Dongdong Lin | Academic Homepage

Bio

Dongdong Lin is currently a lecturer at Huaqiao University, working with Hui Tian. He received his B.Sc. and M.Sc. degrees from Xiangtan University, supervised by Chengqing Li, and completed his Ph.D. at Shenzhen University, supervised by Bin Li. From 2020 to 2022, he was a visiting student at the University of Siena, Italy, working with Mauro Barni and Benedetta Tondi.

His research focuses on AI security, deep model watermarking, deepfake detection, adversarial robustness, and multimedia forensics. Recent work spans watermarking for diffusion models, speech synthesis systems, and ownership verification of generative models.

He is currently interested in trustworthy AIGC, robust model ownership protection, and security problems around modern multimodal generative systems.

He serves as a CCF Professional Member, a CSIG Member, and an IEEE Member. He also serves as a reviewer for IEEE T-IFS/DSC/ASLP, SP, CVPR, ACM MM, and ICASSP.

News

2025.04

🎉One paper is accepted to ICASSP 2025.

2024.10

🎉One paper is accepted to ACM MM.

2024.07

🎉One paper is accepted to TDSC.

Preprints

arXiv2026

Figure from the Diffusion-Guided Adversarial Perturbation Injection paper

Diffusion-Guided Adversarial Perturbation Injection for Generalizable Defense Against Facial Manipulations

Adversarial ExampleDeepfake Detection

Yue Li, Linying Xue, Kaiqing Lin, Hanyu Quan, Dongdong Lin, Hui Tian, Hongxia Wang, Bin Wang

arXiv preprint, 2026

PDF Paper

@article{li2026diffusion,
  title={Diffusion-Guided Adversarial Perturbation Injection for Generalizable Defense Against Facial Manipulations},
  author={Li, Yue and Xue, Linying and Lin, Kaiqing and Quan, Hanyu and Lin, Dongdong and Tian, Hui and Wang, Hongxia and Wang, Bin},
  journal={arXiv preprint arXiv:2604.01635},
  year={2026}
}

arXiv2025

Figure from the Imperceptible Adversarial Defense paper

Towards Imperceptible Adversarial Defense: A Gradient-Driven Shield against Facial Manipulations

Adversarial ExampleDeepfake Detection

Yue Li, Linying Xue, Dongdong Lin, Qiushi Li, Hui Tian, Hongxia Wang

arXiv preprint, 2025

PDF Paper

@article{li2025towards,
  title={Towards Imperceptible Adversarial Defense: A Gradient-Driven Shield against Facial Manipulations},
  author={Li, Yue and Xue, Linying and Lin, Dongdong and Li, Qiushi and Tian, Hui and Wang, Hongxia},
  journal={arXiv preprint arXiv:2510.01699},
  year={2025}
}

arXiv2024

Figure from the latent diffusion watermarking paper

An Efficient Watermarking Method for Latent Diffusion Models via Low-Rank Adaptation and Dynamic Loss Weighting

Model WatermarkingText-to-image ModelLow-Rank Adaptation

Dongdong Lin, Yue Li, Benedetta Tondi, Kaiqing Lin, Bin Li, Mauro Barni

arXiv preprint, 2024

PDF Paper

@article{lin2024efficient,
  title={An efficient watermarking method for latent diffusion models via low-rank adaptation},
  author={Lin, Dongdong and Li, Yue and Tondi, Benedetta and Li, Bin and Barni, Mauro},
  journal={arXiv preprint arXiv:2410.20202},
  year={2024}
}

arXiv2025

TriniMark: A Robust Generative Speech Watermarking Method for Trinity-Level Attribution

Speech Watermarking

Yue Li, Weizhi Liu, Dongdong Lin

arXiv preprint, 2025

PDF Paper

@article{li2025trinimark,
  title={TriniMark: A Robust Generative Speech Watermarking Method for Trinity-Level Attribution},
  author={Li, Yue and Liu, Weizhi and Lin, Dongdong},
  journal={arXiv preprint arXiv:2504.20532},
  year={2025}
}

arXiv2025

SOLIDO: A Robust Watermarking Method for Speech Synthesis via Low-Rank Adaptation

Speech Watermarking

Yue Li, Weizhi Liu, Dongdong Lin

arXiv preprint, 2025

PDF Paper

@article{li2025solido,
  title={SOLIDO: A Robust Watermarking Method for Speech Synthesis via Low-Rank Adaptation},
  author={Li, Yue and Liu, Weizhi and Lin, Dongdong},
  journal={arXiv preprint arXiv:2504.15035},
  year={2025}
}

arXiv2025

Figure from the Protecting Your Voice paper

Protecting Your Voice: Temporal-aware Robust Watermarking

Speech Watermarking

Yue Li, Weizhi Liu, Dongdong Lin, Hui Tian, Hongxia Wang

arXiv preprint, 2025

PDF Paper

@article{li2025protecting,
  title={Protecting Your Voice: Temporal-aware Robust Watermarking},
  author={Li, Yue and Liu, Weizhi and Lin, Dongdong and Tian, Hui and Wang, Hongxia},
  journal={arXiv preprint arXiv:2504.14832},
  year={2025}
}

Publications

ICASSP2025Fine Minima Optimizer

Exploiting Robust Model Watermarking Against the Model Fine-Tuning Attack via Flat Minima Aware Optimizers

Model WatermarkingWatermark Robustness

Dongdong Lin, Yue Li, Bin Li, Jiwu Huang

2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Paper

@inproceedings{10890107,
  author={Lin, Dongdong and Li, Yue and Li, Bin and Huang, Jiwu},
  booktitle={ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  title={Exploiting Robust Model Watermarking Against the Model Fine-Tuning Attack via Flat Minima Aware Optimizers},
  year={2025},
  pages={1-5},
  doi={10.1109/ICASSP49660.2025.10890107}
}

ACMMM2024

GROOT: Generating Robust Watermark for Diffusion-Model-Based Audio Synthesis

Audio WatermarkingACM MM 2024CCF-A

Weizhi Liu, Yue Li, Dongdong Lin, Hui Tian, Haizhou Li

Proceedings of the 32nd ACM International Conference on Multimedia (ACMMM)

Paper

@inproceedings{10.1145/3664647.3680596,
author = {Liu, Weizhi and Li, Yue and Lin, Dongdong and Tian, Hui and Li, Haizhou},
title = {GROOT: Generating Robust Watermark for Diffusion-Model-Based Audio Synthesis},
year = {2024},
doi = {10.1145/3664647.3680596},
booktitle = {Proceedings of the 32nd ACM International Conference on Multimedia},
pages = {3294-3302},
numpages = {9}
}

TDSC2025

Figure from the CycleGAN watermarking paper

A CycleGAN Watermarking Method for Ownership Verification

GAN WatermarkingOwnership VerificationTDSC 2025CCF-A

Dongdong Lin, Benedetta Tondi, Bin Li, Mauro Barni

IEEE Transactions on Dependable and Secure Computing

Paper

@article{10591362,
  author={Lin, Dongdong and Tondi, Benedetta and Li, Bin and Barni, Mauro},
  journal={IEEE Transactions on Dependable and Secure Computing},
  title={A CycleGAN Watermarking Method for Ownership Verification},
  year={2025},
  volume={22},
  number={2},
  pages={1040-1054},
  doi={10.1109/TDSC.2024.3424900}
}

PR2023Bi-Granularity Artifacts

Watching the big artifacts: Exposing deepfake videos via bi-granularity artifacts

Deepfake DetectionPR 2023CCF-B

Han Chen, Yuezun Li, Dongdong Lin, Bin Li, Junqiang Wu

Pattern Recognition

Paper

@article{CHEN2023109179,
title = {Watching the BiG artifacts: Exposing DeepFake videos via Bi-granularity artifacts},
journal = {Pattern Recognition},
volume = {135},
pages = {109179},
year = {2023},
doi = {https://doi.org/10.1016/j.patcog.2022.109179},
author = {Han Chen and Yuezun Li and Dongdong Lin and Bin Li and Junqiang Wu}
}

IJCB2022Temporal Transferability

Exploiting temporal information to prevent the transferability of adversarial examples against deep fake detectors

Adversarial ExampleIJCB 2022CCF-C

Dongdong Lin, Benedetta Tondi, Bin Li, Mauro Barni

2022 IEEE International Joint Conference on Biometrics (IJCB)

Paper

@inproceedings{10007959,
  author={Lin, Dongdong and Tondi, Benedetta and Li, Bin and Barni, Mauro},
  booktitle={2022 IEEE International Joint Conference on Biometrics (IJCB)},
  title={Exploiting temporal information to prevent the transferability of adversarial examples against deep fake detectors},
  year={2022},
  pages={1-8},
  doi={10.1109/IJCB54206.2022.10007959}
}

IEEEMM2018Cryptanalysis

Cryptanalyzing an image encryption algorithm based on autoblocking and electrocardiography

Cryptanalysis🏆 ESI Highly Cited Paper

Chengqing Li, Dongdong Lin, Jinhu Lu, Feng Hao

IEEE MultiMedia, 2018

Paper

@article{8495009,
  author={Li, Chengqing and Lin, Dongdong and Lu, Jinhu and Hao, Feng},
  journal={IEEE MultiMedia},
  title={Cryptanalyzing an Image Encryption Algorithm Based on Autoblocking and Electrocardiography},
  year={2018},
  volume={25},
  number={4},
  pages={46-56},
  doi={10.1109/MMUL.2018.2873472}
}

IEEEMM2017Pixel Bits Cipher

Cryptanalyzing an image-scrambling encryption algorithm of pixel bits

Cryptanalysis🏆 ESI Highly Cited Paper

Chengqing Li, Dongdong Lin, Jinhu Lu

IEEE MultiMedia, 2017

Paper

@article{7999153,
  author={Li, Chengqing and Lin, Dongdong and Lu, Jinhu},
  journal={IEEE MultiMedia},
  title={Cryptanalyzing an Image-Scrambling Encryption Algorithm of Pixel Bits},
  year={2017},
  volume={24},
  number={3},
  pages={64-71},
  doi={10.1109/MMUL.2017.3051512}
}