Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.21375 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914057086304256 |
|---|---|
| author | Jelaca, Aleksa Jiao, Ying Tian, Chang Moens, Marie-Francine |
| author_facet | Jelaca, Aleksa Jiao, Ying Tian, Chang Moens, Marie-Francine |
| contents | Text-to-image generation has advanced rapidly with large-scale multimodal training, yet fine-grained controllability remains a critical challenge. Counterfactual controllability, defined as the capacity to deliberately generate images that contradict common-sense patterns, remains a major challenge but plays a crucial role in enabling creativity and exploratory applications. In this work, we address this gap with a focus on counterfactual size (e.g., generating a tiny walrus beside a giant button) and propose an automatic prompt engineering framework that adapts base prompts into revised prompts for counterfactual images. The framework comprises three components: an image evaluator that guides dataset construction by identifying successful image generations, a supervised prompt rewriter that produces revised prompts, and a DPO-trained ranker that selects the optimal revised prompt. We construct the first counterfactual size text-image dataset and enhance the image evaluator by extending Grounded SAM with refinements, achieving a 114 percent improvement over its backbone. Experiments demonstrate that our method outperforms state-of-the-art baselines and ChatGPT-4o, establishing a foundation for future research on counterfactual controllability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21375 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Automated Prompt Generation for Creative and Counterfactual Text-to-image Synthesis Jelaca, Aleksa Jiao, Ying Tian, Chang Moens, Marie-Francine Computer Vision and Pattern Recognition Artificial Intelligence Text-to-image generation has advanced rapidly with large-scale multimodal training, yet fine-grained controllability remains a critical challenge. Counterfactual controllability, defined as the capacity to deliberately generate images that contradict common-sense patterns, remains a major challenge but plays a crucial role in enabling creativity and exploratory applications. In this work, we address this gap with a focus on counterfactual size (e.g., generating a tiny walrus beside a giant button) and propose an automatic prompt engineering framework that adapts base prompts into revised prompts for counterfactual images. The framework comprises three components: an image evaluator that guides dataset construction by identifying successful image generations, a supervised prompt rewriter that produces revised prompts, and a DPO-trained ranker that selects the optimal revised prompt. We construct the first counterfactual size text-image dataset and enhance the image evaluator by extending Grounded SAM with refinements, achieving a 114 percent improvement over its backbone. Experiments demonstrate that our method outperforms state-of-the-art baselines and ChatGPT-4o, establishing a foundation for future research on counterfactual controllability. |
| title | Automated Prompt Generation for Creative and Counterfactual Text-to-image Synthesis |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2509.21375 |