个人简介
聂何望,男,汉族,籍贯江西省九江市,博士/讲师。2025年6月毕业于华中科技大学,获网络与信息安全博士学位。长期致力于人工智能安全、可逆信息隐藏、神经网络模型水印等领域的研究。
主持中央网信办创新资助计划项目,参与国家重点研发计划、国家科学自然基金地区科学基金项目、广西高校中青年教师基础能力提升项目等重要科研项目研究。已在国内外著名期刊和会议发表论文十余篇,其中SCI一区TOP国际期刊论文5篇、CCF-B类国际期刊/会议论文6篇,获授权发明专利3件,软件著作权3件。
学术任职
- 中国图象图形学学会数字媒体取证与安全专委会委员
- “新一代图像安全技术研讨会” 程序委员会委员 (PC Member)
期刊审稿人
- IEEE Transactions on Artificial Intelligence
- IEEE Access
- Knowledge-Based Systems
- Expert Systems with Applications
- Engineering Applications of Artificial Intelligence
- Information Processing & Management
- Information Fusion
- Information Sciences
- Applied Soft Computing
- Neural Networks
- Neurocomputing
- Signal Processing
- Ad Hoc Networks
- Cluster Computing
- Complex & Intelligent Systems
- EURASIP Journal on Image and Video Processing
- IET Image Processing
- PLOS ONE
- Journal of King Saud University -- Computer and Information Sciences
- Smart Health
- Computer Law & Security Review
- International Journal of Computational Intelligence Systems
会议审稿人
科研项目
- 中央网信办,网络安全学院学生创新资助计划,AI数据模型防泄露检测工具,2024.04-2025.04,5万元,结题,主持
- 国家重点研发计划 (2021YFB2012200),开放式数控系统安全可信技术,2021.11-2024.10,2297万元,结题,参与
- 广西教育厅 (2021KY0861),基于注意力卷积神经网络的目标检测算法研究,2021.01-2025.04,2万元,结题,参与
- 国家自然科学基金 (61962008),基于矩阵分解的彩色图像哈希算法研究,2020.01-2023.12,40万元,结题,参与
论文发表
† 通讯作者 | ∗ 共同一作
期刊论文 (按年份)
-
1.
Compression Is No Barrier: Dataset Copyright Protection with Compression-Resistant Backdoor Watermarks
-
2.
Beyond Protection: Unveiling Neural Network Copyright Trading
-
3.
Secure Industrial Federated Learning: Label Encryption for Model Protection
-
4.
Federated Learning with Bilateral Defense via Blockchain
-
5.
Deep Model Intellectual Property Protection with Compression-Resistant Model Watermarking
Nie Hewang, Songfeng Lu, Junjun Wu, Jianxin Zhu.
IEEE Transactions on Artificial Intelligence, 2024.
-
6.
PersistVerifty: Federated model ownership verification with spatial attention and boundary sampling
-
7.
FedCRMW: Federated model ownership verification with compression-resistant model watermarking
-
8.
Securing IP in Edge AI: Neural Network Watermarking for Multimodal Models
-
9.
Color image reversible data hiding with double-layer embedding
会议论文 (按年份)
-
1.
Optimized Dynamic Watermarking for Audio DNNs with Adaptive Embedding and Boundary Sampling
-
2.
FedDiT: Federated Learning by Distillation Token Enhanced Vision Transformer
-
3.
VeriChroma: Ownership Verification for Federated Models via RGB Filters
-
4.
Lightweight Byzantine-Robust and Privacy-Preserving Federated Learning
发明专利与软件著作权
- 路松峰, 路直, 聂何望, 杨豪. 一种基于国产密码的工控数据安全防护系统及其工作方法. CN118133298A (2024).
- 路松峰, 周立天, 朱建新, 罗勇, 聂何望. 一种基于LSTM的数控系统日志审计方法及终端. CN116781321A (2023).
- 路松峰, 肖珏, 路直, 聂何望, 杨豪. 一种适用于数控系统的流加密机及其工作方法. CN116684076A (2023).
- 聂何望, 唐振军, 梁晓萍, 广西师范大学. 抗压缩的联邦学习模型水印系统V1.0,登记号:2025R11L2606607
- 聂何望, 唐振军, 陈旅, 广西师范大学. 基于边界采样的音频深度神经网络模型水印系统V1.0,登记号:2025SR2186923
- 聂何望, 唐振军, 凌曼, 广西师范大学. 基于SVM的车牌识别软件V1.0,登记号:2018SR219059
荣誉奖项
- 优秀毕业博士研究生 (2025)
- 博士研究生国家奖学金 (2023-2024)
教学与指导
指导本科生获奖:
- 高锐 (2019级):全国蓝桥杯大赛广西赛区三等奖
- 韩林荣 (2019级):全国蓝桥杯大赛广西赛区二等奖
招生启事
欢迎对人工智能安全、联邦学习与模型水印等方向感兴趣的同学报考与加入课题组!同时欢迎踏实勤奋、积极上进的本科生加入。
- 研究方向: 模型版权保护、联邦学习所有权验证、数据/模型水印、AI安全评测等
- 期望背景: 具备良好的编程基础(Python/PyTorch优先),对科研有热情与自驱力
- 联系方式: 请附个人简历、成绩单及代表性成果发送至:nhw@gxnu.edu.cn
About
Hewang Nie is a Lecturer with a Ph.D. in Cyberspace Security from Huazhong University of Science and Technology (June 2025). His research focuses on AI security, reversible information hiding, and neural network watermarking.
He has led a student innovation funding project from the Cyberspace Administration of China and participated in major research programs including the National Key R&D Program of China and the National Natural Science Foundation of China. He has published over ten papers in well-known journals and conferences, including five papers in SCI Q1 (TOP) journals and six papers in CCF-B journals/conferences. He holds 3 authorized invention patents and 3 software copyrights.
Research Interests
- AI Security; Intellectual Property Protection for Neural Network Models
- Multimedia Information Security; Neural Network Watermarking; Reversible Information Hiding; Image Watermarking
- Federated Learning Security; Model Ownership Verification
- Artificial Intelligence in Healthcare; Medical Image Processing
Academic Service
- Member of the Digital Media Forensics and Security Committee, CSIG.
- PC Member for Workshop on New-Generation Image Security Technologies.
Journal Reviewer
- IEEE Transactions on Artificial Intelligence
- IEEE Access
- Knowledge-Based Systems
- Expert Systems with Applications
- Engineering Applications of Artificial Intelligence
- Information Processing & Management
- Information Fusion
- Information Sciences
- Applied Soft Computing
- Neural Networks
- Neurocomputing
- Signal Processing
- Ad Hoc Networks
- Cluster Computing
- Complex & Intelligent Systems
- EURASIP Journal on Image and Video Processing
- IET Image Processing
- PLOS ONE
- Journal of King Saud University -- Computer and Information Sciences
- Smart Health
- Computer Law & Security Review
- International Journal of Computational Intelligence Systems
Conference Reviewer
Projects
- Cyberspace Administration of China, AI Data/Model Leakage Detection Tool, 2024.04-2025.04, PI.
- National Key R&D Program (2021YFB2012200), Security for Open CNC Systems, 2021.11-2024.10, Participant.
- Guangxi Education Department, Object Detection with Attention-based CNN, 2021.01-2025.04, Participant.
- National Natural Science Foundation of China (61962008), Color Image Hashing, 2020.01-2023.12, Participant.
Publications
† Corresponding Author | ∗ Co-first Author
Journal Articles
-
[1]
Compression Is No Barrier: Dataset Copyright Protection with Compression-Resistant Backdoor Watermarks
-
[2]
Beyond Protection: Unveiling Neural Network Copyright Trading
-
[3]
Secure Industrial Federated Learning: Label Encryption for Model Protection
-
[4]
Federated Learning with Bilateral Defense via Blockchain
-
[5]
Deep Model Intellectual Property Protection with Compression-Resistant Model Watermarking
Hewang Nie, Songfeng Lu, Junjun Wu, Jianxin Zhu.
IEEE Transactions on Artificial Intelligence, 2024.
-
[6]
PersistVerifty: Federated model ownership verification with spatial attention and boundary sampling
-
[7]
FedCRMW: Federated model ownership verification with compression-resistant model watermarking
-
[8]
Securing IP in Edge AI: Neural Network Watermarking for Multimodal Models
-
[9]
Color image reversible data hiding with double-layer embedding
Conference Papers
-
[10]
Optimized Dynamic Watermarking for Audio DNNs with Adaptive Embedding and Boundary Sampling
-
[11]
FedDiT: Federated Learning by Distillation Token Enhanced Vision Transformer
-
[12]
VeriChroma: Ownership Verification for Federated Models via RGB Filters
-
[13]
Lightweight Byzantine-Robust and Privacy-Preserving Federated Learning
Patents & Software Copyrights
Patents
- Songfeng Lu, Zhi Lu, Hewang Nie, Hao Yang. Industrial Control Data Security Protection System. CN118133298A (2024).
- Songfeng Lu, Litian Zhou, Jianxin Zhu, Yong Luo, Hewang Nie. Log Audit Method for CNC Systems based on LSTM. CN116781321A (2023).
- Songfeng Lu, Jue Xiao, Zhi Lu, Hewang Nie, Hao Yang. Stream Cipher Device for CNC Systems. CN116684076A (2023).
Software Copyrights
- Hewang Nie, Zhenjun Tang, Xiaoping Liang. Compression-Resistant Federated Learning Model Watermarking System V1.0. (No. 2025R11L2606607)
- Hewang Nie, Zhenjun Tang, Lv Chen. Audio DNN Model Watermarking System Based on Boundary Sampling V1.0. (No. 2025SR2186923)
- Hewang Nie, Zhenjun Tang, Man Ling. License Plate Recognition Software V1.0 based on SVM. (No. 2018SR219059)
Honors
- Outstanding Graduating Ph.D. Student (2025)
- National Scholarship for Doctoral Students (2023-2024)
Teaching & Supervision
Undergraduate Supervision:
- Rui Gao (Class of 2019): Third Prize, LanQiao Cup (Guangxi Division).
- Linrong Han (Class of 2019): Second Prize, LanQiao Cup (Guangxi Division).
Admission
We welcome motivated students interested in AI security, federated learning, and model watermarking to join our group. Diligent and proactive undergraduates are also welcome.
- Topics: Model IP protection, FL ownership verification, Data/Model watermarking, AI security evaluation
- Requirements: Solid programming skills (Python/PyTorch preferred) and strong research motivation
- Contact: Please email your CV and transcript to nhw@gxnu.edu.cn