5 papers
From Unlearning to UNBRANDING: A Benchmark for Trademark-Safe Text-to-Image Generation
Dawid Malarz, Filip Manjak, Maciej ZiÄba +2
The rapid progress of text-to-image diffusion models raises significant concerns regarding the unauthorized reproduction of trademarked content. While prior work targets general co…
UnGuide: Learning to Forget with LoRA-Guided Diffusion Models
Agnieszka Polowczyk, Alicja Polowczyk, Dawid Malarz +4
Recent advances in large-scale text-to-image diffusion models have heightened concerns about their potential misuse, especially in generating harmful or misleading content. This un…
PALATE: Peculiar Application of the Law of Total Expectation to Enhance the Evaluation of Deep Generative Models
Tadeusz Dziarmaga, Marcin KÄ dzioÅka, Artur Kasymov +1
Deep generative models (DGMs) have caused a paradigm shift in the field of machine learning, yielding noteworthy advancements in domains such as image synthesis, natural language p…
Classifier-free Guidance with Adaptive Scaling
Dawid Malarz, Artur Kasymov, Maciej ZiÄba +2
Classifier-free guidance (CFG) is an essential mechanism in contemporary text-driven diffusion models. In practice, in controlling the impact of guidance we can see the trade-off b…
AutoLoRA: AutoGuidance Meets Low-Rank Adaptation for Diffusion Models
Artur Kasymov, Marcin Sendera, MichaÅ StypuÅkowski +2
Low-rank adaptation (LoRA) is a fine-tuning technique that can be applied to conditional generative diffusion models. LoRA utilizes a small number of context examples to adapt the…