Martin Trapp
Almost surely
Assistant Professor in Machine Learning & WASP Fellow
KTH Royal Institute of Technology
I am an Assistant Professor in Machine Learning at KTH Royal Institute of Technology, WASP fellow, and a member of the ELLIS society. Before joining KTH, I was an Academy of Finland postdoctoral researcher at Aalto University working with Arno Solin. My research is centred around representing, quantifying, and reducing uncertainties to make machine learning more trustworthy. I am particularly interested in efficient and principled approaches for large-scale models such as LLMs & VLMs.
See biography for more details.
Research Interests
- Tractable Models: Probabilistic circuits, neurosymbolics, and exact & approximate Bayesian inference.
- Bayesian Learning: Uncertainty quantification & reduction in deep learning, inductive biases, and non-parametric & function-space methods.
- Probabilistic Numerics: Probabilistic programming, numerical imprecision, error and uncertainty quantification.
News & Updates
ICLR Paper accepted!
January 2026Our paper on Post-hoc Probabilistic Vision-Language Models (BayesVLM) has been accepted to ICLR 2026. Check out the paper
CVPR Workshop on Uncertainty Quantification for Computer Vision.
June 2025Together with collegues we organised the 4th Workshop on Uncertainty Quantification for Computer Vision at CVPR 2025.
UAI Paper accepted!
May 2025Our BitVI paper has been accepted to UAI 2025. Check out the paper, the code will be released soon.