Tackling Structural Hallucination in Image Translation with Local Diffusion
This addresses a critical issue in medical imaging and image translation by mitigating hallucination risks, though it is incremental as it builds on existing diffusion models.
The paper tackles the problem of structural hallucination in image translation, where diffusion models struggle with out-of-distribution regions like unseen tumors, by proposing a training-free framework with Local Diffusion processes. The result is a reduction in misdiagnosis by 40% in medical datasets and 25% in natural image datasets.
Recent developments in diffusion models have advanced conditioned image generation, yet they struggle with reconstructing out-of-distribution (OOD) images, such as unseen tumors in medical images, causing "image hallucination" and risking misdiagnosis. We hypothesize such hallucinations result from local OOD regions in the conditional images. We verify that partitioning the OOD region and conducting separate image generations alleviates hallucinations in several applications. From this, we propose a training-free diffusion framework that reduces hallucination with multiple Local Diffusion processes. Our approach involves OOD estimation followed by two modules: a "branching" module generates locally both within and outside OOD regions, and a "fusion" module integrates these predictions into one. Our evaluation shows our method mitigates hallucination over baseline models quantitatively and qualitatively, reducing misdiagnosis by 40% and 25% in the real-world medical and natural image datasets, respectively. It also demonstrates compatibility with various pre-trained diffusion models.