arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

Cornell University(康奈尔大学)

共收录 951
2603.18514 2026-04-28 stat.ML cs.LG

On the Peril of (Even a Little) Nonstationarity in Satisficing Regret Minimization

关于在满足满意遗憾最小化中即使少量非平稳性的危险

Yixuan Zhang, Ruihao Zhu, Qiaomin Xie

机构 * Department of Industrial & Systems Engineering, University of Wisconsin-Madison(威斯康星大学麦迪逊分校工业与系统工程系) SC Johnson College of Business, Cornell University(康奈尔大学SC约翰逊商学院)

AI总结 研究非平稳K臂老虎机的满足满意遗憾保证,发现当L≥2时最优遗憾为Θ(L log T),而L=1时可实现T无关的Θ(1)遗憾。

Comments 20 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.22714 2026-04-27 cs.CV

Long-tail Internet photo reconstruction

长尾互联网照片重建

Yuan Li, Yuanbo Xiangli, Hadar Averbuch-Elor, Noah Snavely, Ruojin Cai

机构 * Cornell University(康奈尔大学) Kempner Institute, Harvard University(哈佛大学凯普勒研究所)

AI总结 本文针对长尾分布的互联网照片重建问题,提出MegaDepth-X数据集和采样策略,提升极端稀疏场景下的3D重建鲁棒性,并增强对对称重复场景的重建可靠性。

Comments Project page: https://megadepth-x.github.io/

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.22202 2026-04-27 cs.CV

ArchSym: Detecting 3D-Grounded Architectural Symmetries in the Wild

ArchSym:在真实场景中检测3D基础的建筑对称性

Hanyu Chen, Ruojin Cai, Steve Marschner, Noah Snavely

机构 * Cornell University(康奈尔大学) Kempner Institute, Harvard University(哈佛大学凯普勒研究所)

AI总结 本文提出首个基于单张真实RGB图像检测3D基础反射对称性的框架,通过自动生成大规模建筑对称性数据集ArchSym,并引入单视角对称检测器,实现对3D对称性的精确定位。

Comments project page: https://hanyuc.com/archsym/

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.14135 2026-04-27 cs.LG cs.AI cs.GT cs.MA

AdaFair-MARL: Enforcing Adaptive Fairness Constraints in Multi-Agent Reinforcement Learning

AdaFair-MARL:在多智能体强化学习中强制适应性公平性约束

Promise Ekpo, Saesha Agarwal, Felix Grimm, Lekan Molu, Angelique Taylor

机构 * Cornell Tech(康奈尔科技) Cornell University(康奈尔大学) Bala-Cynwyd, PA, USA(巴拉-辛威德, 美国宾夕法尼亚州)

AI总结 本文提出AdaFair-MARL框架,通过将工作负载公平性作为显式约束,使智能体在优化团队性能的同时保持贡献平衡,实验表明其在医院协调环境中的有效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.07298 2026-04-27 cs.LG cs.AI

Pre-trained Large Language Models Learn Hidden Markov Models In-context

预训练大语言模型通过上下文学习学习隐马尔可夫模型

Yijia Dai, Zhaolin Gao, Yahya Sattar, Sarah Dean, Jennifer J. Sun

机构 * Cornell University(康奈尔大学)

AI总结 本文研究预训练大语言模型通过上下文学习有效建模由隐马尔可夫模型生成的数据,并揭示了其在复杂数据中挖掘隐藏结构的潜力。

Comments NeurIPS 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.20904 2026-04-24 cs.LG cs.AI

Reinforcing privacy reasoning in LLMs via normative simulacra from fiction

通过小说中的规范仿像强化大语言模型的隐私推理

Matt Franchi, Madiha Zahrah Choksi, Harold Triedman, Helen Nissenbaum

机构 * Cornell Tech(康奈尔科技)

AI总结 本文通过从小说中提取规范仿像,结合监督学习和GRPO强化学习,提升大语言模型的隐私推理能力,验证了小说来源的规范仿像在现实领域中的有效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.16797 2026-04-24 cs.LG cs.CV

Adaptive Moments are Surprisingly Effective for Plug-and-Play Diffusion Sampling

自适应矩在即插即用扩散采样中出人意料地有效

Christian Belardi, Justin Lovelace, Kilian Q. Weinberger, Carla P. Gomes

机构 * Cornell University(康奈尔大学)

AI总结 本文提出利用自适应矩估计来稳定扩散采样中的噪声似然分数,通过实验证明该方法在图像修复和类别条件生成任务中优于复杂方法,提升了对齐效果。

详情

展开后加载摘要…

URL PDF HTML 收藏
2504.07940 2026-04-24 cs.CV

Beyond the Frame: Generating 360 Panoramic Videos from Perspective Videos

超越帧限:从视角视频生成360度全景视频

Rundong Luo, Matthew Wallingford, Ali Farhadi, Noah Snavely, Wei-Chiu Ma

机构 * Cornell University(康奈尔大学) University of Washington(华盛顿大学)

AI总结 本文提出从视角视频生成完整全景视频的方法,通过设计几何与运动感知操作提升生成质量,实现空间与时间一致性,应用于视频稳定、视角控制和交互式视觉问答。

Comments Project page: https://red-fairy.github.io/argus/

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.15339 2026-04-23 cs.CL cs.AI

AutoGraph-R1: End-to-End Reinforcement Learning for Knowledge Graph Construction

AutoGraph-R1:知识图谱构建的端到端强化学习

Hong Ting Tsang, Jiaxin Bai, Haoyu Huang, Qiao Xiao, Tianshi Zheng, Baixuan Xu, Shujie Liu, Yangqiu Song

机构 * The Hong Kong University of Science and Technology(香港理工大学) Cornell University(康奈尔大学) Microsoft Research(微软研究院)

AI总结 本文提出AutoGraph-R1框架,通过强化学习直接优化知识图谱构建,提升检索增强生成任务性能。设计了两种任务感知奖励函数,使图谱作为知识载体和索引,实现构建与应用的闭环。

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.16658 2026-04-23 math.ST cs.LG stat.ML stat.TH

Multi-Armed Bandits With Machine Learning-Generated Surrogate Rewards

多臂老虎机与机器学习生成的代理奖励

Wenlong Ji, Yihan Pan, Ruihao Zhu, Lihua Lei

机构 * Stanford University(斯坦福大学) Northwestern University(西北大学) Cornell University(康奈尔大学)

AI总结 本文提出MLA-UCB算法,利用机器学习模型将辅助信息转换为代理奖励,解决代理奖励偏差问题,改进累积遗憾并实现渐近最优。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.04983 2026-04-22 eess.IV cs.AI cs.LG

AI-Based Detection of Temporal Changes in MR-Linac Images Acquired During Routine Prostate Radiotherapy

基于人工智能的MR-Linac图像中时间变化检测研究

Seungbin Park, Peilin Wang, Ryan Pennell, Emily S. Weg, Himanshu Nagar, Timothy McClure, Mert R. Sabuncu, Daniel Margolis, Heejong Kim

机构 * Radiology, Weill Cornell Medicine(放射科,威尔·康奈尔医学大学) Department of Health Technology and Informatics, The Hong Kong Polytechnic University(健康科技与信息学系,香港理工大学) Radiation Oncology, Weill Cornell Medicine(放射肿瘤学,威尔·康奈尔医学大学) Department of Radiation Oncology, Memorial Sloan Kettering Cancer Center(放射肿瘤学部,纪念斯隆凯特琳癌症中心) School of Electrical and Computer Engineering, Cornell University and Cornell Tech(电气与计算机工程学院,康奈尔大学和康奈尔科技)

AI总结 本文研究了基于AI的方法在MR-Linac图像中检测放射治疗期间细微时间变化的可行性,通过深度学习模型和放射科医生的对比,验证了AI在前列腺放疗中的应用潜力。

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.17869 2026-04-21 eess.IV cs.CV cs.LG

Integrating Feature Selection and Machine Learning for Nitrogen Assessment in Grapevine Leaves using In-Field Hyperspectral Imaging

整合特征选择与机器学习用于葡萄叶氮含量评估的田间高光谱成像

Atif Bilal Asad, Achyut Paudel, Safal Kshetri, Chenchen Kang, Salik Ram Khanal, Nataliya Shcherbatyuk, Pierre Davadant, R. Paul Schreiner, Santosh Kalauni, Manoj Karkee, Markus Keller

机构 * organization= Center for Precision Automated Agricultural Systems , addressline= Washington State University , city= Prosser , postcode= 99350 , state= WA , country= USA organization= Biological \& Environmental Engineering Department , addressline= Cornell University , city= Ithaca , postcode= 14853 , state= NY , country= USA organization= Fruit Research Extension Center , addressline= The Penn State University , city= Biglerville , postcode= 17307 , state= PA , country= USA organization= School of Business Technology , addressline= Curry College , city= Milton , postcode= 02186 , state= MA , country= USA organization= Department of Viticulture Enology , addressline= Washington State University , city= Prosser , postcode= 99350 , state= WA , country= USA Genetic Improvement Research Unit (HCPGIRU) , city= Corvallis , postcode= 973300 , state= OR , country= USA organization= Mid-Columbia Agricultural Research Extension Center , addressline = Oregon State University , city= Corvallis , postcode= 97031 , state= OR , country= USA organization= Department of Plant Sciences , addressline = University of Tennessee, Institute of Agriculture , city= Knoxville , postcode= 37996 , state= TN , country= USA

AI总结 本文通过田间高光谱成像与机器学习整合,开发了特征选择框架以提高葡萄叶氮含量预测精度,验证了在不同生长阶段和品种间的方法有效性。

Comments Major Revision

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.16106 2026-04-21 cs.CY cs.AI

Reckoning with the Political Economy of AI: Avoiding Decoys in Pursuit of Accountability

审视人工智能的政治经济学:避免诱饵以实现问责

Janet Vertesi, danah boyd, Alex Taylor, Benjamin Shestakofsky

机构 * Sociology Department(社会学系) Princeton University(普林斯顿大学) Department of Communication(传播学系) Cornell University(康奈尔大学) Institute of Design Informatics(设计信息研究所) University of Edinburgh(爱丁堡大学) Department of Information Science(信息科学系)

AI总结 本文探讨人工智能项目中诱饵对权力与经济的影响,指出需识别诱饵的误导性并直接面对其物质政治经济,以促进更公平的AI发展。

Comments To be presented at ACM FAccT, Montréal, Canada, June 25 to June 28, 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.10906 2026-04-21 math.OC cs.LG cs.SY eess.SY

Distributionally Robust Regret Optimal Control Under Moment-Based Ambiguity Sets

在基于矩的模糊集下实现分布鲁棒的后悔最优控制

Feras Al Taha, Eilyan Bitar

机构 * School of Electrical and Computer Engineering, Cornell University(康奈尔大学电气与计算机工程学院)

AI总结 本文研究了有限时间 horizon 的线性二次随机控制问题,设计了因果仿射控制策略以最小化模糊集内所有分布的最坏预期后悔,将其转化为可解的凸优化问题,并提出可扩展的投影子梯度方法进行计算。

Comments 24 pages, 4 figures, to appear in the Proceedings of the 8th Annual Learning for Dynamics & Control Conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2407.06048 2026-04-21 cs.CL cs.CV

Vision-Braille: A Curriculum Learning Toolkit and Braille-Chinese Corpus for Braille Translation

Vision-Braille: 一种课程学习工具包及用于盲文翻译的盲文-中文语料库

Alan Wu, Ye Yuan, Zhiping Xiao, Ming Zhang

机构 * State Key Laboratory for Multimedia Information Processing, School of Computer Science, PKU-Anker LLM Lab, Peking University(多媒体信息处理国家重点实验室,计算机学院,PKU-Anker LLM实验室,北京大学) College of Agriculture and Life Sciences, Cornell University(农业与生命科学学院,康奈尔大学) Computer Science Department, University of California at Los Angeles(计算机科学系,加州大学洛杉矶分校)

AI总结 本文提出Vision-Braille,首个可公开获取的端到端系统,用于将图像中提取的中文盲文翻译成书面中文。该系统解决有限标注资源和声调缺失问题,结合鲁棒的盲文OCR流程与针对序列到序列翻译微调的LLM。

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.02871 2026-04-21 cs.CL cs.AI

Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoning

位置:多模态大语言模型可以显著推动科学推理

Yibo Yan, Shen Wang, Jiahao Huo, Jingheng Ye, Zhendong Chu, Xuming Hu, Philip S. Yu, Carla Gomes, Bart Selman, Qingsong Wen

机构 * Squirrel AI HKUST(GZ)(香港科技大学(广州)) HKUST(香港科技大学) Tsinghua University(清华大学) University of Illinois at Chicago(伊利诺伊大学香槟分校) Cornell University(康奈尔大学)

AI总结 本文探讨多模态大语言模型在科学推理中的应用,提出四阶段研究路线,指出当前模型在跨领域推理中的潜力与挑战,为实现通用人工智能提供新视角。

Comments Accepted by The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026, Findings)

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.16403 2026-04-21 cs.AI cs.CY

Computational Hermeneutics: Evaluating generative AI as a cultural technology

计算阐释学:评估生成式AI作为文化技术

Cody Kommers, Ruth Ahnert, Maria Antoniak, Emmanouil Benetos, Steve Benford, Mercedes Bunz, Baptiste Caramiaux, Shauna Concannon, Martin Disley, James Dobson, Yali Du, Edgar Duéñez-Guzmán, Kerry Francksen, Evelyn Gius, Jonathan W. Y. Gray, Ryan Heuser, Sarah Immel, Richard Jean So, Sang Leigh, Dalaki Livingston, Hoyt Long, Meredith Martin, Georgia Meyer, Daniela Mihai, Ashley Noel-Hirst, Kirsten Ostherr, Deven Parker, Yipeng Qin, Jessica Ratcliff, Emily Robinson, Karina Rodriguez, Adam Sobey, Ted Underwood, Aditya Vashistha, Matthew Wilkens, Youyou Wu, Yuan Zheng, Drew Hemment

机构 * The Alan Turing Institute(艾伦·图灵研究所) Queen Mary University of London(伦敦玛丽女王大学) University of Colorado(科罗拉多大学) University of Nottingham(诺丁汉大学) King’s College London(伦敦国王学院) Sorbonne Université(索邦大学) Durham University(杜伦大学) University of Edinburgh(爱丁堡大学) Dartmouth College(达特茅斯学院) Gibran AI(吉布兰人工智能) University of Coventry(科文特大学) Technische Universität Darmstadt(德累斯顿技术大学) University of Cambridge(剑桥大学) McGill University(麦吉尔大学) Cornell University(康奈尔大学) University of Utah(犹他大学) University of Chicago(芝加哥大学) Princeton University(普林斯顿大学) London School of Economics(伦敦经济学院) University of Southampton(南安普顿大学) Rice University(德克萨斯大学稻谷分校) University of Glasgow(格拉斯哥大学) Cardiff University(卡迪夫大学) University of Exeter(埃克塞特大学) University of Brighton(布里斯托尔大学) University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校) University College London(伦敦大学学院) University of Sheffield(谢菲尔德大学)

AI总结 本文提出计算阐释学框架,旨在通过阐释学理论评估生成式AI作为文化技术的含义生成与理解挑战。

Comments Published in Frontiers in Artificial Intelligence

Journal ref Front. Artif. Intell. 9:1753041

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.08626 2026-04-21 cs.CV

WildDet3D: Scaling Promptable 3D Detection in the Wild

WildDet3D: 在真实世界中扩展可提示的3D检测

Weikai Huang, Jieyu Zhang, Sijun Li, Taoyang Jia, Jiafei Duan, Yunqian Cheng, Jaemin Cho, Matthew Wallingford, Rustin Soraki, Chris Dongjoo Kim, Shuo Liu, Donovan Clay, Taira Anderson, Winson Han, Ali Farhadi, Bharath Hariharan, Zhongzheng Ren, Ranjay Krishna

机构 * Allen Institute for AI(人工智能研究院) University of Washington(华盛顿大学) Cornell University(康奈尔大学) Johns Hopkins University(约翰霍普金斯大学)

AI总结 本文提出WildDet3D,一种统一的几何感知架构,支持文本、点和框提示,并在推理时整合辅助深度信号。同时,构建了最大的开放3D检测数据集WildDet3D-Data,包含13.5K类别的100万张图像,在多个基准测试中取得新突破。

Comments code: https://github.com/allenai/WildDet3D website: https://allenai.github.io/WildDet3D/

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.15672 2026-04-20 cs.LG cs.CL

Faster LLM Inference via Sequential Monte Carlo

通过序列蒙特卡洛方法加速大语言模型推理

Yahya Emara, Mauricio Barba da Costa, Chi-Chih Chang, Cameron Freer, Tim Vieira, Ryan Cotterell, Mohamed S. Abdelfattah

机构 * Cornell University(康奈尔大学) Makora(马科拉) MIT(麻省理工学院) ETH Zürich(苏黎世联邦理工学院)

AI总结 本文提出SMC-SD方法,通过重要性加权重采样替代传统拒绝采样,提升推理速度,实验证明在多个基准测试中保持高精度的同时,比推测解码和自回归解码分别快2.36倍和5.2倍。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.15555 2026-04-20 cs.CV

CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification

CXR-LT 2026挑战:多中心长尾和零样本胸片分类

Hexin Dong, Yi Lin, Pengyu Zhou, Fengnian Zhao, Alan Clint Legasto, Juno Cho, Dohui Kim, Justin Namuk Kim, Mingeon Kim, Sunwoo Kwak, Gabriel Moyà-Alcover, Ky Trung Nguyen, Thanh-Huy Nguyen, Ha-Hieu Pham, Huy-Hieu Pham, Huy Le Pham, Nikhileswara Rao Sulake, Aina Tur-Serrano, Ruichi Zhang, Ang Zu, Adam E. Flanders, Zhiyong Lu, Ronald M. Summers, Mingquan Lin, Hao Chen, Yuzhe Yang, George Shih, Yifan Peng

机构 * Department of Population Health Sciences, Weill Cornell Medicine(韦尔·科恩医学中心流行病学与公共卫生科学系) Department of Radiology, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College(中国医学科学院阜外医院心血管病国家中心放射科) Department of Radiology, West China School of Medicine, Sichuan University(四川大学西昌医学院放射科) Sichuan University Affiliated Chengdu Second People’s Hospital(四川大学附属成都第二人民医院) Department of Radiology, Weill Cornell Medicine(韦尔·科恩医学中心放射科) School of Electrical Engineering, Korea Advanced Institute of Science and Technology(韩国科学技术院电子工程学院) Gwangju Institute of Science and Technology(光州科学技术院) Department of Biomedical Engineering, Case Western Reserve University(凯斯西储大学生物医学工程系) School of Electrical and Computer Engineering, Cornell Tech(康奈尔科技学院电气与计算机工程系) Department of Mathematics and Computer Science, Universitat de les Illes Balears(巴利阿里大学数学与计算机科学系) School of Computer Science and Engineering, VNU-HCM International University(VNU-HCM国际大学计算机科学与工程系) Vietnam National University, Ho Chi Minh City(越南国家大学河内市分校) School of Computer Science, Carnegie Mellon University(卡内基梅隆大学计算机科学系) VNU-HCM University of Science, Ho Chi Minh City(VNU-HCM科技大学河内市分校)

AI总结 本文提出CXR-LT 2026挑战,通过多中心数据集和零样本任务,研究长尾分布和开放世界下的胸片分类问题,评估视觉语言模型在罕见疾病检测中的表现。

Comments 25 pages, 6 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.19104 2026-04-20 cs.LG stat.ML

Online Distributionally Robust LLM Alignment via Regression to Relative Reward

通过回归相对奖励实现在线分布鲁棒LLM对齐

Sharan Sahu, Martin T. Wells

机构 * Department of Statistics and Data Science(统计与数据科学系) Cornell University(康奈尔大学)

AI总结 本文提出DRO-REBEL方法,通过类型-p Wasserstein、KL和χ²模糊集实现分布鲁棒优化,证明了在偏好转移下的参数误差界,并在多个基准测试中优于现有基线。

Comments 70 pages, 7 figures, 1 table

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.14474 2026-04-17 cs.LG

Scouting By Reward: VLM-TO-IRL-Driven Player Selection For Esports

通过奖励进行 scouting:由 VLM-TO-IRL 驱动的电子竞技玩家选拔

Qing Yan, Wenyu Yang, Yufei Wang, Wenhao Ma, Linchong Hu, Yifei Jin, Anton Dahbura

机构 * Johns Hopkins University(约翰霍普金斯大学) University of Pennsylvania(宾夕法尼亚大学) Cornell University(康奈尔大学)

AI总结 本文提出一种基于逆强化学习的电子竞技玩家选拔框架,通过学习专业选手的奖励函数,利用多模态双分支架构和 GAIL 目标实现玩家风格匹配评估,提升人才发现效率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.06296 2026-04-17 cs.LG cs.AI cs.MA cs.SE

AgentOpt v0.1 Technical Report: Client-Side Optimization for LLM-Based Agent

AgentOpt v0.1 技术报告:基于LLM的代理的客户端优化

Wenyue Hua, Sripad Karne, Qian Xie, Armaan Agrawal, Nikos Pagonas, Kostis Kaffes, Tianyi Peng

机构 * Microsoft Research, AI Frontiers(微软研究院,人工智能前沿) Cornell University(康奈尔大学) Columbia University(哥伦比亚大学)

AI总结 本文提出AgentOpt框架,用于解决LLM代理客户端资源分配问题,通过多种搜索算法优化模型选择和资源分配,提升效率与效果。

Comments 24 pages, 1 figure

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.04442 2026-04-16 cs.CV cs.CL

Addressing Overthinking in Large Vision-Language Models via Gated Perception-Reasoning Optimization

通过门控感知-推理优化解决大视觉-语言模型中的过度思考

Xingjian Diao, Zheyuan Liu, Chunhui Zhang, Weiyi Wu, Keyi Kong, Lin Shi, Kaize Ding, Soroush Vosoughi, Jiang Gui

机构 * Dartmouth College(达特茅斯学院) University of Notre Dame(诺丁汉大学) Cornell University(康奈尔大学) Northwestern University(西北大学)

AI总结 本文提出GPRO方法,通过动态路由计算路径提升大视觉-语言模型的推理效率与准确性,减少过度思考带来的冗余响应。

Comments Accepted to Annual Meeting of the Association for Computational Linguistics (ACL 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.04671 2026-04-16 cs.RO cs.AI cs.CV

X-Diffusion: Training Diffusion Policies on Cross-Embodiment Human Demonstrations

X-Diffusion:在跨具身人类示范上训练扩散策略

Maximus A. Pace, Prithwish Dan, Chuanruo Ning, Atiksh Bhardwaj, Audrey Du, Edward W. Duan, Wei-Chiu Ma, Kushal Kedia

机构 * Cornell University(康奈尔大学)

AI总结 本文提出X-Diffusion框架,通过在噪声人类动作上训练扩散策略,提升机器人任务成功率16%。

Comments ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.11715 2026-04-15 cs.CV

Point Prompting: Counterfactual Tracking with Video Diffusion Models

点提示:基于视频扩散模型的反事实跟踪

Ayush Shrivastava, Sanyam Mehta, Daniel Geng, Andrew Owens

机构 * University of Michigan(密歇根大学) Cornell University(康奈尔大学)

AI总结 本文提出利用视频扩散模型进行零样本点跟踪,通过提示模型在视频中标记移动点,实现反事实生成,实验表明其跟踪效果优于传统方法且能克服遮挡。

Comments ICLR 2026. Project link: https://point-prompting.github.io

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.11993 2026-04-15 cs.CV physics.optics

Ultra-low-light computer vision using trained photon correlations

利用训练的光子相关性实现超低光照计算机视觉

Mandar M. Sohoni, Jérémie Laydevant, Mathieu Ouellet, Shi-Yuan Ma, Ryotatsu Yanagimoto, Benjamin A. Ash, Tatsuhiro Onodera, Tianyu Wang, Logan G. Wright, Peter L. McMahon

机构 * School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853, USA(应用与工程物理学院,康奈尔大学,纽约州伊萨卡市,14853,美国) NTT Physics and Informatics Laboratories, NTT Research Inc., Sunnyvale, CA 94085, USA(NTT物理与信息实验室,NTT研究公司,硅谷,加利福尼亚州94085,美国) Department of Electrical and Computer Engineering, Boston University, Boston, MA 02215, USA(电气与计算机工程系,波士顿大学,马萨诸塞州波士顿市,02215,美国) Department of Applied Physics, Yale University, New Haven, CT 06511, USA(应用物理系,耶鲁大学,新 Haven,康涅狄格州06511,美国) Kavli Institute at Cornell for Nanoscale Science, Cornell University, Ithaca, NY 14853, USA(康奈尔大学纳米科学研究所,康奈尔大学,纽约州伊萨卡市,14853,美国)

AI总结 本文提出通过训练光子相关性增强的光子照明源与Transformer后端进行物体识别,提升低光照和噪声条件下的分类准确率,比传统方法提高15个百分点。

Comments 49 pages, 47 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.20635 2026-04-15 cs.CL cs.AI

Why Did Apple Fall: Evaluating Curiosity in Large Language Models

苹果为何坠落:评估大语言模型中的好奇心

Haoyu Wang, Sihang Jiang, Yuyan Chen, Xiaojun Meng, Jiansheng Wei, Yitong Wang, Yanghua Xiao

机构 * Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University(上海数据科学 key laboratory,计算机科学学院,复旦大学) Huawei Large Model Data Technology Lab(华为大模型数据技术实验室) Cornell University(康奈尔大学)

AI总结 本文基于人类好奇心评估量表,设计评估框架,发现大语言模型对知识有更强的渴求,但在不确定环境中仍保守,好奇心可提升推理与学习能力。

Comments ACL 2026 findings paper

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.01152 2026-04-14 cs.CL

MASH: Modeling Abstention via Selective Help-Seeking

MASH:通过选择性求助进行模型退避建模

Mustafa Omer Gul, Claire Cardie, Tanya Goyal

机构 * Department of Computer Science, Cornell University(康奈尔大学计算机科学系)

AI总结 本文提出MASH框架,通过强化学习实现LLM退避决策,提升多跳问答准确率7.6%,展示出优于传统方法的退避性能。

Comments 25 pages, with 15 dedicated to citations and appendix. 17 tables and 11 figures. Preprint, under review. Paper updated to reflect new title and results

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.08805 2026-04-13 cs.CR cs.AI

Building Better Environments for Autonomous Cyber Defence

构建更好的自主网络防御环境

Chris Hicks, Elizabeth Bates, Shae McFadden, Isaac Symes Thompson, Myles Foley, Ed Chapman, Nickolas Espinosa Dice, Ankita Samaddar, Joshua Sylvester, Himanshu Neema, Nicholas Butts, Nate Foster, Ahmad Ridley, Zoe M, Paul Jones

机构 * The Alan Turing Institute(艾伦·图灵研究所) University College London(伦敦大学学院) Cornell University(康奈尔大学) Vanderbilt University(范德比尔特大学) Microsoft(微软) NSA(美国国家安全局)

AI总结 本文通过工作坊总结,提出构建更有效的强化学习环境框架和最佳实践指南,以提升自主网络防御系统的训练与评估。

详情

展开后加载摘要…

URL PDF HTML 收藏