Skip to content
View pedrocurvo's full-sized avatar
👋
👋
12 followers · 14 following

Achievements

Achievement: StarstruckAchievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: StarstruckAchievement: Pair ExtraordinaireAchievement: Pull Sharkx2Achievement: YOLOAchievement: Quickdraw

Highlights

  • Pro

Block or report pedrocurvo

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
pedrocurvo/README.md

👋 Pedro Curvo

Incoming PhD Candidate

Website · GitHub · LinkedIn · Email · Resume

I'm an incoming PhD candidate, currently finishing my MSc in Artificial Intelligence at the University of Amsterdam, working on generative modeling — mainly flow matching and diffusion models — with AMLab, under Jan-Willem van de Meent. My physics background (BSc, Engineering Physics, IST Lisbon) shows up everywhere in how I think about these models: as dynamical systems, not black boxes.

The thread running through my work: models that adapt through data, not just through parameter updates.

Currently

  • 🎓 Finishing my MSc thesis on reference-guided flow matching
  • 🏆 Presented an MSPT poster at CVPR 2026 (Highlight)
  • 🧪 TA for Mechanistic Interpretability @ UvA
  • 🌍 Based in Amsterdam, from Portugal

Selected Publications

Paper Venue Link
Follow the Mean: Reference-Guided Flow Matching 2026 arXiv:2605.10302
Multi-Scale Patch Transformer CVPR 2026 — Highlight arXiv:2512.01738
Deep Learning for High Aspect Ratio Fusion Devices 2024 arXiv:2409.00564

Stack

Python PyTorch C++ TypeScript React

top languages

If you're into flow matching, diffusion models, or just want to talk physics-meets-ML — my inbox is open.

Pinned Loading

  1. follow-the-mean follow-the-mean Public

    Follow the Mean: controlling flow-matching generative models by shifting endpoint means with reference examples

    Python 22 5

  2. mspt mspt Public

    Near-linear scaling neural PDE solver for large unstructured meshes via multi-scale attention with ball-tree partitioning

    Python 13

  3. TheTraitors TheTraitors Public

    The Traitors: Deception and Trust

    Python 1

  4. MLStellaratorDesign MLStellaratorDesign Public

    Stellarator Construction via Deep Learning

    Python 4