Adding Conditional Control to Diffusion Models with Reinforcement Learning

doi:doi:10.57702/rwgr1k3a

Adding Conditional Control to Diffusion Models with Reinforcement Learning

Diffusion models are powerful generative models that allow for precise control over the characteristics of the generated samples. While these diffusion models trained on large datasets have achieved success, there is often a need to introduce additional controls in downstream fine-tuning processes, treating these powerful models as pre-trained diffusion models.

BibTex:

@dataset{Yulai_Zhao_and_Masatoshi_Uehara_and_Gabriele_Scalia_and_Tommaso_Biancalani_and_Sergey_Levine_and_Ehsan_Hajiramezanali_2024,
    abstract = {Diffusion models are powerful generative models that allow for precise control over the characteristics of the generated samples. While these diffusion models trained on large datasets have achieved success, there is often a need to introduce additional controls in downstream fine-tuning processes, treating these powerful models as pre-trained diffusion models.},
    author = {Yulai Zhao and Masatoshi Uehara and Gabriele Scalia and Tommaso Biancalani and Sergey Levine and Ehsan Hajiramezanali},
    doi = {10.57702/rwgr1k3a},
    institution = {No Organization},
    keyword = {'conditional control', 'diffusion models', 'reinforcement learning'},
    month = {dec},
    publisher = {TIB},
    title = {Adding Conditional Control to Diffusion Models with Reinforcement Learning},
    url = {https://service.tib.eu/ldmservice/dataset/adding-conditional-control-to-diffusion-models-with-reinforcement-learning},
    year = {2024}
}