: Task-Aligned Local Visual Conditioning for Robust Multitask Robot Manipulation
arXiv 2609.34297 · 2026-09-29Dr.Credit：面向深度研究智能体的基于评分标准的流程信用分配
Dr.Credit: Rubric-Grounded Process Credit Assignment for Deep Research Agents
arXiv 2609.34296 · 2026-09-29单调Sobolev函数：逼近、临界点与水平集
Monotone Sobolev functions: approximation, critical points, and level sets
arXiv 2609.34295 · 2026-09-29语义模态补偿用于非配对设置下的无监督可见光-红外行人重识别
Semantic Modality Compensation for Unsupervised Visible-Infrared Person Re-identification under Unpaired Settings
arXiv 2609.34294 · 2026-09-29有序搜索的最优量子算法
Optimal Quantum Algorithms for Ordered Search
arXiv 2609.34293 · 2026-09-29固态物理启发的大语言模型压缩的预注册测试：小语言模型尺度上的聚类级阴性结果
Pre-registered tests of solid-state-physics-inspired LLM compression: a cluster-level negative result at small-language-model scale
arXiv 2609.34292 · 2026-09-29超越多面体：紧凸集上的阻尼牛顿 Frank--Wolfe 方法
Beyond Polytopes: Damped Newton Frank--Wolfe Methods over Compact Convex Sets
arXiv 2609.34290 · 2026-09-29典范预三角商中的更高正交性与截断
Higher orthogonality and truncation in canonical pretriangulated quotients
arXiv 2609.34289 · 2026-09-29无线搜索的非线性预处理梯度下降与半自动几何设计
Linesearch-Free Nonlinearly Preconditioned GD with Semiautomatic Geometry Design
arXiv 2609.34288 · 2026-09-29ReScraper：面向高效LLM预训练的网络数据统一抓取与清洗
ReScraper: Unified Scraping and Cleaning of Web Data for Effective LLM Pretraining
arXiv 2609.34287 · 2026-09-29灵巧触觉世界模型
Dexterous Tactile World Model
arXiv 2609.34286 · 2026-09-29从不精确标注中的认知学习
Epistemic Learning from Imprecise Annotation
arXiv 2609.34285 · 2026-09-29$\mathbb{H}^n$ 中稳定常平均曲率超曲面
Stable Constant Mean Curvature Hypersurfaces in $\mathbb{H}^n$
arXiv 2609.34283 · 2026-09-29基于假设与证据引导搜索的智能体高维贝叶斯优化
Agentic High-Dimensional Bayesian Optimization with Hypothesis- and Evidence-Guided Search
arXiv 2609.34281 · 2026-09-29MoSPR：利用形态-空间宏状态和低秩分子程序进行组织学到基因表达的预测
MoSPR: Histology-to-Gene Expression Prediction with Morpho-Spatial Macrostates and Low-Rank Molecular Programs
arXiv 2609.34280 · 2026-09-29直接自进化优化：无需挑战者训练即可进化大语言模型
Direct Self-Evolving Optimization: Evolving LLMs without Challenger Training
arXiv 2609.34279 · 2026-09-29除环上一侧 Nullstellensatz 的强形式与显式形式
Strong and Explicit Forms of the One-Sided Nullstellensatz over Division Rings
arXiv 2609.34278 · 2026-09-29观察、测量与推理：学习病理学中的视觉接地推理
See, Measure, and Reason: Learning Visually Grounded Reasoning in Pathology
arXiv 2609.34277 · 2026-09-29NavHarness：迈向终身具身导航
NavHarness: Towards Lifelong Embodied Navigation
arXiv 2609.34276 · 2026-09-29一种单因子分解预测-校正长步弧搜索方法及其曲率放大变体用于具有齐次自对偶嵌入的半定规划
A One-Factorization Predictor-Corrector Long-Step Arc-Search Method and a Curvature-Amplified Variant for Semidefinite Programming with a Homogeneous Self-Dual Embedding
arXiv 2609.34275 · 2026-09-29BIABench：评估AI代理在真实生物图像分析任务中的表现
BIABench: Evaluating AI agents on real-world bioimage analysis tasks
arXiv 2609.34274 · 2026-09-29查询扩展与键特化在Transformer注意力几何中的表现
Query Expansion and Key Specialization in Transformer Attention Geometry
arXiv 2609.34273 · 2026-09-29BF16注意力中的对称性破缺：为什么FlashAttention梯度在训练后期会爆炸
Broken Symmetry in BF16 Attention: Why FlashAttention Gradients Blow Up Late in Training
arXiv 2609.34272 · 2026-09-29扩展通用3D资产生成编辑：百万级数据集
Scaling Versatile 3D Assets Editing with a Million-Scale Dataset
arXiv 2609.34271 · 2026-09-29交互式机器人规划中意图消歧的贝叶斯主动学习
Bayesian Active Learning for Intent Disambiguation in Interactive Robot Planning
arXiv 2609.34270 · 2026-09-29舍入公平最大最小多样化问题的球线性规划
Rounding the Ball LP for Fair Max-Min Diversification
arXiv 2609.34269 · 2026-09-29SAGE：面向LLM任务规划器的符号动作门控与编辑
SAGE: Symbolic Action-Gating and Editing for LLM Task Planners
arXiv 2609.34268 · 2026-09-29上三角矩阵李群上的Karcher均值
The Karcher Mean on Lie Groups of Upper Triangular Matrices
arXiv 2609.34267 · 2026-09-29基于最小二乘近似的自适应LASSO惩罚最小密度功率散度估计：应用于SWAN研究的骨密度数据
Adaptive LASSO Penalized Minimum Density Power Divergence Estimation through Least Squares Approximation: Application to Bone Mineral Density Data from the SWAN Study
arXiv 2609.34266 · 2026-09-29ZeroCode：基于零矩阵通过强化学习按需构造纠错码
ZeroCode: On-demand Error-Correcting Code Construction from the Zero Matrix via Reinforcement Learning
arXiv 2609.34265 · 2026-09-29低秩自适应的旋转流形优化
Rotated Manifold Optimization for Low-Rank Adaptation
arXiv 2609.34264 · 2026-09-29通过硬件/软件协同设计改进解释器中的间接分支预测
Improving Indirect Branch Prediction in Interpreters via Hardware/Software Co-Design
arXiv 2609.34263 · 2026-09-29维护基准测试以应对日益强大的智能体：未达标通过的检测与修复
Maintaining Benchmarks Against Increasingly Capable Agents: Detection and Remediation of Unearned Passes
arXiv 2609.34262 · 2026-09-29RoboICL：基于GPT-6 Astra的具身上下文学习
RoboICL: Embodied In-Context Learning with GPT-6 Astra
arXiv 2609.34261 · 2026-09-29QuantaSpike：面向大型语言模型的短窗口脉冲驱动量化
QuantaSpike: Short-Window Spike-Driven Quantization for Large Language Models
arXiv 2609.34259 · 2026-09-29探究人类与AI在多解问题上的差异
Investigating Human--AI Discrepancies via Multiple-Solution Problems
arXiv 2609.34258 · 2026-09-29递归大语言模型在生物医学问答中的退化：一项跨代研究
Recursive LLM Degradation in Biomedical Question Answering: A Cross-Generation Study
arXiv 2609.34257 · 2026-09-29带强制休息的调度：NP-难性与加一近似
Scheduling with Mandatory Breaks: NP-Hardness and an Additive-One Approximation
arXiv 2609.34254 · 2026-09-29DreamingGoose：从自回归Transformer到双向循环扩散语言模型的分阶段蒸馏
DreamingGoose: Staged Distillation from Autoregressive Transformers to Bidirectional Recurrent Diffusion Language Models
arXiv 2609.34253 · 2026-09-29
