殷东硕
邮 件： yinds@ustc.edu.cn
所属单位：人工智能与数据科学学院
个人主页：https://www.yindongshuo.top/
主要研究方向：计算机视觉、遥感图像解译、模型高效迁移、多模态学习
个人简介：殷东硕，中国科学技术大学人工智能与数据科学学院，预聘副教授。2024年博士毕业于中国科学院空天信息创新研究院目标认知与应用技术国家级重点实验室，师从孙显研究员。2024至2026年在清华大学计算机系从事博士后工作，师从胡事民院士。2025年入选人社部博士后创新人才支持计划（博新计划，全国计算机科学与技术学科20人），2024年入选清华大学水木学者计划，同年获中国科学院院长奖。主持国家自然科学基金青年基金项目C类、中国博士后科学基金面上项目各一项，并参与国家自然科学基金重大项目一项。已在NatureCommunications、CVPR、ICCV等国际顶级期刊/会议发表一作论文多篇，并长期担任NeurIPS、CVPR、ICLR等多个国际顶级会议的审稿人。研究方向包括计算机视觉、遥感图像解译以及参数高效微调等领域。相关工作被“中科院之声”、“机器之心”等媒体报导。
近几年代表论著：
[1] Xian Sun*, Dongshuo Yin*, Fei Qin,Hongfeng Yu, Wanxuan Lu, Fanglong Yao, Qibin He, Xingliang Huang, Zhiyuan Yan,Peijin Wang, Chubo Deng, Nayu Liu, Yiran Yang, Wei Liang, Ruiping Wang, ChengWang, Naoto Yokoya, Ronny Hänsch, Kun Fu. Revealing influencing factors onglobal waste distribution via deep-learning based dumpsite detection fromsatellite imagery. Nature Communications, 2023, 14(1): 1444.
[2] Dongshuo Yin, Leiyi Hu, Bin Li, YouqunZhang, Xue Yang. 5%> 100%: Breaking performance shackles of full fine-tuningon visual recognition tasks. Proceedings of the Computer Vision and PatternRecognition Conference. 2025: 20071-20081.
[3] Dongshuo Yin, Yiran Yang, Zhechao Wang,Hongfeng Yu, Kaiwen Wei, Xian Sun. 1% vs 100%: Parameter-efficient low rankadapter for dense predictions. Proceedings of the IEEE/CVF conference oncomputer vision and pattern recognition. 2023: 20116-20126.
[4] Dongshuo Yin, Xueting Han, Bin Li, HaoFeng, Jing Bai. Parameter-efficient is not sufficient: Exploring parameter,memory, and time efficient adapter tuning for dense predictions. Proceedings ofthe 32nd ACM International Conference on Multimedia. 2024: 1398-1406.
[5] Dongshuo Yin, Hongfeng Yu, Nayu Liu,Fanglong Yao, Qibin He, Jihao Li, Yiran Yang, Shiyao Yan, Xian Sun. Gal:Graph-induced adaptive learning for weakly supervised 3d object detection. IEEETransactions on Intelligent Transportation Systems, 2023, 24(9): 9684-9697.
[6] Yiran Yang*, Dongshuo Yin*, Xuee Rong,Xian Sun, Wenhui Diao, Xinming Li. Beyond the Limitation of Monocular 3DDetector via Knowledge Distillation. Proceedings of the IEEE/CVF InternationalConference on Computer Vision. 2023: 9077-9086.
[7] Leiyi Hu, Hongfeng Yu, Wanxuan Lu,Dongshuo Yin, Xian Sun, Kun Fu. AiRs: Adapter in remote sensing forparameter-efficient transfer learning. IEEE Transactions on Geoscience andRemote Sensing, 2024, 62: 1-18.
[8] Leiyi Hu, Wanxuan Lu, Hongfeng Yu,Dongshuo Yin, Xian Sun, Kun Fu. TEA: A training-efficient adapting frameworkfor tuning foundation models in remote sensing. IEEE Transactions on Geoscienceand Remote Sensing, 2024, 62: 1-18.
[9] Dongshuo Yin, Ting-Feng Zhao, Deng-PingFan, Shutao Li, Bo Du, Xian Sun, Shi-Min Hu. Remote sensing tuning: A survey.Computational Visual Media, 2025, 11(5): 897-937.
[10] Zhechao Wang, Yiming Zeng, Lufan Ma,Zeqing Fu, Chen Bai, Dongshuo Yin, Ziyao Lin, Cheng Lu. DrivePTS: A ProgressiveLearning Framework with Textual and Structural Enhancement for Driving SceneGeneration. Proceedings of the IEEE/CVF Conference on Computer Vision andPattern Recognition. 2026: 3699-3708.