研究领域为机器学习、计算机视觉、AI4Science等相关方向，主要兴趣包括异常检测、多任务学习、基于多模态大模型的细粒度视频理解、科学智能计算等。

2016-11--中国科学院自动化研究所   副研究员
2013-07--2016-10   中国科学院自动化研究所   助理研究员
2010-09--2013-06   中国科学院自动化研究所   博士
2009-07--2010-07   上海炬力集成电路设计有限公司   系统工程师
2007-09--2009-06   华中科技大学   硕士
2003-09--2007-06   浙江大学   学士

   

2016-11~现在, 中国科学院自动化研究所, 副研究员
2013-07~2016-10,中国科学院自动化研究所, 助理研究员
2009-07~2010-07,上海炬力集成电路设计有限公司, 系统工程师

2017-12-31-2021-03-30,图象图形学学会国际交流与合作委员会, 秘书长
2017-11-16-今,自动化学会模式识别与机器智能专委会, 委员

模式识别与机器学习

   

（1） A Synergistic Strategy for Data-Constrained Deep Learning in Materials Science, Materials Genome Engineering Advances, 2026, 第 5 作者  通讯作者
（2） Gradient Guided LoRA for Stable Fine-tuning of LLMs, International Conference on Pattern Recognition, 2026, 第 2 作者  通讯作者
（3） Fine-Grained Post-Training Quantization for Large Vision Language Models with Quantization-Aware Integrated Gradients, The IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2026, 第 8 作者  通讯作者
（4） Hybrid Generative Diffusion Modeling for Auto-Regressive Prediction of Supersonic Combustion Dynamics, Physics of Fluids, 2026, 第 4 作者
（5） VAGU & GtS: LLM-Based Benchmark and Framework for Joint Video Anomaly Grounding and Understanding, Annual AAAI Conference on Artificial Intelligence, 2026, 第 2 作者  通讯作者
（6） FARSS: Fisher-Optimized Adaptive Low-Rank and Singular-Vector Selection for Knowledge-Preserving Fine-Tuning, The 64th Annual Meeting of the Association for Computational Linguistics (ACL) Findings, 2026, 第 8 作者
（7） Semi-supervised Video Anomaly Detection With Compact Deformable 3D Convolution, Semi-supervised Video Anomaly Detection With Compact Deformable 3D Convolution, International Conference on Acoustics, Speech and Signal Processing, 2025, 第 2 作者  通讯作者
（8） SUVAD: Semantic Understanding Based Video Anomaly Detection Using MLLM, SUVAD: Semantic Understanding Based Video Anomaly Detection Using MLLM, International Conference on Acoustics, Speech and Signal Processing, 2025, 第 2 作者  通讯作者
（9） Region-Aware Instruction-Guided Image Editing with Attention-Weighted Feature Fusion, Region-Aware Instruction-Guided Image Editing with Attention-Weighted Feature Fusion, International Conference on Neural Information Processing, 2024, 第 2 作者  通讯作者
（10） Scene-Adaptive SVAD Based OnMulti-modal Action-Based Feature Extraction, Scene-Adaptive SVAD Based OnMulti-modal Action-Based Feature Extraction, Asian Conference on Computer Vision, 2024, 第 2 作者  通讯作者
（11） Computing Approximate Nash Equilibrium in Two-Team Zero-Sum Games by NashConv Descent, International Conference on Neural Information Processing, 2024, 第 3 作者
（12） A Stable Long-Term Tracking Method for Group-Housed Pigs, A Stable Long-Term Tracking Method for Group-Housed Pigs, International Conference on Image and Graphics, 2023, 第 3 作者  通讯作者
（13） Document Image Rectification in Complex Scene Using Stacked Siamese Networks, International Conference on Pattern Recognition, 2022, 第 3 作者
（14） One-Stage Open Set Object Detection with Prototype Learning, One-Stage Open Set Object Detection with Prototype Learning, International Conference on Neural Information Processing, 2021, 第 2 作者  通讯作者
（15） Learning-based Tensor Decomposition with Adaptive Rank Penalty for CNNs Compression, IEEE 4th International Conference on Multimedia Information Processing and Retrieval (MIPR), 2021, 第 2 作者
（16） One-Stage Open Set Object Detection with Prototype Learning, International Conference on Neural Information Processing, 2021, 第 2 作者  通讯作者
（17） Insect Recognition Under Natural Scenes Using R-FCN with Anchor Boxes Estimation, IMAGE AND GRAPHICS, ICIG 2019, PT I, 2019, 第 2 作者
（18） LightweightNet: Toward fast and lightweight convolutional neural networks via architecture distillation, PATTERN RECOGNITION, 2019, 第 2 作者
（19） Using High-Frequency Local Field Potentials From Multicortex to Decode Reaching and Grasping Movements in Monkey, IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, 2019, 第 5 作者
（20） A review on multi-task metric learning, BIG DATA ANALYTICS, 2018, 第 1 作者
（21） From Classification to Regression: Model Transfer for Visual Aesthetic Quality Assessment, PROCEEDINGS 2017 4TH IAPR ASIAN CONFERENCE ON PATTERN RECOGNITION (ACPR), 2017, 第 2 作者
（22） Margin-Aware Binarized Weight Networks for Image Classification, IMAGE AND GRAPHICS (ICIG 2017), PT I, 2017, 第 2 作者
（23） A Self-Paced Category-Aware Approach For Unsupervised Adaptation Networks, 2017 17TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), 2017, 第 2 作者
（24） GRMA: Generalized Range Move Algorithms for the Efficient Optimization of MRFs, INTERNATIONAL JOURNAL OF COMPUTER VISION, 2017, 第 3 作者
（25） A Self-Paced Category-Aware Approach For Unsupervised Adaptation Networks, 2017, 第 2 作者
（26） From Classification to Regression: Model Transfer for Visual Aesthetic Quality Assessment, 2017, 第 3 作者
（27） A Semi-Supervised Method for Surveillance-Based Visual Location Recognition., IEEE TRANSACTIONS ON CYBERNETICS, 2017, 第 2 作者
（28） FastLCD: Fast Label Coordinate Descent for the Efficient Optimization of 2D Label MRFs, INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2016, 第 2 作者
（29） ReD-SFA: Relation Discovery Based Slow Feature Analysis for Trajectory Clustering, 2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, 第 4 作者
（30） Adaptive Slice Representation for Human Action Classification, IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2015, 第 3 作者
（31） GRSA: Generalized Range Swap Algorithm for the Efficient Optimization of MRFs, 2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, 2015, 第 4 作者
（32） Cross-Domain Object Recognition Using Object Alignment, THE26THBRITISHMACHINEVISIONCONFERENCE, 2015, 第 3 作者
（33） GRSA: Generalized Range Swap Algorithm for the Efficient Optimization of MRFs, 2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2015, 第 3 作者
（34） Combination of Classification and Clustering Results with Label Propagation, IEEE SIGNAL PROCESSING LETTERS, 2014, 第 2 作者
（35） A real-time EMG pattern recognition method for virtual myoelectric hand control, NEUROCOMPUTING, 2014, 第 2 作者  通讯作者
（36） Robust Object Recognition via Visual Pathway Feedback, ICPR, 2014, 第 4 作者
（37） Geometry preserving multi-task metric learning, MACHINE LEARNING, 2013, 第 1 作者  通讯作者
（38） A multi-task framework for metric learning with common subspace, NEURAL COMPUTING & APPLICATIONS, 2013, 第 1 作者
（39） A Novel Feature Reduction Method for Real-Time EMG Pattern Recognition System, 2013 25TH CHINESE CONTROL AND DECISION CONFERENCE (CCDC), 2013, 第 1 作者
（40） Geometry Preserving Multi-task Metric Learning, MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, 2012, 第 3 作者
（41） Multi-task low-rank metric learning based on common subspace, LECTURE NOTES IN COMPUTER SCIENCE (INCLUDING SUBSERIES LECTURE NOTES IN ARTIFICIAL INTELLIGENCE AND LECTURE NOTES IN BIOINFORMATICS), 2011, 第 2 作者

   

（ 1 ） 连续多体问题智能求解工具, 参与, 中国科学院计划, 2025-02--2027-01
（ 2 ） 基于xx的大模型训练加速和推理压缩技术, 负责人, 国家任务, 2024-12--2026-05
（ 3 ） AI for Science重点领域研究案例与智能组件研发, 参与, 地方任务, 2023-11--2025-10
（ 4 ） 面向典型畜禽疫病的知识推理与诊断方法, 负责人, 国家任务, 2021-12--2024-11
（ 5 ） 场景感知的多语种文字识别技术, 负责人, 国家任务, 2020-11--2023-10
（ 6 ） 检疫昆虫的智能化识别算法与测试, 负责人, 境内委托项目, 2019-11--2022-11
（ 7 ） 短视频场景识别分析, 负责人, 境内委托项目, 2019-07--2019-12
（ 8 ） 入侵生物物种的智能化识别算法与测试, 负责人, 境内委托项目, 2018-04--2020-06

（1）机器学习视角下的数据驱动力学建模与实践   空天飞行高温气动全国重点实验室2025年度学术年会   2026-01-21

已指导学生

朱彦霖  硕士研究生  081104-模式识别与智能系统  

现指导学生

朱涵  硕士研究生  081104-模式识别与智能系统  

徐世龙  硕士研究生  085410-人工智能