10 papers
Attention, May I Have Your Decision? Localizing Generative Choices in Diffusion Models
Katarzyna Zaleska, Åukasz Popek, Monika WysoczaÅska +1
Text-to-image diffusion models exhibit remarkable generative capabilities, yet their internal operations remain opaque, particularly when handling prompts that are not fully descri…
TADA! Tuning Audio Diffusion Models through Activation Steering
Åukasz Staniszewski, Katarzyna Zaleska, Mateusz Modrzejewski +1
Audio diffusion models can synthesize high-fidelity music from text, yet achieving fine-grained control over specific musical attributes remains challenging, as their internal mech…
Precise Parameter Localization for Textual Generation in Diffusion Models
Åukasz Staniszewski, Bartosz CywiÅski, Franziska Boenisch +2
Novel diffusion models can synthesize photo-realistic images with integrated high-quality text. Surprisingly, we demonstrate through attention activation patching that only less th…
Membership and Dataset Inference Attacks on Large Audio Generative Models
Jakub Proboszcz, PaweÅ Kochanski, Karol Korszun +5
Generative audio models, based on diffusion and autoregressive architectures, have advanced rapidly in both quality and expressiveness. This progress, however, raises pressing copy…
ELROND: Exploring and decomposing intrinsic capabilities of diffusion models
PaweÅ SkierÅ, Tomasz TrzciÅski, Kamil Deja
A single text prompt passed to a diffusion model often yields a wide range of visual outputs determined solely by stochastic process, leaving users with no direct control over whic…
Joint Diffusion models in Continual Learning
PaweÅ SkierÅ, Kamil Deja
In this work, we introduce JDCL - a new method for continual learning with generative rehearsal based on joint diffusion models. Neural networks suffer from catastrophic forgetting…