About Me
I am a Machine Learning Engineer at TikTok, with a Ph.D. from the National University of Singapore. My work bridges research and industry, with a focus on building robust and secure AI systems. My research interests include diffusion models, large language models, graph-based learning, digital watermarking, and AI security. In the industry, I have applied these areas to real-world systems, contributing to recommendation systems, content moderation, and risk detection.
Publications
-
Mitigating Adversarial Attacks by Distributing Different Copies to Different BuyersProceedings of the 2023 ACM on Asia Conference on Computer and Communications Security (AsiaCCS)
-
Confusing and Detecting ML Adversarial Attacks with Injected AttractorsProceedings of the 2022 ACM on Asia Conference on Computer and Communications Security (AsiaCCS)
-
Common Component in Black-Boxes Is Prone to Attacks26th European Symposium on Research in Computer Security (ESORICS)
-
Finding Input Data Domains of Image Classification Models with Hard-Label Black-Box AccessProceedings of the 32nd ACM International Conference on Multimedia, 2024 (MM)
-
Neural Network Inversion in Adversarial Setting via Background Knowledge AlignmentProceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security (CCS)
-
Tracing the origin of adversarial attack for forensic investigation and deterrenceIEEE/CVF International Conference on Computer Vision, 2023 (ICCV)
-
Flow-Based Robust Watermarking with Invertible Noise Layer for Black-Box DistortionsProceedings of the AAAI Conference on Artificial Intelligence, 2023 (AAAI)
-
De-END: Decoder-driven Watermarking NetworkIEEE Transactions on Multimedia, 2022 (TMM)
-
DP 2 Dataset Protection by Data PoisoningIEEE Transactions on Dependable and Secure Computing, 2024 (TDSC)
Patents
-
Interactive Silent Liveness DetectionUS Patent 11,922,732, 2024
-
Watermark as Honeypot for Adversarial DefenseUS Patent 11,501,136, 2022
-
Automatic document processingUS Patent 11,455,713, 2022
-
Ensemble Method for Face Recognition Deep Learning ModelsUS Patent 11,689,526, 2023
-
Image forgery detection via headpose estimationUS Patent 12,067,475, 2024
-
Removing Clarity Issues From Images To Improve ReadabilityUS Patent App. 17/645,484, 2023
-
Real-time Media Alteration Using Generative TechniquesUS Patent App. 18/512,007, 2023
-
Fraud Detection for Signed DocumentsUS Patent App. 18/051,580, 2024
-
Document Similarity Learning by Heterogeneous Group RelationshipsUS Patent App. 18/461,299, 2023
-
Seller Risk Detection by Product Community and Supply Chain ModellingUS Patent App. 18/082,870, 2022
-
Two-Sided Machine Learning Framework for Pointer Movement-Based Bot DetectionUS Patent App. 18/146,738, 2022
-
Detection of Physical Tampering on DocumentsUS Patent App. 17/850,602, 2023
-
Image Forgery Detection via Pixel-Metadata Consistency AnalysisUS Patent App. 17/374,439, 2023
-
Machine Learning Model for Image Forgery DetectionUS Patent App. 18/189,044, 2023