About
Heyo! I鈥檓 Matias 馃憢. I鈥檓 interested in a couple of topics, mainly around Computer Vision. Lately, I鈥檝e been broadly interested in diffusion and flow-matching methods, differential equations in machine learning, and generative methods in general. I recently finished my Master鈥檚 in CS at Universidad de Buenos Aires, and I鈥檓 currently looking for PhD opportunities.
This site鈥檚 purpose is to be a polished digital notebook. It somewhat reflects my different thoughts on the various subjects I鈥檓 interested in learning or researching. I sometimes take notes and add them here as blog posts. They take an incredibly long time to produce, and, as I don鈥檛 have much free time, I post much more infrequently than I want to.
Papers and Publications
FLOWING 馃寠: Implicit Neural Flows for Structure-Preserving Morphing
NeurIPS 2025
Nuevas Variantes y Optimizaci贸n de M茅todos Qss para Geant4: Simulaci贸n por Cuantificaci贸n de Estados para F铆sica de Altas Energ铆as
Master鈥檚 Thesis. 2025
A Lightweight Diffusion Model with Modified Sampler
MCPR 2025
This work was a spin-off from my participation in the LatinX in AI (LXAI) Mentorship Program 2024.
Notes / Contact
If you鈥檇 like to get in touch regarding my posts or any other issue, feel free to contact me at mgrynberg@dc.uba.ar