具身智能，世界模型，多模态大模型

2016-09--2022-06   中国科学院计算技术研究所   工学博士
2012-09--2016-06   西北工业大学   工学学士

博士

工学博士

2024-10~现在, 中国科学院计算技术研究所, 副研究员
2022-09~2024-10,中国科学院计算技术研究所, 博士后

   

（1） Humanpcr: Probing mllm capabilities in diverse human-centric scenes,  International Conference on Learning Representations (ICLR, CCF-A), 2026, 第 4 作者
（2） Revisiting Multimodal Positional Encoding in Vision-Language Models, International Conference on Learning Representations (ICLR, CCF-A), 2026, 第 4 作者  通讯作者
（3） MATS: An Audio Language Model under Text-only Supervision, International Conference on Machine Learning (ICML,CCF-A), 2025, 第 2 作者  通讯作者
（4） UniPose: A Unified Multimodal Framework for Human Pose Comprehension, Generation and Edit, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR, CCF-A), 2025, 第 2 作者  通讯作者
（5） HIS-GPT: Towards 3D Human-In-Scene Multimodal Understanding, International Conference on Computer Vision (ICCV, CCF-A) , 2025, 第 2 作者  通讯作者
（6） Morph: A Motion-free Physics Optimization Framework for Human Motion Generation, International Conference on Computer Vision (ICCV, CCF-A)), 2025, 第 3 作者  通讯作者
（7） Revisiting logit distributions for reliable out-of-distribution detection, Advances in Neural Information Processing Systems (NeurIPS, CCF-A), 2025, 第 2 作者  通讯作者
（8） Knowmol: Advancing molecular large language models with multi-level chemical knowledge, Advances in Neural Information Processing Systems (NeurIPS, CCF-A), 2025, 第 3 作者
（9） unCLIP: Improving CLIP's Visual Detail Capturing Ability via Inverting unCLIP, Advances in Neural Information Processing Systems (NeurIPS, CCF-A), 2025, 第 4 作者
（10） Clothes-changing person re-identification with feasibility-aware intermediary matching, IEEE Transactions on Multimedia (TMM, CCF-A), 2025, 第 2 作者  通讯作者
（11） UMFC: Unsupervised Multi-Domain Feature Calibration for Vision-Language Models, Neural Information Processing Systems (NeurIPS, CCF-A), 2024, 第 2 作者  通讯作者
（12） MGPT: An Advanced Multimodal, Multitask Framework for Motion Comprehension and Generation, Neural Information Processing Systems (NeurIPS, CCF-A), 2024, 第 2 作者  通讯作者
（13） Triplet adaptation framework for robust semi-supervised learning, IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI, CCF-A)), 2024, 第 1 作者
（14） 基于时序多尺度互补特征的视频行人重识别, Temporal Multi-Scale Complementary Feature for Video Person Re-Identification, 计算机学报, 2023, 第 1 作者
（15） Dual Compensation Residual Networks for Class Imbalanced Learning, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 第 1 作者
（16） Generalized Semi-Supervised Learning via Self-Supervised Feature Adaptation, NeurIPS, 2023, 第 2 作者
（17） Feature Completion for Occluded Person Re-Identification, IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 第 1 作者
（18） Bicnet-tks: Learning efficient spatial-temporal representation for video person re-identification, Proceedings of thComputer Vision and Pattern Recognition (CVPR, CCF-A), 2021, 第 1 作者
（19） VRSTC: Occlusion-Free Video Person Re-Identification, 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, 第 1 作者  通讯作者
（20） Interaction-and-Aggregation Network for Person Re-identification, 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, 第 1 作者  通讯作者
（21） Video Prediction with Bidirectional Constraint Network, 2019 14TH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE AND GESTURE RECOGNITION (FG 2019), 2019, 第 1 作者  通讯作者
（22） Cross Attention Network for Few-shot Classification, ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019, 第 1 作者  通讯作者