Dario Shariatian

Dario Shariatian

PhD Student. Developing methodologies for generative models, focus on diffusion models and related approaches

Publications

You can also find my articles on my Google Scholar profile.

Preprints

Denoising Markov Probabilistic Models (DMPM)

Published in Arxiv, 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

Shariatian D., Pham L.T.N., Ocello A., Conforti G., Durmus A.O. (2025). Denoising Markov Probabilistic Models. ArXiv, abs/2502.07939.
See paper | GitHub Repository

Conference Papers

Denoising Levy Probabilistic Models (DLPM)

Published in The Thirteenth International Conference on Learning Representations (ICLR), 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.

Shariatian, D., Simsekli, U., & Durmus, A.O. (2025). Denoising Lévy Probabilistic Models. ICLR 2025
See paper | See slides | GitHub Repository

Piecewise Deterministic Generative Models

Published in The Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS), 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.

Bertazzi, A., Shariatian, D., Durmus, A.O., Simsekli, U., & Moulines, É. (2024). Piecewise deterministic generative models. NeurIPS 2024
See paper | GitHub Repository