> About
I am a first-year PhD researcher at the Université de Neuchâtel, supervised by Prof. Lydia Y. Chen. My research focuses on AI Safety & Alignment, computer vision, and NLP.
> Publications
An Optimal Transport View of Activation Steering In Masked Diffusion Models
ICLR Workshop TTU 2026Gert Lek, Chaoyi Zhu, Pin-Yu Chen, Robert Birke, Lydia Y. Chen
Introduces an Optimal Transport view of activation steering for masked diffusion models, learning a lightweight affine map that transports activation distributions between behaviors. Unifies common steering rules as special cases and improves instruction-following accuracy across LLaDA and Dream-Instruct.
Detective SAM: Adaptive AI-Image Forgery Localization
ICLR 2026Gert Lek, Nicolas van Schaik, Chaoyi Zhu, Pin-Yu Chen, Robert Birke, Lydia Y. Chen
Extends SAM2 for image forgery localization with perturbation-driven forensic embeddings, lightweight feature adapters, and a learnable prompt module. Introduces AutoEditForge, an automated diffusion edit generation pipeline. Achieves SOTA on common benchmarks and modern editing models such as NanoBanana and Qwen-Image-Edit.
Detective SAM: Adapting SAM to Localize Diffusion-based Forgeries via Embedding Artifacts
ICML Workshop DIG-BUGS 2025Gert Lek, Chaoyi Zhu, Pin-Yu Chen, Robert Birke, Lydia Y. Chen
Extends SAM with blur-driven forensic embedding signals, hierarchical learnable prompts, and lightweight adapters for automatic forgery mask generation. Outperforms prior methods on MagicBrush and CoCoGlide.
> Experience
PhD Researcher @ Université de Neuchâtel
2024 – PresentResearch on AI Safety & Alignment, computer vision, and NLP. Supervised by Prof. Lydia Y. Chen.
R&D @ Ortec Finance
2023 – 2024Reinforcement learning, bandit problems, and software development.
> Education
MSc Quantitative Finance @ Erasmus School of Economics
2022 – 2023Thesis: "Interpretable High-dimensional Continuous RL" - Grade: 9.0/10
BSc Mathematics @ TU Delft
2019 – 2022Thesis: "Robust Optimal Classification Trees" - Grade: 9.0/10
> Posts
OT Activation Steering accepted at ICLR 2026 TTU Workshop
Our paper on optimal transport-based activation steering for masked diffusion language models was accepted at the ICLR 2026 TTU Workshop.
Detective SAM accepted at ICLR 2026
Our full paper extending Detective SAM with AutoEditForge and adaptive forgery localization was accepted at ICLR 2026.
Detective SAM accepted at ICML 2025 DIG-BUGS Workshop
Our paper on adapting SAM for diffusion-based image forgery localization was accepted at the ICML 2025 DIG-BUGS Workshop.
> Contact
Feel free to reach out if you're interested in collaborating or have questions about my research.