Diffusion models have demonstrated remarkable success in high-fidelity image synthesis and prompt-guided generative modeling. However, ensuring adequate diversity in generated samples of prompt-guided diffusion models remains a challenge, particularly when the prompts span a broad semantic spectrum and the diversity of generated data needs to be evaluated in a prompt-aware fashion across semantically similar prompts. Recent methods have introduced guidance via diversity measures to encourage more varied generations.
We extend the diversity measure-based approaches by proposing the
Scalable Prompt-Aware Renyi Kernel
Entropy Diversity Guidance (SPARKE) method for prompt-aware diversity guidance.
SPARKE utilizes conditional entropy for diversity guidance, which dynamically conditions diversity measurement on similar prompts and enables prompt-aware diversity control.
While the entropy-based guidance approach enhances prompt-aware diversity, its reliance on the matrix-based entropy scores poses computational challenges in large-scale generation settings.
To address this, we focus on the special case of Conditional latent RKE Score Guidance, reducing entropy computation and gradient-based optimization complexity from
𝒪(n³)
of general entropy measures to 𝒪(n)
. The reduced computational complexity allows for diversity-guided sampling over potentially thousands of generation rounds on different prompts.
We numerically test the SPARKE method on several text-to-image diffusion models, demonstrating that the proposed method improves the prompt-aware diversity of the generated data without incurring significant computational costs.
To cite this work, please use the following BibTeX entries:
SPARKE Diversity Guidance:
@article{jalali2025sparke,
author = {Mohammad Jalali and Haoyu Lei and Amin Gohari and Farzan Farnia},
title = {SPARKE: Scalable Prompt-Aware Diversity Guidance in Diffusion Models via RKE Score},
journal = {arXiv preprint arXiv:2506.10173},
year = {2025},
url = {https://arxiv.org/abs/2506.10173},
}
RKE Score:
@inproceedings{jalali2023rke,
author = {Jalali, Mohammad and Li, Cheuk Ting and Farnia, Farzan},
booktitle = {Advances in Neural Information Processing Systems},
pages = {9931--9943},
title = {An Information-Theoretic Evaluation of Generative Models in Learning Multi-modal Distributions},
url = {https://openreview.net/forum?id=PdZhf6PiAb},
volume = {36},
year = {2023}
}