Figure 2: Overview of GenRevisit. Panel (a) presents the overall workflow. At each decision step, the multimodal planner selects the next action based on the decision context and, when editing, jointly selects a source image as the continuation state and a corresponding edit specification. The resulting image is verified and recorded in the interaction history to inform subsequent decisions, while submission returns the historical GM-best image. Panel (b) illustrates an editing example. The planner selects I₁, which is neither the latest candidate I₂ nor the GM-best candidate I₀, and corrects the backpack count while preserving the other constraints. The resulting candidate I₃ passes all constraint-level checks and becomes the new GM-best image.