18 papers
Rethinking Pixel Mean Flows via Interval Denoiser
Alexander Zaytsev, Dmitry Baranchuk, Alexander Korotin +1
Modern diffusion and flow-based models are increasingly moving toward few-step, latent-free generation to bypass the computational overhead of multi-step sampling and the reconstru…
DifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing
Artyom Mazur, Nina Konovalova, Aibek Alanov
Mechanistic interpretability seeks to explain neural network behavior by decomposing model computations into interpretable features and circuits. While transcoder-based circuit tra…
ReCache: Learning Budget-Aware Caching Schedules for Diffusion Models via REINFORCE
Mishan Aliev, Eva Neudachina, Ilya Bykov +4
Modern diffusion models generate high-quality images and videos, but their iterative denoising process makes inference expensive. Feature caching accelerates sampling by reusing or…
ATATA: One Algorithm to Align Them All
Boyi Pang, Savva Ignatyev, Vladimir Ippolitov +8
We suggest a new multi-modal algorithm for joint inference of paired structurally aligned samples with Rectified Flow models. While some existing methods propose a codependent gene…
SHIFT: Steering Hidden Intermediates in Flow Transformers
Nina Konovalova, Andrey Kuznetsov, Aibek Alanov
Diffusion models have become leading approaches for high-fidelity image generation. Recent DiT-based diffusion models, in particular, achieve strong prompt adherence while producin…
OrthoFuse: Training-free Riemannian Fusion of Orthogonal Style-Concept Adapters for Diffusion Models
Ali Aliev, Kamil Garifullin, Nikolay Yudin +5
In a rapidly growing field of model training there is a constant practical interest in parameter-efficient fine-tuning and various techniques that use a small amount of training da…