Prompt Evolution for Generative AI: A Classifier-Guided Approach
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911589525880832 |
|---|---|
| author | Wong, Melvin Ong, Yew-Soon Gupta, Abhishek Bali, Kavitesh K. Chen, Caishun |
| author_facet | Wong, Melvin Ong, Yew-Soon Gupta, Abhishek Bali, Kavitesh K. Chen, Caishun |
| contents | Synthesis of digital artifacts conditioned on user prompts has become an important paradigm facilitating an explosion of use cases with generative AI. However, such models often fail to connect the generated outputs and desired target concepts/preferences implied by the prompts. Current research addressing this limitation has largely focused on enhancing the prompts before output generation or improving the model's performance up front. In contrast, this paper conceptualizes prompt evolution, imparting evolutionary selection pressure and variation during the generative process to produce multiple outputs that satisfy the target concepts/preferences better. We propose a multi-objective instantiation of this broader idea that uses a multi-label image classifier-guided approach. The predicted labels from the classifiers serve as multiple objectives to optimize, with the aim of producing diversified images that meet user preferences. A novelty of our evolutionary algorithm is that the pre-trained generative model gives us implicit mutation operations, leveraging the model's stochastic generative capability to automate the creation of Pareto-optimized images more faithful to user preferences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2305_16347 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Prompt Evolution for Generative AI: A Classifier-Guided Approach Wong, Melvin Ong, Yew-Soon Gupta, Abhishek Bali, Kavitesh K. Chen, Caishun Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Neural and Evolutionary Computing I.2 Synthesis of digital artifacts conditioned on user prompts has become an important paradigm facilitating an explosion of use cases with generative AI. However, such models often fail to connect the generated outputs and desired target concepts/preferences implied by the prompts. Current research addressing this limitation has largely focused on enhancing the prompts before output generation or improving the model's performance up front. In contrast, this paper conceptualizes prompt evolution, imparting evolutionary selection pressure and variation during the generative process to produce multiple outputs that satisfy the target concepts/preferences better. We propose a multi-objective instantiation of this broader idea that uses a multi-label image classifier-guided approach. The predicted labels from the classifiers serve as multiple objectives to optimize, with the aim of producing diversified images that meet user preferences. A novelty of our evolutionary algorithm is that the pre-trained generative model gives us implicit mutation operations, leveraging the model's stochastic generative capability to automate the creation of Pareto-optimized images more faithful to user preferences. |
| title | Prompt Evolution for Generative AI: A Classifier-Guided Approach |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Neural and Evolutionary Computing I.2 |
| url | https://arxiv.org/abs/2305.16347 |