Replace in Translation: Boost Concept Alignment in Counterfactual Text-to-Image
This addresses the problem of generating plausible counterfactual scenes for AI-generated content users, though it appears incremental as it builds on existing controllable T2I models.
The paper tackles concept alignment in counterfactual text-to-image generation by proposing a strategy to replace objects in synthesized images step-by-step using Explicit Logical Narrative Prompt (ELNP) generated by DeepSeek, which boosts concept alignment as demonstrated through experiments and a new evaluation metric.
Text-to-Image (T2I) has been prevalent in recent years, with most common condition tasks having been optimized nicely. Besides, counterfactual Text-to-Image is obstructing us from a more versatile AIGC experience. For those scenes that are impossible to happen in real world and anti-physics, we should spare no efforts in increasing the factual feel, which means synthesizing images that people think very likely to be happening, and concept alignment, which means all the required objects should be in the same frame. In this paper, we focus on concept alignment. As controllable T2I models have achieved satisfactory performance for real applications, we utilize this technology to replace the objects in a synthesized image in latent space step-by-step to change the image from a common scene to a counterfactual scene to meet the prompt. We propose a strategy to instruct this replacing process, which is called as Explicit Logical Narrative Prompt (ELNP), by using the newly SoTA language model DeepSeek to generate the instructions. Furthermore, to evaluate models' performance in counterfactual T2I, we design a metric to calculate how many required concepts in the prompt can be covered averagely in the synthesized images. The extensive experiments and qualitative comparisons demonstrate that our strategy can boost the concept alignment in counterfactual T2I.