The stunning qualitative improvement of recent text-to-image models has led to their widespread attention and adoption. However, we lack a comprehensive quantitative understanding of their capabilities and risks. To fill this gap, we introduce a new benchmark, Holistic Evaluation of Text-to-Image Models (HEIM). Whereas previous evaluations focus mostly on text-image alignment and image quality, we identify 12 aspects, including text-image alignment, image quality, aesthetics, originality, reasoning, knowledge, bias, toxicity, fairness, robustness, multilinguality, and efficiency. We curate 62 scenarios encompassing these aspects and evaluate 26 state-of-the-art text-to-image models on this benchmark. Our results reveal that no single model excels in all aspects, with different models demonstrating different strengths. We release the generated images and human evaluation results for full transparency at https://crfm.stanford.edu/heim/v1.1.0 and the code at https://github.com/stanford-crfm/helm, which is integrated with the HELM codebase.
@article{arxiv.2311.04287,
title = {Holistic Evaluation of Text-To-Image Models},
author = {Tony Lee and Michihiro Yasunaga and Chenlin Meng and Yifan Mai and Joon Sung Park and Agrim Gupta and Yunzhi Zhang and Deepak Narayanan and Hannah Benita Teufel and Marco Bellagente and Minguk Kang and Taesung Park and Jure Leskovec and Jun-Yan Zhu and Li Fei-Fei and Jiajun Wu and Stefano Ermon and Percy Liang},
journal= {arXiv preprint arXiv:2311.04287},
year = {2023}
}
Comments
NeurIPS 2023. First three authors contributed equally