Pedro M. P. Curvo
Personal notebookAmsterdam · 52.3676° N, 4.9041° E

Hello 👋

Welcome to my estate on the World Wide Web.

I'm Pedro—Pedro M. P. Curvo in my publications.

Current workMost of the times lost in the woods. The road ahead? Trying to teach machines how to imagine
Now reading1984 · George Orwell
ElsewhereWell, usually travelling or looking for something new to eat in the good company of my (foodie) girlfriend

Here is a visual reference

Fig. 01 · visual reference, more or less

Who am I, you ask?

Sometimes, I also do not know, but the shorter answer—even though it's more about what I do than who I am, and probably the one you care about—is that I'm a PhD candidate in Artificial Intelligence at the University of Amsterdam, collaborating between VISLab and AMLab, with my research focusing on World Models1.

I've created this place to share my thoughts, projects, and writing, both in and out of research. The goal was to have a space where I can share random and not-so-random texts. This also means the writing will not always be perfect, but it will get better over time. But what matters most is that it's my place—a place I can look back on one day and remember. Whether you enjoy it or not, I hope you find it interesting.

A note on World Models

Well, it seems there is no consensus on what exactly a World Model is, and if it had already been figured out, then I would be out of a job—or, in this case, out of a PhD.

The way I see it, or at least the way I can describe it, is that a World Model mimics our way of predicting the future. Not the future in the sense of predicting the next lottery numbers, but the future of the simple actions we take: imagining what to write before we do it, where a ball will land if we throw it, or what a painting will look like if we apply a certain brushstroke, and so on and so forth.

Well, maybe predict is not the right word. Maybe imagination is.

Continue reading
  • Our new preprint, LiFT: Loop Flow Transformers, is out! ♾️ Loop in depth, flow in time.

  • Accepted at NeurIPS 2026 main: Follow the Mean 🐘! See you in Sydney 🇦🇺

  • I started my PhD in Artificial Intelligence at the University of Amsterdam, researching World Models with VISLab and AMLab.

  • MSPT is a CVPR 2026 Highlight! We study efficient large-scale physical modelling through parallelised multi-scale attention.

Selected research

All publications

Recently added

From the notebook

See all writing