article

Capacitor Chain Field Matching for Electrostatic Generative Modeling

Daniil Shlenskii, S. I. Manukhov, Alexander Kolesov, Alexander Varlamov, Nazar Buzun, V. V. Palyulin, Alexander Korotin

Equilibrium Diffusion Language Models

Denis Basharin, Mikhail Konovalov, Alexander Korotin, Nikita Gushchin

Discrete Action Matching: Learning Stochastic Dynamics from Samples via State Graphs

Learning population dynamics from unpaired temporal marginals is an ill-posed inverse problem that requires structural assumptions on …

Mikhail Persiianov, Alexander Korotin

One-Step Generation via Riemannian Wasserstein Gradient Flows

Recently, Drifting Models and Wasserstein Gradient Flows have attracted substantial attention because they move iterative …

David Li, Chanhyuk Lee, Jaehoon Yoo, Nikita Gushchin, Floor Eijkelboom, Eric Moulines, Maxim Panov, Alexander Korotin, Seunghoon Hong, Jinwoo Kim

IDRF: Inverse-Distilled Reward Fine-tuning of Masked Discrete Diffusion Models

Masked discrete diffusion models offer a promising alternative to autoregressive generation, but iterative sampling can be costly, and …

Vladislav Gromadskii, David Li, Samson Gourevitch, Yazid Janati, Eric Moulines, Maxim Panov, Alexander Korotin

Light Entropic Optimal Transport on Riemannian Manifolds

Entropic Optimal Transport (EOT) has become a practical framework for learning stochastic couplings between complex distributions, with …

Xavier Aramayo Carrasco, Petr Mokrov, Alexander Korotin

Alpha Diffusion Language Models: Factorization Alone Is Not the Problem

Discrete diffusion language models can generate multiple tokens in parallel, but reducing the number of denoising steps can lead to …

Nikita Gushchin, Dmitry Baranchuk, Alexander Korotin

E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models

Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse …

Arseny Ivanov, Alexander Kolesov, Alexander Korotin, Ivan Oseledets, Mikhail Goncharov

Simultaneous Neural Optimal Transport

Optimal Transport (OT) provides a principled framework for learning transformations between probability distributions from unpaired …

Milena Gazdieva, Kirill Sokolov, Jiawei Chen, Evgeny Burnaev, Alexander Korotin

Data Unlearning via Inverse Distillation

Multi-step matching models, including flow and diffusion models, produce high-quality outputs but incur substantial inference costs and …

Aleksei Leonov, Nikita Kornilov, Zhenhe Zhang, Evgeny Burnaev, Iaroslav Koshelev, Alexander Korotin