Selected Papers

2026 - Accepted, spotlight

MIRA: A Score for Conditional Distribution Accuracy and Model Comparison

Sammy Nasser Sharief, Justine Zeghal, Gabriel Missael Barco, Pablo Lemos, Yashar Hezaveh, Laurence Perreault-Levasseur

International Conference on Machine Learning

Co-author. A method for checking conditional distribution estimates and comparing probabilistic models.

2025 - Journal article

Tackling the Problem of Distributional Shifts: Correcting Misspecified, High-Dimensional Data-Driven Priors for Inverse Problems

Gabriel Missael Barco, Alexandre Adam, Connor Stone, Yashar Hezaveh, Laurence Perreault-Levasseur

The Astrophysical Journal, 980, 108

First author. An update procedure that reduces bias from misspecified simulation-trained priors.

2025 - Peer-reviewed workshop paper

Blind Strong Gravitational Lensing Inversion: Joint Inference of Source and Lens Mass with Score-Based Models

Gabriel Missael Barco, Ronan Legin, Connor Stone, Yashar Hezaveh, Laurence Perreault-Levasseur

NeurIPS Workshop on Machine Learning and the Physical Sciences; ML4Astro, co-located with ICML

First author. Joint inference of source galaxies and lens-mass parameters from noisy lensing images.

2024 - Software paper

Caustics: A Python Package for Accelerated Strong Gravitational Lensing Simulations

Connor Stone, Alexandre Adam, Adam Coogan, M. J. Yantovski-Barth, Andreas Filipp, Landung Setiawan, Cordero Core, Ronan Legin, Charles Wilson, Gabriel Missael Barco, Yashar Hezaveh, Laurence Perreault-Levasseur

Journal of Open Source Software, 9(103), 7081

Contributor. Open-source, GPU-accelerated software for differentiable strong-lensing simulations.

Theses

2025 - M.Sc. thesis

Correction, guidée par les données, de distributions a priori biaisées en haute dimension pour l'astrophysique

Gabriel Missael Barco

Université de Montréal, M.Sc. thesis

Author. MSc thesis on correcting biased high-dimensional priors for astrophysical inverse problems.

Additional Papers

2025 - Journal article

The EDIBLES Survey IX: Simulations of the lambda 6614 DIB Profile Variations: A Surprising Connection with CH+

Charmi Bhatt, Jan Cami, Peter J. Sarre, Harold Linnartz, Nick L. J. Cox, Amin Farhang, Jonathan Smoker, Heather MacIsaac, Haoyu Fan, Alexander Ebenbichler, Harshit Khandelwal, Alex Romanec, Meriem Elyajouri, Pascale Ehrenfreund, Bernard H. Foing, Ana Monreal-Ibero, Gabriel Missael Barco, Jacco Th. van Loon

ACS Earth and Space Chemistry, 9(2), 227-240

Co-author. Paper from my Mitacs Globalink work on molecular spectra and diffuse interstellar bands.

2022 - Journal article

Online Education in Times of COVID: Adapting and Deploying a Data Science Program in Mexico

Andrea Monserrat Arredondo-Rodriguez, Gabriel Missael Barco, Claudia I. Garcia-Gil, Alicia del Carmen Hernandez-Guzman, Rogelio Antonio Hernandez-Lopez, Benjamin Manuel Sanchez-Lengeling, Carla Marquez-Luna

Harvard Data Science Review, 4(2)

Co-author. Lessons from moving a hands-on data science program in Mexico online during COVID.

2025 - Preprint

Mind the Information Gap: Unveiling Detailed Morphologies of z 0.5-1.0 Galaxies with SLACS Strong Lenses and Data-Driven Analysis

Ronan Legin, Connor Stone, Alexandre Adam, Gabriel Missael Barco, Adam Coogan, Nikolay Malkin, Laurence Perreault-Levasseur, Yashar Hezaveh

Submitted manuscript

Co-author. Uses generative priors to reconstruct source galaxies, lens light, and PSFs in real lensing systems.

2025 - Preprint

Pixellated Posterior Sampling of Point Spread Functions in Astronomical Images

Connor Stone, Ronan Legin, Alexandre Adam, Nikolay Malkin, Gabriel Missael Barco, Laurence Perreault-Levasseur, Yashar Hezaveh

Submitted manuscript

Co-author. Bayesian pixel-level PSF modeling with a generative prior for astronomical imaging.

2025 - Peer-reviewed workshop paper

Pokie: Posterior Accuracy and Model Comparison

Sammy Nasser Sharief, Justine Zeghal, Gabriel Missael Barco, Pablo Lemos, Yashar Hezaveh, Laurence Perreault-Levasseur

ML4Astro, co-located with ICML; NeurIPS Frontiers in Probabilistic Inference

Co-author. Sample-based posterior checking and model comparison from simulator joint samples.

2025 - Peer-reviewed workshop paper

Addressing Misspecified Physical Models: Correcting Underspecified Lens Convergence Models via Data-Driven Updates

Nicolas Payot, Gabriel Missael Barco, Laurence Perreault-Levasseur, Yashar Hezaveh

ML4Astro, co-located with ICML

Co-author. Applies prior updates to misspecified lens convergence models in strong-lensing analysis.