I'm Gianluca! I'm a Research Scientist at Genesis Molecular AI working on generative models for drug discovery. I hold a Ph.D. student @ Italian Institute of Technology in Computer Vision and Reinforcement Learning. In 2022, I interned at Google DeepMind.
This is my linkedin
This is my blog
This is my scholar
📫 email: gianluca@scarpellini.dev
🔭 Research on Generative models, object-detection, robotics exploration, reinforcement learning
🌱 Computer Vision, Image processing
Programming Languages: Python, C++
Courses: Udacity - Deep Reinforcement Learning, Udacity - Robotics\
Publications:
- Few-step Cofolding with All-Atom Flow Maps Arxiv
- Triangle Multiplication Is All You Need For Biomolecular Structure Representations ICLR2025
- Condensation of Force Field Parameters from Machine Learning Predicted Distributions for High-Throughput Virtual Screening Applications Journal of Chemical Information and Modeling
- Re-assembling the past: The RePAIR dataset and benchmark NeurIPS2024
- Positional diffusion: Graph-based diffusion models for set ordering Pattern Recognition
- Look Around and Learn ECCV2024
- DiffAssemble: A Unified Graph-Diffusion Model for 2D and 3D Reassembly CVPR2024
- π2vec: Policy Representation with Successor Features ICLR2024
- Lifting events to 3D Human Poses CVPRw 2021
Contributing to:
⚡ Fun fact: I love reading. Currently heading toward 35 books this year. Add me on goodreads. Sometimes I write too. You can read something