English

ODIN: On-demand Data Formulation to Mitigate Dataset Lock-in

Machine Learning 2023-03-17 v2 Artificial Intelligence

Abstract

ODIN is an innovative approach that addresses the problem of dataset constraints by integrating generative AI models. Traditional zero-shot learning methods are constrained by the training dataset. To fundamentally overcome this limitation, ODIN attempts to mitigate the dataset constraints by generating on-demand datasets based on user requirements. ODIN consists of three main modules: a prompt generator, a text-to-image generator, and an image post-processor. To generate high-quality prompts and images, we adopted a large language model (e.g., ChatGPT), and a text-to-image diffusion model (e.g., Stable Diffusion), respectively. We evaluated ODIN on various datasets in terms of model accuracy and data diversity to demonstrate its potential, and conducted post-experiments for further investigation. Overall, ODIN is a feasible approach that enables Al to learn unseen knowledge beyond the training dataset.

Keywords

Cite

@article{arxiv.2303.06832,
  title  = {ODIN: On-demand Data Formulation to Mitigate Dataset Lock-in},
  author = {SP Choi and Jihun Lee and Hyeongseok Ahn and Sanghee Jung and Bumsoo Kang},
  journal= {arXiv preprint arXiv:2303.06832},
  year   = {2023}
}

Comments

15 pages

R2 v1 2026-06-28T09:13:21.052Z