-
MISIS / SBER AI / AIRI
- Moscow, Russia
-
23:50
(UTC +03:00) - levnovitskiy@gmail.com
- https://t.me/leffffffffffff
- in/lev-novitskiy-022289261
- https://t.me/mlball_days
Highlights
- Pro
Lists (11)
Sort Name ascending (A-Z)
Starred repositories
Just another reasonably minimal repo for class-conditional training of pixel-space diffusion transformers.
The official implementation of "Relay Diffusion: Unifying diffusion process across resolutions for image synthesis" [ICLR 2024 Spotlight]
Official Implementation for "Electrostatics from Laplacian Eigenbasis for Neural Network Interatomic Potentials"
Official implementation of "Latent Action Learning Requires Supervision in the Presence of Distractors", ICML 2025
[CVPR 2022] StyleGAN-V: A Continuous Video Generator with the Price, Image Quality and Perks of StyleGAN2
Official code of "Categorical Schrödinger Bridge Matching" paper
Inpaint images with ControlNet
[TITS2025] Pattern-Matching Dynamic Memory Network for Dual-Mode Traffic Prediction
[ACM MM 2024] Frame Interpolation with Consecutive Brownian Bridge Diffusion Model
Continuous Thought Machines, because thought takes time and reasoning is a process.
[ICML 2025] Official PyTorch Implementation of "History-Guided Video Diffusion"
[ICLR'25] ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation
Efficient vision foundation models for high-resolution generation and perception.
Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models
Visualization of DiT self attention features
Collection of publicly available Discrete Choice Datasets
The authors official implementation of Unsupervised Discovery of Interpretable Directions in the GAN Latent Space
PyTorch Implementation of AudioLCM (ACM-MM'24): a efficient and high-quality text-to-audio generation with latent consistency model.
A PyTorch native platform for training generative AI models
Flow Generator Matching (FGM), an innovative approach designed to accelerate the sampling of flow-matching models into a one-step generation.
[ICLR 2024] Code for our paper: GNRI: Lightning-Fast Image Inversion and Editing for Text-to-Image Diffusion Models