arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Wisconsin-Madison(威斯康星大学麦迪逊分校)

2026-01-09 至 2026-01-09 共收录 2
2601.04277 2026-01-09 cs.LG

Unlocking the Pre-Trained Model as a Dual-Alignment Calibrator for Post-Trained LLMs

解封预训练模型作为后训练LLMs的双对齐校准器

Beier Luo, Cheng Wang, Hongxin Wei, Sharon Li, Xuefeng Du

机构 * Department of Statistics and Data Science, Southern University of Science and Technology(统计与数据科学系,南方科技大学) School of Computing, National University of Singapore(计算学院,新加坡国立大学) Department of Computer Sciences, University of Wisconsin-Madison(计算机科学系,威斯康星大学麦迪逊分校) College of Computing and Data Science, Nanyang Technological University(计算与数据科学学院,南洋理工大学)

AI总结 本文提出Dual-Align方法,通过双对齐策略校正后训练LLMs的置信度漂移和过程漂移,提升校准性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.22009 2026-01-09 cs.CV

StreamFlow: Theory, Algorithm, and Implementation for High-Efficiency Rectified Flow Generation

StreamFlow:高效率校正流生成的理论、算法与实现

Sen Fang, Hongbin Zhong, Yalin Feng, Yanxin Zhang, Dimitris N. Metaxas

机构 * Rutgers University, New Jersey, USA(罗杰斯大学) Georgia Institute of Technology, Atlanta, Georgia, USA(佐治亚理工学院) Nanyang Technological University, Singapore(南洋理工大学) University of Wisconsin-Madison, Wisconsin, USA(威斯康星大学麦迪逊分校)

AI总结 本文提出StreamFlow,通过理论、算法和实现的综合优化,显著提升了基于流模型的图像生成效率,达到611%的加速效果。

Comments Improved the quality. Project Page at https://world-snapshot.github.io/StreamFlow/

详情

展开后加载摘要…

URL PDF HTML 收藏