Published:
This paper introduces a novel framework to use heavy-tailed noise in the denoising diffusion paradigm, which constitutes a generalization of the original DDPM method. Using heavy-tailed noise is shown to bring benefits in various contexts: heavy-tailed data distributions, better robustness to class imbalance, and smaller computational time.
Published:
We introduce a novel class of generative models based on piecewise deterministic Markov processes (PDMPs), which combine deterministic motion with random jumps. Like diffusions, PDMPs can be reversed in time. We derive explicit expressions for jump rates and kernels in the time-reversed processes and propose efficient training methods and approximate simulation techniques. Additionally, we provide bounds on the total variation distance between the data and model distributions, supported by promising numerical simulations.
Published:
Finished TA’ing MAA106 numerical analysis course in Ecole Polytechnique. Thanks Maxime Breden for the teaching experience and congrats to the students!
Published:
Attending the Alan Turing Institute Summer School on Diffusion Models in London.
Published:
Our generative PDMP project is accepted to NeurIPS 2024!
Published:
Presenting PDMP at both NeurIPS in Paris and official NeurIPS in Vancouver.
Published:
Heavy-tailed diffusion with DLPM accepted at ICLR 2025!
Published:
Presenting DLPM at Ecole Polytechnique.
Published:
Launching the diffusion model reading group in Inria Paris. Feel free to reach out if you want to join!
Published:
Presenting our discrete diffusion DMPM project at Ecole Polytechnique.
Published:
Attending the mini-workshop ‘Statistical Challenges for Deep Generative Models’ in legendary Oberwolfach, Germany.
Published:
Invited in Padova by Giovanni Conforti to explore cosmological applications of diffusion models. Looking forward to working with such a great person!
Published:
Presenting DLPM at ICLR 2025.
Published:
Our discrete diffusion project DMPM is accepted in ICML 2025!
Published:
I am starting an internship at Sakana AI in Tokyo, advised by legendary Stefano Peluchetti (who discovered flow matching a year before… flow matching).
Published:
I am co-organising a 5-day research retreat with Sakana’s research staff!
Published:
Our work on algorithm and data dependent generalization bounds for diffusion models is accepted to NeurIPS 2025!
Short description of portfolio item number 1
Short description of portfolio item number 2
Published in NeurIPS 2024, 2024
We introduce a novel class of generative models based on piecewise deterministic Markov processes (PDMPs), which combine deterministic motion with random jumps. Like diffusions, PDMPs can be reversed in time. We derive explicit expressions for jump rates and kernels in the time-reversed processes and propose efficient training methods and approximate simulation techniques. Additionally, we provide bounds on the total variation distance between the data and model distributions, supported by promising numerical simulations.
Published in ICLR 2025, 2025
This paper introduces a novel framework to use heavy-tailed noise in the denoising diffusion paradigm, which constitutes a generalization of the original DDPM method. Using heavy-tailed noise is shown to bring benefits in various contexts: heavy-tailed data distributions, better robustness to class imbalance, and smaller computational time.
Published in ICML 2025, 2025
This paper introduces a novel framework for discrete data generation on the hypercube ${0, 1}^d$. We establish theoretical and methodological alignment with classical continuous score-based modesls. We demonstrate the effectiveness of this approach on low and high dimensional datasets (Binary MNIST), beating other state-of-the-art methods like Discrete Flow Matching
Published in NeurIPS 2025, 2025
This paper studies the generalization property of diffusion models through the lens of statistical learning. We develop a perspective that enables using existing tools to characterize the generalization ability of these generative models.
Published:
Published:
Published:
Published:
Admission procedure, Ecole Polytechnique - HEC, 2024
Assisting the admission team during the multiple rounds of the selection process by conducting the mathematical interviews.
Undergraduate course, Ecole Polytechnique, CMAP, 2024
TA’d a 4 months course for 1st year students on numerical analysis.
Admission procedure, Ecole Polytechnique - HEC, 2025
Assisting the admission team during the multiple rounds of the selection process by conducting the mathematical interviews.