可信机器学习、大语言模型可解释性、算法博弈

希望学生自驱力强、多问多思

招生专业：

0812-计算机科学与技术

范鑫鑫，中国科学院计算技术研究所副研究员，研究方向涵盖：可信机器学习、LLM可解释性、算法博弈。主持国家自然基金、中科院先导课题等项目5项，经费总计900余万。在学术方面，在USENIX Security、ICML、SIGKDD、WWW、CIKM、ACM CSUR (2024 IF=28, Ranked 1/147 in Computer ScienceTheory & Methods)、IEEE TPDS、IEEE TMC、IEEE TSC、IEEE TCE、ACM TWEB、The Innovation(2023 IF=33.2)、计算机研究与发展等一系列计算机学科最具影响力期刊 (ACM CSUR)和计算机安全领域四大顶会 (USENIX Security) 上发表论文60余篇。其中，1篇获得了最佳论文奖/Microsoft Award-Honorable Mention (SCC'18)，2篇获得最佳论文奖提名 (IPCCC'17, KSEM'18)。系统方面，研制了“智能算法安全分析平台”、“网络数据分析挖掘系统”，已部署应用于多个单位。

教育经历：

 2011-09~2013-03   Georgia Institute of Technology   博士联培

 2009-03~2014-04   大连理工大学       工学博士 

 2006-09~2009-01   大连理工大学       工学硕士 

 2002-09~2006-06   郑州轻工业大学   工学学士

工作经历：  

 2021-10~现在         中国科学院计算技术研究所 副研究员

 2017-04~2021-09   中国科学院计算技术研究所 助理研究员  

• 2019年 中科院计算所优秀研究人员
• 2013年 大连理工大学优秀博士学位论文特别奖学金
• 2011年 华为专项奖学金
• 2010年 大连理工大学学术新人奖
•  一种基于强化学习的对抗混淆流量生成方法, 2025
• 一种基于知识图谱的双轨验证多跳推理方法，2025
• 一种加密网络流量分类系统、系统构建方法及分类方法，2025
• 一种基于特征空间覆盖的神经网络测试套件多样性评估方法及装置，2025
• 对抗混淆流量生成方法、装置、存储介质、程序产品, 发明专利, 2025
• 基于特征匹配的I2P流量识别方法及系统, 发明授权, 2023, 专利号: CN112910797B

2026

Xinxin Fan, Wenxiong Chen, Quanliang Jing, Chi Lin, Shaoye Luo, Wenbo Song, Yunfeng Lu. TopFeaRe: Locating Critical State of Adversarial Resilience for Graphs Regarding Topology-Feature Entanglement. The 35th USENIX Security Symposium, Accepted to appear (Acceptance Rate: 20/163=12.3% in Primary Field of Security of ML). [USENIX Security'26]

Changhao Wang, Yanfang Liu, Xinxin Fan, Lanzhi Zhou, Ao Tian, Yunfeng Lu. DTKG: Dual-Track Knowledge Graph-Verified Reasoning Framework for Multi-Hop QA. Forty-third International Conference on Machine Learning, Accepted to appear (Acceptance Rate: 6352/23918=26.6%). [arXiv version] [ ICML 2026]

Ao Tian, Yunfeng Lu, Xinxin Fan, Changhao Wang, Lanzhi Zhou, Yeyao Zhang, Yanfang Liu. RGMem: Renormalization Group–inspired Memory Evolution for Language Agents. Forty-third International Conference on Machine Learning, Accepted to appear (Acceptance Rate: 6352/23918=26.6%) [arXiv version] [ ICML 2026]

Wenqi Wei, Tiansheng Huang, Sihao Hu, Xinxin Fan, Rui Zhang, Jingya Zhou, Ling Liu. GradCloak: Gradient Obfuscation for Privacy-Preserving Distributed Learning as a Service. IEEE Transactions on Services Computing, 19(2): 1277-1290, 2026. [TSC 2026]

Wei Yang, Chi Lin, Jing Deng, Haipeng Dai, Liming Chen, Xinxin Fan. Secure Charging Scheduling in Wireless Rechargeable Sensor. IEEE Transactions on Mobile Computing, 25(2): 1870-1884, 2026. [TMC 2026]

2025

Jincai Huang*, Yongjun Xu*, Qi Wang*, Qi (Cheems) Wang*, Xingxing Liang*, Fei Wang*, Zhao Zhang*, Wei Wei*, Boxuan Zhang*, Libo Huang*, Jingru Chang*, Liantao Ma*, Ting Ma*, Yuxuan Liang*, Jie Zhang*, Jian Guo*, Xuhui Jiang*, Xinxin Fan*, Zhulin An*, Tingting Li* (*Co-First Author), et al. Foundation models and intelligent decision-making: Progress, challenges, and perspectives. The Innovation, 6(6): 2666-6758, 100948, 2025 (Cover Paper). [The Innovation 2025]

Shaoye Luo, Xinxin Fan, Quanliang Jing, Men Niu, Chi Lin, Yunfeng Lu. SifterNet: Model-Agnostic Defense against Backdoor Attack in Vision Large Model. 12th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation, BuildSys, 389-393, 2025. [BuildSys'25]

Mengfan Li, Xinxin Fan, Quanliang Jing, Shaoye Luo, Yunfeng Lu, Jingping Bi. LDP: Latent Diffusion-based Adversarial Purification towards Transformer-based Visual Encoders. 24th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, 532-539, 2025. [TrustCom'25]

Qining Sun, Zhiming Gui, Yanfang Liu, Xinxin Fan, Yunfeng Lu. Graph Neural Network-based Multi-agent Reinforcement Learning for Adversarial Policy Detection Algorithm (In Chinese). Computer and Modernization, 0(4): 42-49, 2025. [ C&M 2025]

2024

Xinxin Fan, Mengfan Li, Jia Zhou, Quanliang Jing, Chi Lin, Yunfeng Lu, Jingping Bi. GCSA: A New Adversarial Example-Generating Scheme Toward Black-Box Adversarial Attacks. IEEE Trans. Consumer Electron. 70(1): 2038-2048, 2024. [TCE 2024]

Chi Lin, Zhouhe Sun, Asfandeyar Ahmad, Xinxin Fan, Yi Wang, Lei Wang, Xin Fan, Guowei Wu. AirWrite: An Aerial Handwriting Trajectory Tracking and Recognition System With mmWave. IEEE Trans. Mob. Comput. 23(12): 13325-13341, 2024. [TMC 2024]

Yubo Zhang, Yanfang Liu, Xinxin Fan, Yunfeng Lu. CPLCS: Contrastive Prompt Learning-based Code Search with Cross-modal Interaction Mechanism. 2024 International Joint Conference on Neural Networks. [IJCNN 2024]

2023

Yunfeng Lu, Xinxin Fan, Quanliang Jing. TAEffect: Quantifying interaction risks in trust-enabled communication systems. Int. J. Commun. Syst. 36(4), 2023. [IJCS 2023]

2022

Quanliang Jing, Shuo Liu, Xinxin Fan, Jingwei Li, Di Yao, Baoli Wang, Jingping Bi. Can Adversarial Training benefit Trajectory Representation?: An Investigation on Robustness for Trajectory Similarity Computation. Proceedings of the 31st ACM International Conference on Information & Knowledge Manageme, 2022. [CIKM 2022]

Xiaokai Chu, Jiashu Zhao, Xinxin Fan, Di Yao, Zhihua Zhu, Lixin Zou, Dawei Yin, Jingping Bi. Contrastive Disentangled Graph Convolutional Network for Weakly-Supervised Classification. Database Systems for Advanced Applications, 2022. [DASFAA 2022 ]

Baoli Wang, Xinxin Fan, Quanliang Jing, Yueyang Su, Jingwei Li, Jingping Bi. GTAT: Adversarial Training with Generated Triplets. 2022 International Joint Conference on Neural Networks. [IJCNN 2022 ]

Yubo Zhang, Yanfang Liu, Xinxin Fan, Yunfeng Lu. RetCom: Information Retrieval-Enhanced Automatic Source-Code Summarization. IEEE 22nd International Conference on Software Quality, Reliability and Security . [QRS 2022 ]

2021

Quanliang Jing, Yao Di, Chang Gong, Xinxin Fan, Baoli Wang, Haining Tan, and Jingping Bi. TRAJCROSS: Trajecotry Cross-Modal Retrieval with Mulit-level Contrastive Learning. 2021 IEEE International Conference on Big Data, 2021. [BigData 21]

Xiaokai Chu, Xinxin Fan, Jingping Bi. Position-Aware Network Representation Learning via K-Step Mutual Information Estimation (In Chinese). Journal of Computer Research and Development (计算机研究与发展), 58 (8): 1612-1623, 2021. [JCRD 21]

Xiaokai Chu, Xinxin Fan, Zhihua Zhu and Jingping Bi. Variational Cross-Network Embedding for Anonymized User Identity Linkage. The 30th International Conference on Information and Knowledge Management, Pages 2955–2959, 2021. (Short Papers Track, Acceptance Rate: 177/626=28%). [CIKM 21]

Baoli Wang, Xinxin Fan, Quanliang Jing, Haining Tan, Jingping Bi: AdvCGAN: An Elastic and Covert Adversarial Examples Generating Framework. IEEE International Joint Conference on Neural Network, 1-8, 2021. [IJCNN 21]

Quanliang Jing, Di Yao, Xinxin Fan, Baoli Wang, Haining Tan, Xiangpeng Bu, Jingping Bi. TRANSFAKE: Multi-task Transformer for Multimodal Enhanced Fake News Detection. IEEE International Joint Conference on Neural Network, 1-8, 2021. [IJCNN 21]

2020

Xinxin Fan, Ling Liu, Rui Zhang, Quanliang Jing, Jingping Bi. Decentralized Trust Management: Risk Analysis and Trust Aggregation. ACM Computing Surveys, 53(1): 2:1-2:33, 2020. [CSUR 20]

Zhihua Zhu, Xinxin Fan, Xiaokai Chu, Jingping Bi. HGCN: A Heterogeneous Graph Convolutional Network-Based Deep Learning Model Toward Collective Classification. The 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 1161-1171, 2020. (Research Track, Acceptance Rate: 216/1279=17%).[KDD 20]

Zhihua Zhu, Xinxin Fan, Xiaokai Chu, Jianhui Huang, Jingping Bi. LRHNE: A Latent-Relation Enhanced Embedding Method for Heterogeneous Information Networks. The 29th International Conference on Information and Knowledge Management, 1923-1932, 2020. (Full Paper Research Track, Acceptance Rate: 193/920=21%). [CIKM 20]

2019

Xiaokai Chu, Xinxin Fan, Di Yao, Zhihua Zhu, Jianhui Huang, Jingping Bi. "Cross-Network Embedding for Multi-Network Alignment", The 28th International World Wide Web Conferences, San Francisco, CA, USA, May 13-17, 2019. [WWW 19]

Xiaokai Chu, Xinxin Fan, Di Yao, Chen-Lin Zhang, Jianhui Huang, Jingping Bi. "Noise-Aware Network Embedding for Multiplex Network", The International Joint Conference on Neural Networks, Budapest, Hungary, July 14-19, 2019. [IJCNN 19]

2018

Xinxin Fan, Jingping Bi. "Identification for Strategically Malicious Participants", The 25th International Conference on Web Service, Seattle, WA, USA, June 25-30, 2018. [ICWS 18]

Xinxin Fan, Danyang He, Jingping Bi. "Trustworthiness and Untrustworthiness Inference with Group Assignment", The 25th International Conference on Web Service, Seattle, WA, USA, June 25-30, 2018. [ICWS 18]

Zixuan Ding, Xinxin Fan, Jinping Yu, Jingping Bi. "Update Cost-Aware Cache Replacement for Wildcard Rules in Software-Defined Networking", The 23th IEEE Symposium on Computer and Communications, Natal, Brazil, June 25-28, 2018. (Microsoft Award-Honorable Mention/Best Paper Award) [ ISCC 18]

Haining Tan, Weiqiang Tang, Xinxin Fan, Quanliang Jing, Jingping Bi. "SERL: Semantic-Path Biased Representation Learning of Heterogeneous Information Network", 11th International Conference on Knowledge Science, Engineering and Management, Changchun, China, August 17-19, 2018. (Best Paper Runner Up) [KSEM 18]

Jinping Yu, Jingping Bi, Guoqiang Zhang, Xinxin Fan, Zhongcheng Li. "Loop-Free Rerouting of Multiple Policies With Overlaps", The 36th IEEE International Performance Computing and Communications Conference, San Diego, CA, USA, December 10-12, 2017. [IEEE Communications Letters 18]

2017

Xinxin Fan, Ling Liu, Mingchu Li, Zhiyuan Su. "GroupTrust: Dependable Trust Management", IEEE Transactions on Parallel and Distributed Systems, 28(4): 1076-1090, 2017. [TPDS 17]

Jinping Yu, Xinxin Fan, Guoqiang Zhang, Jingping Bi. "Scheduling Loop-Free Updates for Multiple Policies with Overlaps in Software-Defined Networks", The 36th IEEE International Performance Computing and Communications Conference, San Diego, CA, USA, December 10-12, 2017. (Best Paper Runner Up) [IPCCC 17]

Before 2017

Zhiyuan Su, Ling Liu, Mingchu Li, Xinxin Fan, Yang Zhou. "Reliable and Resilient Trust Management in Distributed Service Provision Networks", ACM Transactions on the Web, 9(3): 14:1-14:37, 2015. [TWEB 15]

Xinxin Fan, Mingchu Li, Hui Zhao, Xiaodong Chen, Zhenzhou Guo, Dong Jiao, Weifeng Sun. "PeerCluster: A Maximum Flow-Based Trust Mechanism in P2P File Sharing Networks", Security and Communication Networks 6(9): 1126-1142, 2013. [SCN (Wiley) 13]

Xiaodong Chen, Mingchu Li, Xin Ma, Xinxin Fan. "Hamiltonian Connectedness in Claw-Free Graphs", Graphs and Combinatorics 29(5): 1259-1267, 2013. [Graphs & Com. (Springer) 13]

Zhenzhou Guo, Mingchu Li, Xinxin Fan. "Attribute-Based Ring Signcryption Scheme", Security and Communication Networks, 6(6): 790-796, 2013. [SCN (Wiley) 13]

Zhiyuan Su, Ling Liu, Mingchu Li, Xinxin Fan, Yang Zhou. "ServiceTrust: Trust Management in Service Provision Networks", The IEEE International Conference on Services Computing, Santa Clara, CA, USA, June 28-July 3, 2013. [SCC 13]

Xinxin Fan, Mingchu Li, Jianhua Ma, Yizhi Ren, Hui Zhao, Zhiyuan Su. "Behavior-based Reputation Management in P2P File-Sharing Networks", Journal of Computer and System Sciences, 78(6): 1737-1750, 2012. [JCSS (Elsevier) 12]

Xinxin Fan, Ling Liu, Mingchu Li, Zhiyuan Su. "EigenTrust^++: Attack Resilient Trust Management", Proceedings of 8th IEEE International Conference on Collaborative Computing, October 14-17, Pittsburgh, Pennsylvania, United States, 2012. [CollaborateCom 12]

Hui Zhao, Mingchu Li, Xinxin Fan. "A Formal Separation Method of Protocols to Eliminate Parallel Attacks in Virtual Organization", Security and Communication Networks, 4(12): 1461-1468, 2011. [SCN (Wiley) 11]

Xinxin Fan, Mingchu Li, Yizhi Ren, Jianhua Ma. "Dual-EigenRep: A Reputation-Based Trust Model for P2P File-Sharing Networks", UIC/ATC Workshops, Xi'an, China, 2010. [UIC/ATC Workshops 10]

Xianchao Zhang, Xinxin Fan, Xinyue Liu, Hong Yu. "A Ranking Algorithm via Changing Markov Probability Matrix Based on Distribution Factor", Proceedings of The Fifth International Conference on Fuzzy Systems and Knowledge Discovery, October 18-20, Jinan, Shandong, China, 2008. [FSKD 08]

• 人机协同的算法博弈, 负责人, 中国科学院计划（先导）, 2023-12--2028-11
• 智能算法隐私窃取及对抗攻击研究, 负责人, 研究所自主部署（所创新课题）, 2022-05--2023-04
• 边缘计算弹性网络架构及协同交互安全可信机制研究, 负责人, 国家任务（国家自然科学基金）, 2018-01--2020-12

Peer Review:

--Journal

• ACM Transaction on Web (TWeb)
• ACM Transactions on Internet Technology (ToIT)
• IEEE Transactions on Services Computing (TSC)
• IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
• SCIENCE CHINA Information Sciences (中国科学.信息科学)
• Security and Communication Networks (John Wiley)
• Journal of Computer and System Sciences (Elsevier)
• Computer Networks (Elsevier)
• The Innovation (Cell Press Partner Journal)
• Scientific Reports (Nature Portfolio Journal)

--Conference

• ICML'26 (Gold Reviewer)
• NeurIPS'26
• AIoTSys'25

现指导学生

渠雨枫  硕士研究生  085400-电子信息  

2026级：丁子瑞 (MS)

2025级：渠雨枫 (MS) 

2024级：钮  门   (MS, co-supervided with Prof. Jingping Bi)  

2024级：骆绍业 (MS, co-supervided with Prof. Jingping Bi)   

2022级：李梦帆 (PhD, co-supervided with Prof. Jingping Bi) 

2022级：陈闻雄 (Visiting MS, co-supervided with Prof. Chi Lin) 

2022级：宋文博 (Visiting MS, co-supervided with Prof. Chi Lin)   毕业去向：宁德时代

2018级：何丹阳 （BS, co-supervided with Prof. Chi Lin）          毕业去向：Columbia Univ. 攻读硕士

2016级：储晓恺 （PhD, co-supervided with Prof. Jingping Bi） 毕业去向：Tencent, China 

2015级：唐威强 （MS, co-supervided with Prof. Jingping Bi）  毕业去向：VMware

2015级：仇文凯   (MS, co-supervided with Prof. Jingping Bi）  毕业去向：Alibaba

2015级：王保利 （PhD, co-supervided with Prof. Jingping Bi） 毕业去向：Microsoft Research Asia (MSRA)

2015级：朱志华 （PhD, co-supervided with Prof. Jingping Bi） 毕业去向：Tencent, China