arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

Northeastern University(东北大学)

2026-05-25 至 2026-05-25 共收录 7
2605.23771 2026-05-25 cs.CV cs.AI cs.MA

PhotoFlow: Agentic 3D Virtual Photography Missions

PhotoFlow: 智能体式3D虚拟摄影任务

Jiarui Guo, Haojia Wei, Yiming Zhang, Yifei Liu, Yuning Gong, Hongjie Zhang, Xue Yang, Zhihang Zhong

机构 * Shanghai Jiao Tong University(上海交通大学) Northeastern University(东北大学) University of California, Los Angeles(加州大学洛杉矶分校) Cornell University(康奈尔大学) Shanghai AI Laboratory(上海人工智能实验室) Sichuan University(四川大学)

AI总结 提出PhotoFlow智能体框架,通过导演-评审-反思闭环搜索机制,在任意Blender场景中根据语言指令自动完成虚拟摄影,并引入VPhotoBench基准,实验表明其在外观质量对齐和成功率上优于现有方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.14180 2026-05-25 cs.CV

Progressive $\mathcal{J}$-Invariant Self-supervised Learning for Low-Dose CT Denoising

渐进式 $\mathcal{J}$-不变自监督学习用于低剂量CT去噪

Yichao Liu, Zongru Shao, Yueyang Teng, Junwen Guo

机构 * organization= IWR, Heidelberg University , city= Heidelberg , postcode= 69120 , state= Baden Württemberg , country= Germany organization= Silicon Austria Labs , city= Linz , postcode= 4040 , state= Upper Austria , country= Austria organization= Institute of Science Tokyo , addressline= , city= Tokyo , country= Japan organization= College of Medicine Biological Information Engineering, Northeastern University , city= Shenyang , postcode= 110169 , state= Liaoning , country= China organization= Key Laboratory of Intelligent Computing in Medical Image, Ministry of Education , city= Shenyang , postcode= 110169 , state= Liaoning , country= China organization= Department of Epidemiology \& Global Health, Umeå University , addressline= , city= Umeå , postcode= 90187 , country= Sweden

AI总结 提出渐进式 $\mathcal{J}$-不变学习,通过逐步盲点去噪机制和噪声注入正则化,提升低剂量CT去噪性能,在Mayo数据集上优于现有自监督方法并接近监督方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.23663 2026-05-25 cs.HC cs.LG

Detecting Drunk Driving Using Off-the-Shelf Smartwatches

使用现成智能手表检测酒驾

Robin Deuber, Lanlan Yang, Michal Bechny, Christoph Heck, Matthias Pfäffli, Matthias Bantle, Florian von Wangenheim, Elgar Fleisch, Wolfgang Weinmann, Manuel Günther, Felix Wortmann, Varun Mishra

机构 * University of Bern(伯尔尼大学) University of St. Gallen(施特加尔伦大学) Northeastern University(东北大学)

AI总结 提出利用智能手表的加速度计和心率变异性数据,通过逻辑回归和双塔一维CNN模型检测酒精影响下的驾驶行为,CNN在检测任何酒精中毒和超过WHO推荐限值时的AUROC分别达到0.88和0.86。

Comments 27 pages, 7 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.23067 2026-05-25 cs.CL

What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA

训练数据教会RL记忆体代理什么:记忆增强问答中课程效应的实证研究

Xinjie He, Zhiyuan Lin, Su Liu, Jialun Wu, Qiyang Xie, Weikai Zhou, Shuai Xiao

机构 * Columbia University(哥伦比亚大学) Independent Researcher(独立研究者) Johns Hopkins University(约翰霍普金斯大学) Northeastern University(东北大学)

AI总结 通过控制实验,研究训练课程(领域内、混合、领域外)对强化学习记忆体代理在记忆增强问答中的技能获取和专业化影响,发现课程组成作为专业化的细粒度杠杆,混合课程在整体F1上最优,而窄领域外课程可转移特定技能(如时间推理),并报告了将GRPO适配到单GPU环境的实用经验。

Comments 14 pages, 2 figures, 11 tables. Code, checkpoints, and evaluation artifacts available at https://github.com/EvaxHe/rl-memory-curriculum

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.22962 2026-05-25 cs.CV cs.CE cs.HC cs.SE q-bio.NC

GazeBehavior Annotation Toolkit (GBAT): AI-powered toolkit for automatic annotation of egocentric eye-tracking and video data of child-caregiver interaction

凝视行为注释工具包 (GBAT): 基于AI的自动注释工具,用于自我中心眼动追踪和儿童-照顾者互动视频数据

Iba Baig, Kevin Li, Yanbin Xu, Seiji Cattelain, Marie Hallo, Hayato Ono, Sho Tsuji, Ming Bo Cai

机构 * Department of Psychology, University of Miami(迈阿密大学心理学系) Northeastern University(东北大学) Ecole Normale Supérieure, PSL University, EHESS, CNRS(巴黎高等师范学院(PSL大学)、EHESS、CNRS) International Research Center for Neurointelligence (WPI-IRCN), The University of Tokyo Institutes for Advanced Study(神经智能国际研究中心(WPI-IRCN)、东京大学高级研究机构)

AI总结 提出GBAT工具包,利用深度学习实现多视频同步、注视目标半自动注释和参与者姿态手部动作分类,提高从自我中心眼动和视频数据中提取特征的效率和可扩展性。

Comments submitted to IEEE International Conference on Development and Learning (ICDL), 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.18911 2026-05-25 cs.LG cs.AI

Does Your Wildfire Prediction Model Actually Work, or Just Score Well?

你的野火预测模型真的有效,还是只是得分高?

Yangshuang Xu, Yuyang Dai, Liling Chang, Qi Wang, Yushun Dong

机构 * Florida State University(佛罗里达州立大学) Northeastern University(东北大学)

AI总结 针对现有地球基础模型不专门用于野火预测的问题,提出首个专门预训练的野火基础模型WILDFIRE-FM,并引入固定合约评估框架以解决评估设计对结论敏感性的问题。

Comments 25 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.08036 2026-05-25 cs.LG

Heterogeneous Sheaf Neural Networks

异质层丛神经网络

Luke Braithwaite, Alessio Borgi, Gabriele Onorato, Kristjan Tarantelli, Francesco Restuccia, Fabrizio Silvestri, Pietro Liò

机构 * Department of Computer Science and Technology, University of Cambridge(计算机科学与技术系,剑桥大学) Department of Electrical and Computer Engineering, Northeastern University(电气与计算机工程系,东北大学) Department of Computer, Control and Management Engineering, Sapienza University of Rome(计算机、控制与管理工程系,罗马萨皮恩扎大学)

AI总结 提出HetSheaf框架,通过细胞层丛表示异质图的类型感知局部特征空间和学习限制映射,在节点分类、链接预测和图分类任务上优于现有方法且参数更少。

Comments 48 pages, 2 figures

详情

展开后加载摘要…

URL PDF HTML 收藏