Research
I'm currently a master's student at DTU in Human-centered AI and a researcher at BrainCapture, where we're building the next foundation model for EEG data.
Selected work
- Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders →
arXiv preprint · 2026
EEG foundation models are clinically state-of-the-art but opaque. We apply TopK sparse autoencoders to SleepFM, REVE, and LaBraM to extract human-readable feature dictionaries from their embeddings.
- Clinical Utility and Feasibility of Smartphone-based EEG in Kenya: A Multicenter Observational Study →
arXiv preprint · 2026
Smartphone-grade EEG hardware deployed across 29 Kenyan clinical sites by non-specialist operators: 96% of 3,036 recordings were interpretable, validating point-of-care EEG in resource-limited settings.
- Membership Inference Attacks for Face Images Against Fine-Tuned Latent Diffusion Models →
arXiv preprint · 2025
Can you tell whether a specific set of face images was used to fine-tune a Latent Diffusion Model? We test the privacy boundary of generative image models.