刘熙尧

刘熙尧，中南大学计算机学院副教授, 博士生导师。北京大学本、博毕业。

长期致力于生成式人工智能（AIGC）及可信内容治理研究，聚焦AIGC大模型开发、多模态内容生成、数字版权保护及深度伪造识别核心领域。

关研究成果在CVPR,AAAI,ECCV,IJCAI,PR,TMM,TDSC等国际知名学术刊物会议上发表40余篇论文。现担任中南大学学报英文版青年编委；CSIG数字取证和安全专委会委员。包括国家自然科学青年基金项目1项，国家自然科学基金子课题1项，国家军事科学院子课题1项，湖南省自然科学面上项目1项和湖南省自然科学青年基金项目1项，参与过多项国家自然科学基金面上项目和湖南省自然科学面上项目。与北大、拉夫堡大学等海内外顶尖团队深度合作，支持研究生联合培养与国际交流，获权威专家指导，优秀者可推荐至北京外企AI算法岗实习。

本年可招收1名博士研究生、3名硕士研究生，要求诚信踏实、爱钻研、有编程基础、能胜任算法研究与工程项目实践。

欢迎本科生加入团队进行科研训练。

联系邮箱：lxyzoewx@csu.edu.cn 微信lxyzoewx

研究方向简介：

1）生成式大模型理论研究：

混合模型协同方法研究：通过整合不同类型的人工智能模型(例如，预训练大模型、领域专用小横型、先验知识模块等) ，可以弥补大模型的不足。

2）生成式大模型应用研究：

文-图生图大模型用于教育领域，如教材生成：

  文-图生图大模型,应用于文创领域（刺绣、国画、唐卡、脸谱）：

3）生成式大模型数字内容安全：

生成式大模型中的数据样本保护，在扩散模型中嵌入内生水印和外置水印结合并进行联合训练，对数据进行有效的版权保护和溯源。

4）伪造视频检测：

 深度视频合成技术的滥用会给国防、军事、民生等多方面带来全新安全的挑战。基于深度生成网络和分类网络实现深度合成视频的主动和被动检测，相关成果对深度合成视频的安全管控有重要作用，是现在人工智能和AIGC安全方面研究热点问题。

5）图像视频指纹：

将数字内容版权保护视为二分类问题，基于深度学习网络提取图像视频数字内容的鲁棒性特征，实现对数字内容的的可靠保护，是人工智能在信息安全中应用的热点问题。

6）生成式信息隐藏方法：

 传统的信息隐藏会修改信息内容，无法达到可证的安全，会受到隐写分析器的威胁。而基于对抗式生成网络可以将特定的信息生成图像，并使用生成图像传递和还原秘密信息，从而实现可证安全的隐蔽通讯，是人工智能在信息安全中应用的热点问题。

近五年主要发表论文：

1. Ting Yang, Siyu Yang, Xiyao Liu*, Songtao Wu, Gerald Schaefer, Kuanhong Xu, Hui Fang. ACID-Style: An Adaptive Condition Injection Diffusion Model for Arbitrary Style Transfer [C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2026. (CCF-A类会议)

2. Xiyao Liu, Li an Zhong, Xiangui Kang, Gerald Schaefer, Da Huang, Ziping Ma. High-Capacity Generative Image Steganography Approach for Hiding Multiple Secret Images. [J]. IEEE Transactions on Multimedia, 2026. (CCF-A类JCR1区期刊)

3. Xiyao Liu, Ziping Ma, Jian Zhang, Gerald Schaefer, Kehua Guo, Yuesheng Zhu, Shichao Zhang. Robust and Diversified Image Steganography without Embedding Through a Disentanglement Autoencoder. [J]. IEEE Transactions on Dependable and Secure Computing, 2026. (CCF-A类JCR1区期刊)

4. Siyu Yang, Chunchen Ke, Jian Zhang, Chunwei Miao, Xiyao Liu*, Songtao Wu, Kuanhong Xu, Da Huang. UniStyleDiff A Unified Diffusion-Driven Framework for Image and Video Style Transfer. [J]. Expert Systems with Applications, 2026, (JCR1区期刊)

5. Yang T, Gao Z, Xiyao Liu*, et al. AHAI: Adaptive Hybrid-Attention Inference For Diffusion-Based Arbitrary Style Transfer [C]//ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2026: 11322-11326. (CCF-B类会议)

6. Xiyao Liu, Junxing Ma, Xinda Wang, Qianyu Lin, Jian Zhang, Gerald Schaefer, Cagatay Turkay, Hui Fang. Recoverable Facial Identity Protection via Adaptive Makeup Transfer Adversarial Attacks. [C]. Proceedings of the AAAI Conference on Artificial Intelligence, 2025, (CCF-A类会议)

7. Cundian Yang, Guibo Luo, Yuesheng Zhu, Jiaqi Li, Xiyao Liu*. Robust Image Hashing based on Contrastive Masked Autoencoder with Weak-Strong Augmentation Alignment. [C]. Proceedings of the AAAI Conference on Artificial Intelligence, 2025,(CCF-A类会议)

8. Xiyao Liu, Qingyu Dang, Huiyi Wang, Xiaoheng Deng, Xunli Fan, Cundian Yang, Zhihong Chen, Hui Fang. An Adversarial Contrastive Learning Based Cross-Modality Zero-Watermarking Scheme for DIBR 3D Video Copyright Protection. [J]. Neurocomputing, 2025,(JCR1区期刊)

9. Hao Zhanga, Zenghui Liu, Zhihang Yana, Songrui Guoc, ChunMing Gao, Xiyao Liu*. Chinese sign language recognition and translation with virtual digital human dataset. [J]. Displays, 2025. (JCR1区期刊)

10. Cundian Yang, Guibo Luo, Yuesheng Zhu, Xiyao Liu*. Robust Image Hashing based on Mixture-of-Experts with Hard-sample Mining Contrastive Learning. [C]. International Conference on Computer Supported Cooperative Work in Design (CSCWD), 2025. (CCF-C类会议)

11. Xiyao Liu, Cundian Yang, Jianbiao He, Hui Fang, Gerald Schaefer, Jian Zhang, Yuesheng Zhu, Shichao Zhang. Attack-Defending Contrastive Learning for Volumetric Medical Image Zero-Watermarking. [J]. ACM Transactions on Multimedia Computing, Communications and Applications, 2024, doi: 10.1145/3702230. (JCR1区期刊)

12. Xiyao Liu, Fengkai Dong, Xin Liao, Yuhan Guo, Jianbiao He, Jian Zhang, Gerald Schaefer, Hui Fang. Multi-Strategy Adversarial Learning for Robust Face Forgery Detection Under Heterogeneous and Composite Attacks. [C]. IEEE International Conference on Multimedia and Expo (ICME), 2024, doi: 10.1109/ICME57554.2024.10688131. (CCF-B类会议)

13. Lei Wang, Tianfu Cai, Pinyi Huang, Xiyao Liu*, Wangyang Cai. Two-Stage Facial Expression Spotting with Spectrum-Based Post-Processing. [C]. IEEE International Conference on Multimedia and Expo (ICME), 2024,(CCF-B类会议)

14. Xiyao Liu, Siyu Yang, Jian Zhang, Gerald Schaefer, Jiya Li, Xunli Fan, Songtao Wu, Hui Fang. Towards compact reversible image representations for neural style transfer. [C]. European Conference on Computer Vision (ECCV), 2024,(CCF-B类会议)

15. Ziping Ma, Yuesheng Zhu, Guibo Luo, Xiyao Liu*, Gerald Schaefer, Hui Fang. Robust Steganography without Embedding based on Secure Container Synthesis and Iterative Message Recovery. [C]. Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence (IJCAI), 2023. (CCF-A类会议)

16. Wenyuan Yang, Shuo Shao, Yue Yang, Xiyao Liu*, Ximeng Liu, Zhihua Xia, Gerald Schaefer, Hui Fang. Watermarking in Secure Federated Learning: A Verification Framework based on Client-Side Backdooring. [J]. ACM Transactions on Intelligent Systems and Technology, 2023,JCR1区期刊)

17. Hao Zhang, Yixiang Sun, Zenghui Liu, Qiyuan Liu, Xiyao Liu*, Ming Jiang, Gerald Schafer, Hui Fang. Heterogeneous Attention Based Transformer for Sign Language Translation. [J]. Applied Soft Computing, 2023. (JCR1区期刊)

18. Li, Fangfang, Guikai Chen, and Xiyao Liu*. "An event-based automatic annotation method for datasets of interpersonal relation extraction." Applied Intelligence 53.3 (2023). (JCR2区期刊)

19. Xiyao Liu, Yang Liu, Yuhao Zheng, Ting Yang, Jian Zhang, Victoria Wang, Hui Fang. Semantics-guided generative diffusion model with a 3DMM model condition for face swapping. [J]. Computer Graphics Forum, 2023, doi: 10.1111/cgf.14949. (JCR2区期刊)

20. Xiyao Liu, Ma Z, Ma J, et al. Image disentanglement autoencoder for steganography without embedding [C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022. (CCF-A类会议)

21. Xiyao Liu, Zhang Y, Du S, et al. DIBR zero-watermarking based on invariant feature and geometric rectification [J]. IEEE MultiMedia, 2022. (JCR1区期刊论文)

22. Li, F., Shan, Y., Mao, X., Ren, X., Xiyao Liu*, Zhang, S. Multi-task joint training model for machine reading comprehension. Neurocomputing, 2022. (JCR1区期刊论文)

23. Xiyao Liu, Ma Z, Chen Z, et al. Hiding multiple images into a single image via joint compressive autoencoders [J]. Pattern Recognition, 2022. (CCF-B类期刊)

24. Xiyao Liu, Zhang Y, Wang J, et al. Multiple-feature-based zero-watermarking for robust and discriminative copyright protection of DIBR 3D videos [J]. Information Sciences, 2022. (CCF-B类期刊)

25. Xiyao Liu, Zhang Y, Du S, et al. Discriminative and geometrically robust zero-watermarking scheme for protecting DIBR 3D videos [J]. 2022. (CCF-B类会议)

26. Xiyao Liu, Wang Y, Sun Z, et al. Robust and discriminative zero-watermark scheme based on invariant features and similarity-based retrieval to protect large-scale DIBR 3D videos [J]. Information Sciences, 2022. (CCF-B类期刊)

主要主持项目：