团队力量

副高(副教授等)

当前位置: 首页 -> 团队力量 -> 全体教工 -> 正文

张小明

发布日期:2019-06-16

张小明 Xiaoming Zhang

  

副教授

       

博士

       

010-82338247

办公室  

 

       

太阳成官网

       

 

电子信箱

yolixs@buaa.edu.cn

       

 

个人主页

 

  学习经历

  1999/092003/07 国防科技大学计算机科学与技术学士学位  

  2004/092007/01 国防科技大学计算机应用技术硕士学位  

  2007/092012/03 太阳成官网计算机应用技术博士学位  

  2009/092010/09 澳大利亚昆士兰大学联合培养博士  

  工作经历

  2012/07-2017/01,太阳成官网,计算机学院,讲师

  2017/01-至今太阳成官网,计算机学院,助理教授  

  2017/06-2019/05太阳成官网,计算机学院,博士生导师

  2019/05-至今太阳成官网,太阳成官网,副教授、博士生导师.

  研究领域

大模型安全、多模态大模型后门攻击检测与修复、RAG、多模态数据深度融合学习、智能渗透测试、多模态敏感信息检测。

  荣誉及奖励

  2012年,太阳成官网“蓝天新秀”;

▪  2025年,北京市科技进步二等奖;

▪  2025年,中国商业联合会科技进步一等奖

 

  开授课程

网络内容安全、机器学习  

  教学及科研成果

长期从事机器学习、文本数据挖掘等领域的研究,已在著名国际学术会议ICMLAAAIACLMMCIKMIJCAISDM等、以及著名国际期刊TKDETIPTOISTMMTCYBPRIF等发表70余篇学术论文

已承担的科研项目包括:国家科技重大专项课题、国家自然科学基金重点项目、面上项目、北京市自然基金面上项目、装备预研、国家242信息安全专项等20余项课题。项目主要研究基于多智能体协同的夸模态敏感信息检测、特定领域知识图谱、垂直领域大模型训练、多模态大大模型威胁定位于实时修复等。研究成果获北京市科技进步二等奖1项、中国商业联合会科技进步一等奖2项,并在多个部门得到广泛应用。

近年主要论文:

(1) Ziyi Zhou, Xioaming Zhang, Hui Pang, et al., Reasoning through Evolution: Automatic Meta-path Discovery for LLM-based Fake News Detection. In proceedings of ACM MM 2026.

(2) Dezhuang Miao, Xiaoming Zhang, Bo Zhang, Yibin Du, Xiang Li, Ruilin Zeng, Yirui Qi. LLM-MatLogic: Executable Exchange Contracts for Knowledge-Graph Query Answering with Scoped Negation. In Proceedings of ICML 2026.

(3) Bingyu Yan, Xiaoming Zhang, Jinyu Hou, Chaozhuo Li, Ziyi Zhou,Yiming Hei, Litian Zhang. Evo-Attacker: Memory-Augmented Reinforcement Learning for Long-Horizon Tool Attacks on LLM-MAS. In proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026).

(4) Yirui Qi1*, Xiaoming Zhang1†, Ruilin Zeng1, Mengyao Liu1,Ziyi Zhou1, Dezhuang Miao1, Bingyu Yan1, Zhenyu Guan. Beyond Static Persona Consistency: Dynamic Persona Coherence in LLM Role-Playing. In proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026).

(5) Bingyu Yan, Xiaoming Zhang, Ziyi Zhou, Chaozhuo Li, Ruilin Zeng, Yirui Qi, Tianbo Wang, Litian Zhang. Attack the Messages, Not the Agents: A Multi-round Adaptive Stealthy Tampering Framework for LLM-MAS. In proceedings of AAAI 2026.

(6) Z  Zhou, X  Zhang, L  Zhang, et al., FineFake: A knowledge-enriched dataset for fine-grained multi-domain fake news detection, Information Fusion, 132, 2026, 104253.

(7) X Li, X Zhang, D Miao, X Cheng, D Li, H Han, Z Li, Mixture of disentangled experts with missing modalities for robust multimodal sentiment analysis, Information Fusion, Volume 135, 2026, 104495, .

(8) Litian Zhang, Xiaoming Zhang, Ziyi Zhou, Xi Zhang, Senzhang Wang, Philip S. Yu, Chaozhuo Li. Early Detection of Multimodal Fake News via Reinforced Propagation Path Generation. IEEE Trans. Knowl. Data Eng. 37(2): 613-625 (2025)

(9) Litian Zhang, Xiaoming Zhang, Ziyi Zhou, Xi Zhang, Philip S. Yu, Chaozhuo Li: Knowledge-aware multimodal pre-training for fake news detection. Inf. Fusion 114: 102715 (2025)

(10) Ziyi Zhou, Xiaoming Zhang, Shenghan Tan, Litian Zhang, Chaozhuo Li. Collaborative Evolution:Multi-Round Learning between Largeand Small Language Models for Emergent Fake News Detection. AAAI 2025.

(11)  Xiang Li, Haijun Zhang, Zhiqiang Dong, Xianfu Cheng, Yun Liu, Xiaoming Zhang: Learning fine-grained representation with token-level alignment for multimodal sentiment analysis. Expert Syst. Appl. 269: 126274 (2025)

(12) Litian Zhang, Xiaoming Zhang, Ziyi Zhou, Xi Zhang, Philip S. Yu, Chaozhuo Li: Knowledge-aware multimodal pre-training for fake news detection. Inf. Fusion 114: 102715 (2025)

(13) Bo Zhang, Xiaoming Zhang, Zhibo Zhou, Yun Liu, Yancong Li, Feiran Huang: Moment matching of joint distributions for unsupervised domain adaptation. Inf. Process. Manag. 62(1): 103944 (2025)

(14) Litian Zhang, Xiaoming Zhang, Ziyi Zhou, Xi Zhang, Senzhang Wang, Philip S. Yu, Chaozhuo Li: Early Detection of Multimodal Fake News via Reinforced Propagation Path Generation. IEEE Trans. Knowl. Data Eng. 37(2): 613-625 (2025)

(15) Litian Zhang and Xiaoming Zhang and Chaozhuo Li and Ziyi Zhou and Jiacheng Liu and Feiran Huang and Xi Zhang, Mitigating Social Hazards: Early Detection of Fake News via Diffusion-Guided Propagation Path Generation, Proceedings of the 32nd ACM International Conference on Multimedia, 2024.

(16) Yancong Li, Xiaoming Zhang, Ying Cui, Shuai Ma.Hyperbolic Graph Neural Network for Temporal Knowledge Graph Completion. In proceedings of COLING2024.

(17) Zhipeng Liu,Xiaoming Zhang, Litian Zhang, Zelong Yu. MDS:A Fine-Grained Dataset for Multi-Modal Dialogue Summarization.In proceedings of COLING2024.

(18) Litian Zhang, Xiaoming Zhang, Linfeng Han, Ziyi Zhou, Feiran Huang, Chaozhuo Li. Reinforced Adaptive Knowledge Learning for Multiodal Fake News Detection. In proceedings of AAAI2024.

(19) Bo Zhang, Xiaoming Zhang*, Feiran Huang, Ming Lu, Shuai Ma. Deep Kernel Network Embedding. IEEE Transactions on Knowledge and Data Engineering, 2023.

(20) Litian Zhang, Xiaoming Zhang, Ziming Guo, Zhipeng Liu. CISum: Learning Cross‑modality Interaction to Enhance Multimodal Semantic Coverage for Multimodal Summarization. In Proceedings of the 2023 SIAM International Conference on Data Mining (SDM), 2023.

(21) Litian zhang, Xiaoming Zhang*, Junshu Pan, Feiran Huang. Hierarchical Cross-Modality Semantic Correlation Learning Model for Multimodal Summarization. In proc of AAAI2022.

(22) Yancong Li, Xiaoming Zhang, Bo Zhang, Haiying Ren. Each Snapshot to Each Space: Space Adaptation for Temporal Knowledge Graph Completion[C]. In Proceedings of International Semantic Web Conference (ISWC, candidate best paper), 2022.

(23) Yun Liu, Xiaoming Zhang, Feiran Huang, Bo Zhang and Zhoujun. Li. Cross-Attentional Spatio-Temporal Semantic Graph Networks for Video Question Answering. IEEE Transactions on Image Processing, vol. 31, pp. 1684-1696, 2022.

(24) Bo Zhang, Xiaoming Zhang*,Yun Liu, Lei Cheng, Zhoujun Li. Matching Distributions between Model and Data: Cross-domain Knowledge Distillation for Unsupervised Domain Adaptation. In Proceeding of the 59th Annual Meeting of the Association for Computational Linguistics (ACL), 2021.

(25) Liu Y, Zhang X , Zhang Q , et al. Dual Self-Attention with Co-Attention Networks for Visual Question Answering[J]. Pattern Recognition, 2021(2):107956.

(26) Feiran Huang, Xiaoming Zhang*, Jie Xu, Zhonghua Zhao, Zhoujun Li: Multi-modal Learning of Social Image Representation By Exploiting Social Relations[J]. IEEE Trans. Cybernetics, 2021.

(27) Bo Zhang, Xiaoming Zhang*, Lei Cheng. Discriminative Feature Adaptation via Conditional Mean Discrepancy for Cross-domain Text Classification. In proceedings of DASFAA2021.

(28) Yun LiuXiaoming Zhang*, et al. Visual Question Answering via Combining Inferential Attention and Semantic Space Mapping. 10.1016/j.knosys.2020.106339, Knowledge-Based Systems, 1-12.

(29) Yun LiuXiaoming Zhang*, et al. ALSA: Adversarial Learning of Supervised Attentions for Visual Question Answering. 10.1109/TCYB.2020.3029423,IEEE TRANSACTIONS ON CYBERNETICS, 1-14.

(30) Yun Liu, Xiaoming Zhang*, Feiran Huang, Lei Cheng, Zhoujun Li. Adversarial Learning With Multi-Modal Attention for Visual Question Answering. IEEE Transactions on Neural Networks and Learning Systems. 2020. PP. 1-15. 10.1109/TNNLS.2020.3016083.

(31) Bo Zhang, Xiaoming Zhang*. Hierarchy Response Learning for Neural Conversation Generation. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP), 1772--1781.

 

  学术与社会服务

中关村实验室双聘专家、中国计算机学会会员,中国人工智能学会智能服务专委会委员,网络内容安全专委会委员,国际会议ECMLSDMWISEPC member,国际期刊Applied science等的客座编辑

  科研项目

 

Curriculum Vitae  (CV)

With long-term research experience in machine learning and text data mining, he has published over 70 academic papers in renowned international conferences such as ICML, AAAI, ACL, MM, CIKM, IJCAI, SDM, and top-tier journals including TKDE, TIP, TOIS, TMM, TCYB, PR, and IF. The undertaken research projects include over 20 initiatives such as National Key Science and Technology Major Projects, National Natural Science Foundation Key Projects, General Projects, Beijing Natural Science Foundation General Projects, Equipment Preliminary Research, and the National 242 Information Security Special Program. The projects primarily focus on multimodal sensitive information detection based on multi-agent collaboration, domain-specific knowledge graphs, vertical domain large model training, multimodal large model threat localization and real-time remediation, among others. The research achievements have earned one Beijing Science and Technology Progress Second Prize and two China Council for the Promotion of International Trade Science and Technology Progress First Prizes, and have been widely applied across multiple sectors.

 

上一条:王钢

下一条:姚燕青