近日，计算机视觉领域顶级会议CVPR 2026在美国举行。我院在集成聚类、测试时自适应、视觉大模型训练、图像去噪及多视角学习等方向共有6项研究成果被大会收录。

论文目录

[1]Changhao He, Di Xue, Shuxian Li, Yanji Hao, Xi Peng, Peng Hu. Bootstrapping Multi-view Learning for Test-time NoisyCorrespondence.CVPR 2026.

[2]Haoxiang Sun, Tao Wang, Chenwei Tang, Li Yuan, Jiancheng Lv.Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Design. CVPR 2026.

[3]Jiaxuan Xu, Lei Duan, Xinye Wang, Liang .Large-scale Robust Enhanced Ensemble Clustering via Outlier Decoupling. CVPR 2026.

[4]Xingyu Wang, Tao Wang.FOZO: Forward-Only Zeroth-Order Prompt Optimization for Test-Time Adaptation. CVPR 2026.

[5]Yiwen Shan, Haiyu Zhao, Peng Hu, Xi Peng, Yuanbiao Gou.Next-Scale Prediction: A Self-Supervised Approach for Real-World Image Denoising. CVPR 2026.

[6]Zhenbin Wang, Lei Zhang, Lituan Wang, Zhenwei Zhang, Guangwu Qian, Yan Wang, Wei Huang.Energy Waveify and Redistribution for Test-Time Adaptation: A Control System Perspective. CVPR 2026.