AI RESEARCH
Attention, May I Have Your Decision? Localizing Generative Choices in Diffusion Models
arXiv CS.CV
•
ArXi:2604.06052v1 Announce Type: new Text-to-image diffusion models exhibit remarkable generative capabilities, yet their internal operations remain opaque, particularly when handling prompts that are not fully descriptive. In such scenarios, models must make implicit decisions to generate details not explicitly specified in the text. This work investigates the hypothesis that this decision-making process is not diffuse but is computationally localized within the model's architecture.