Dynamic VLM-Guided Negative Prompting for Diffusion Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915586715418624 |
|---|---|
| author | Chang, Hoyeon Kim, Seungjin Choi, Yoonseok |
| author_facet | Chang, Hoyeon Kim, Seungjin Choi, Yoonseok |
| contents | We propose a novel approach for dynamic negative prompting in diffusion models that leverages Vision-Language Models (VLMs) to adaptively generate negative prompts during the denoising process. Unlike traditional Negative Prompting methods that use fixed negative prompts, our method generates intermediate image predictions at specific denoising steps and queries a VLM to produce contextually appropriate negative prompts. We evaluate our approach on various benchmark datasets and demonstrate the trade-offs between negative guidance strength and text-image alignment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_26052 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dynamic VLM-Guided Negative Prompting for Diffusion Models Chang, Hoyeon Kim, Seungjin Choi, Yoonseok Computer Vision and Pattern Recognition Artificial Intelligence We propose a novel approach for dynamic negative prompting in diffusion models that leverages Vision-Language Models (VLMs) to adaptively generate negative prompts during the denoising process. Unlike traditional Negative Prompting methods that use fixed negative prompts, our method generates intermediate image predictions at specific denoising steps and queries a VLM to produce contextually appropriate negative prompts. We evaluate our approach on various benchmark datasets and demonstrate the trade-offs between negative guidance strength and text-image alignment. |
| title | Dynamic VLM-Guided Negative Prompting for Diffusion Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2510.26052 |