Continual Diffusion with STAMINA: STack-And-Mask INcremental Adapters

arXiv:2311.18763v2 Announce Type: replace-cross
Abstract: Recent work has demonstrated a remarkable ability to customize text-to-image diffusion models to multiple, fine-grained concepts in a sequential (i.e., continual) manner while only providing a few example images for each concept. This setting is known as continual diffusion. Here, we ask the question: Can we scale these methods to longer concept sequences without forgetting? Although prior work mitigates the forgetting of previously learned concepts, we show that its capacity to learn new tasks reaches …

abstract arxiv concept concepts continual cs.ai cs.cv cs.lg diffusion diffusion models example fine-grained image image diffusion images incremental multiple question scale stack text text-to-image type while work