Portrait de Gabriel Sasseville n'est pas disponible

Gabriel Sasseville

gabriel.sasseville@mila.quebec
Doctorat - UdeM
Superviseur⋅e principal⋅e
Pierre-Luc Bacon
Sujets de recherche
Apprentissage par renforcement
Apprentissage profond
Généralisation hors distribution (OOD)
IA pour le changement climatique
Modèles de diffusion
Séries temporelles
Traitement du langage naturel
Vision par ordinateur
Site web
Google Scholar
GitHub

Publications

Secure Agentic Coding through Counterexample-Grounded Feedback
Tue Le
Anh Nguyen Hoang
Gabriel Sasseville
Doseok Jang
Pierre-Luc Bacon
Nghi D. Q. Bui
As coding models and agent harnesses become more capable, reliable security verification is becoming a central bottleneck. Improving secure … (voir plus)code generation requires a signal whose reported progress remains aligned with actual program behavior under optimization; otherwise, an optimizer can improve the signal's verdict without removing the underlying vulnerability. Existing model-based approaches require models to either infer program behavior from source code or specify correct outcomes for adversarial tests, placing substantial reasoning demands on the security signal. We introduce \falcon{}, a counterexample-grounded verification framework that separates attack exploration from behavioral verification and connects them through execution traces of the program under test. In test-time repair, \falcon{} yields substantially stronger held-out functional-security gains than alternative security signals across secure code generation benchmarks, while preserving or improving functionality. As an RL reward, \falcon{} also outperforms static-analysis and learned rewards on held-out functional-security, even when all policies successfully optimize their respective training rewards. Combining \falcon{} with either proxy reward further improves held-out performance, suggesting that execution-grounded feedback can mitigate reward overoptimization while preserving useful guidance from existing security signals.
2026-10-18
arXiv (prépublication)
doi.org
arxiv.org
Probabilistic Interpolation of Sagittarius A*’s Multiwavelength Light Curves Using Diffusion Models
Gabriel Sasseville
Julie Hlavacek-Larrondo
Daryl Haggard
Alexandre Adam
Hadrien Paugnat
Gunther Witzel
Abstract Understanding the variability of Sagittarius A* (Sgr A*) requires coordinated, multiwavelength observations that span the electroma… (voir plus)gnetic spectrum. In this work, we focus on data from four key observatories: Chandra in the X-ray (2–8 keV), GRAVITY on the Very Large Telescope in the near-infrared (2.2 μ m), Spitzer in the infrared (4.5 μ m), and the Atacama Large Millimeter/submillimeter Array in the submillimeter (340 GHz). These multiband observations are essential for probing the physics of accretion and emission near the black hole’s event horizon, yet they suffer from irregular sampling, band-dependent noise, and substantial data gaps. These limitations complicate efforts to robustly identify flares and measure cross-band time lags, key diagnostics of the physical processes driving variability. To address this challenge, we introduce a diffusion-based generative model, for interpolating sparse, multivariate astrophysical time series. This represents the first application of score-based diffusion models to astronomical time series. We also present the first transformer-based model for light-curve reconstruction that includes calibrated uncertainty estimates. The models are trained on simulated light curves constructed to match the statistical and observational characteristics of real Sgr A* data. These simulations capture correlated multiband variability, realistic observation cadences, and wavelength-specific noise. We compare our models against a multi-output Gaussian process. The diffusion model achieves superior accuracy and competitive calibration across both simulated and real datasets, demonstrating the promise of diffusion models for high-fidelity, uncertainty-aware reconstruction of multiwavelength variability in Sgr A*.
2026-02-02
Astrophysical Journal (publié)
doi.org
arxiv.org