PhD @ Tokyo Tech | Multimodal AI & AI Trustworthiness & Privacy
- LLM Security / Trustworthiness
- Diffusion Model Data Privacy
- Generative Models as Service
- PaRaMS (ICCV 2025)
- Parameter-level defense without accuracy trade-off
- Research → System → Infrastructure
- Turning papers into reproducible pipelines
- Email: wei.titech@outlook.com
