David W. Romero

Researcher, Cartesia AI

Joined

September 2019

Names

David W. Romero (Preferred)

Emails

****@vu.nl (Confirmed)

****@google.com (Confirmed)

****@gmail.com (Confirmed)

****@nvidia.com (Confirmed)

Personal Links

Homepage

Google Scholar

DBLP

Career & Education History

Researcher

Cartesia AI (cartesia.ai)

2025 – Present

Researcher

NVIDIA (nvidia.com)

2023 – 2025

Intern

Google Research (google.com)

2022 – 2023

PhD student

Vrije Universiteit Amsterdam (vu.nl)

2018 – 2023

Intern

Mitsubishi Electric Research Labs (merl.com)

2021 – 2021

Intern

Qualcomm AI Research (qti.qualcomm.com)

2021 – 2021

MS student

TU Berlin (tu-berlin.de)

2016 – 2018

Undergrad student

Universidad Nacional de Colombia (unal.edu.co)

2010 – 2016

Advisors, Relations & Conflicts

PhD Advisor

Jakub M. Tomczak

2019 – Present

PhD Advisor

Erik J. Bekkers

2019 – Present

PhD Advisor

Mark Hoogendoorn

2018 – Present

Expertise

Long context

efficient deep learning

group equivariance

2018 – Present

Publications

The Illustrated Evo2: A visual, interactive walkthrough of Evo 2 and the StripedHyena2 architecture.
Jared Wilber, FARHAD RAMEZANGHORBANI, Tyler C. Shimko, John St. John, David W. Romero
- IEEE VIS 2025 Workshop on Visualization for AI Explainability (VISxAI 2025)
- Readers: Everyone
RECON: Robust symmetry discovery via Explicit Canonical Orientation Normalization
Alonso Urbano, David W. Romero, Max Zimmer, Sebastian Pokutta
- ICLR 2026 Poster
- Readers: Everyone
Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning
Alisson G. Azzolini, Hannah Brandon, Prithvijit Chattopadhyay, Huayu Chen, Jinju Chu, Yin Cui, Jenna Diamond, Yifan Ding, Francesco Ferroni, Rama Govindaraju, Jinwei Gu, Siddharth Gururani, Imad El Hanafi, Zekun Hao, Jacob Samuel Huffman, Jingyi Jin, Brendan Johnson, Rizwan Khan, George Kurian, Elena Lantz et al. (25 additional authors not shown)
- CoRR 2025
- Readers: Everyone
Systems and Algorithms for Convolutional Multi-Hybrid Language Models at Scale
Jerome Ku, Eric Nguyen, David W. Romero, Garyk Brixi, Brandon Yang, Anton Vorontsov, Ali Taghibakhshi, Amy X. Lu, Dave P. Burke, Greg Brockman, Stefano Massaroli, Christopher Ré, Patrick D. Hsu, Brian L. Hie, Stefano Ermon, Michael Poli
- CoRR 2025
- Readers: Everyone
Meshtron: High-Fidelity, Artist-Like 3D Mesh Generation at Scale
Zekun Hao, David W. Romero, Tsung-Yi Lin, Ming-Yu Liu
- Submitted to ICLR 2025
- Readers: Everyone
Self-Supervised Detection of Perfect and Partial Input-Dependent Symmetries
Alonso Urbano, David W. Romero
- ICML 2024 Workshop GRaM
- Readers: Everyone
Wavelet Networks: Scale-Translation Equivariant Learning From Raw Time-Series
David W. Romero, Erik J. Bekkers, Jakub M. Tomczak, Mark Hoogendoorn
- Trans. Mach. Learn. Res. 2024
- Readers: Everyone
Fast, Expressive SE(n) Equivariant Networks through Weight-Sharing in Position-Orientation Space
Erik J. Bekkers, Sharvaree P. Vadgama, Rob Hesselink, Putri A. van der Linden, David W. Romero
- ICLR 2024
- Readers: Everyone
Meshtron: High-Fidelity, Artist-Like 3D Mesh Generation at Scale
Zekun Hao, David W. Romero, Tsung-Yi Lin, Ming-Yu Liu
- CoRR 2024
- Readers: Everyone
The Good, The Efficient and the Inductive Biases: Exploring Efficiency in Deep Learning Through the Use of Inductive Biases
David W. Romero
- CoRR 2024
- Readers: Everyone

View all 41 publications

Co-Authors

View all 106 co-authors