arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Toronto(多伦多大学)

2026-05-19 至 2026-05-19 共收录 6
2605.18012 2026-05-19 cs.CV cs.AI cs.LG

SAS: Semantic-aware Sampling for Generative Dataset Distillation

SAS: 语义感知的生成数据集蒸馏

Mingzhuo Li, Guang Li, Linfeng Ye, Jiafeng Mao, Takahiro Ogawa, Konstantinos N. Plataniotis, Miki Haseyama

机构 * Hokkaido University(北海道大学) University of Toronto(多伦多大学) The University of Tokyo(东京大学)

AI总结 本文提出了一种语义感知的数据集蒸馏方法,通过利用CLIP作为语义先验,设计三个语义评分函数来量化类别相关性、类别间分离性和集合内多样性,从而生成紧凑且语义区分度高的数据集。

Comments Published as a journal paper in IEEE OJSP

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.11276 2026-05-19 stat.ML cs.LG

RIE-Greedy: Regularization-Induced Exploration for Contextual Bandits

RIE-Greedy: 基于正则化的探索策略用于上下文老虎机

Tong Li, Thiago de Queiroz Casanova, Eric M. Schwartz, Victor Kostyuk, Dehan Kong, Joseph J. Williams

机构 * University of Toronto(多伦多大学) University of Michigan(密歇根大学)

AI总结 本文提出了一种基于正则化的探索策略(RIE-Greedy),利用模型拟合过程中的随机性作为内在探索源,理论证明其在两臂老虎机情况下等价于Thompson Sampling,并在大规模商业环境中优于epsilon-greedy等基准方法。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.04070 2026-05-19 cs.CL

T-FIX: Text-Based Explanations with Features Interpretable to eXperts

T-FIX:基于文本的可解释性方法,具备可解释的专家特征

Shreya Havaldar, Weiqiu You, Chaehyeon Kim, Anton Xue, Helen Jin, Marco Gatti, Bhuvnesh Jain, Helen Qu, Amin Madani, Daniel A. Hashimoto, Gary E. Weissman, Rajat Deo, Sameed Khatana, Lyle Ungar, Eric Wong

机构 * Department of Computer and Information Science, University of Pennsylvania(宾夕法尼亚大学计算机与信息科学系) Department of Computer Science, University of Texas at Austin(德克萨斯大学奥斯汀分校计算机科学系) Department of Physics and Astronomy, University of Pennsylvania(宾夕法尼亚大学物理与天文学系) Flatiron Institute(Flatiron研究所) Department of Surgery, Perelman School of Medicine, University of Pennsylvania(宾夕法尼亚大学佩雷尔曼医学院外科系) Division of Pulmonary, Allergy, and Critical Care, Perelman School of Medicine, University of Pennsylvania(宾夕菲亚大学佩雷尔曼医学院呼吸、过敏与危重医学科) Division of Cardiovascular Medicine, Perelman School of Medicine, University of Pennsylvania(宾夕法尼亚大学佩雷尔曼医学院心血管医学科) Department of Surgery, University of Toronto(多伦多大学外科系) University Health Network(大学健康网络)

AI总结 本文提出T-FIX框架,用于评估LLM生成的解释是否符合专家的推理方式,通过七个科学任务和三个领域进行验证,实现了自动且可定制的专家对齐评估。

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.06384 2026-05-19 eess.IV cs.CV

Mitigating 3D Prostate Biparametric MRI Data Scarcity through Domain Adaptation using Locally-Trained Latent Diffusion Models for Prostate Cancer Detection

通过使用本地训练的潜在扩散模型进行领域适应以缓解3D前列腺双参数MRI数据稀缺问题

Emerson P. Grabke, Babak Taati, Masoom A. Haider

机构 * Institute of Biomedical Engineering, University of Toronto(多伦多大学生物医学工程研究所) Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital(圣心医院卢内尔-塔内本研究所) KITE Research Institute, Toronto Rehabilitation Institute, University Health Network(多伦多康复研究所、KITE研究所在大学健康网络) Joint Department of Medical Imaging, University of Toronto, Princess Margaret Hospital, and Sinai Health systems(多伦多大学联合医学影像部门、玛格丽特医院及辛纳医疗系统) Department of Computer Science, University of Toronto(多伦多大学计算机科学系) Faculty Affiliate of the Vector Institute, Toronto(向量研究所教职员工)

AI总结 本文提出CCELLA++,一种新的潜在扩散模型流程,用于同时生成3D双参数前列腺MRI(bpMRI),包括轴向T2加权(AxT2)、高b值扩散系列(HighB)和表观扩散系数图(ADC),以克服数据稀缺问题。

Comments This work has been submitted to the IEEE for possible publication

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.17338 2026-05-19 cs.SE cs.CL

Precise Debugging Benchmark: Is Your Model Debugging or Regenerating?

精确调试基准:你的模型是在调试还是重生成?

Wang Bill Zhu, Miaosen Chai, Shangshang Wang, Yejia Liu, Song Bian, Honghua Dong, Willie Neiswanger, Robin Jia

机构 * University of Southern California(南加州大学) Microsoft(微软) University of Wisconsin–Madison(威斯康星大学麦迪逊分校) University of Toronto(多伦多大学)

AI总结 本文提出PDB基准,评估LLM调试能力,发现前沿模型调试精度低,即使受指令引导也难以达到高精度。

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.04149 2026-05-19 cs.CL cs.AI cs.LG

Difficulty-Based Preference Data Selection by DPO Implicit Reward Gap

基于难度的偏好数据选择:通过DPO隐式奖励差距

Xuan Qi, Rongwu Xu, Zhijing Jin

机构 * Paul G. Allen School of Computer Science & Engineering, University of Washington(华盛顿大学计算机科学与工程保罗·G·艾伦学校) Max Planck Institute for Intelligent Systems, Tübingen, Germany(德国图宾根马克斯·普朗克智能系统研究所) Jinesis Lab, University of Toronto & Vector Institute(多伦多大学Jinesis实验室及向量研究所)

AI总结 本文提出基于难度的偏好数据选择方法,利用DPO隐式奖励机制选择奖励差距小的样本,提升数据效率和模型对齐性能,在多个数据集和对齐任务中优于五个基线方法。

Comments Our code and data are available at https://github.com/Difficulty-Based-Preference-Data-Select/Difficulty-Based-Preference-Data-Select

详情

展开后加载摘要…

URL PDF HTML 收藏