Paving the Way for Better AI Models: Insights from HEIM’s 12-Aspect Benchmark

cover
12 Oct 2024

Authors:

(1) Tony Lee, Stanford with Equal contribution;

(2) Michihiro Yasunaga, Stanford with Equal contribution;

(3) Chenlin Meng, Stanford with Equal contribution;

(4) Yifan Mai, Stanford;

(5) Joon Sung Park, Stanford;

(6) Agrim Gupta, Stanford;

(7) Yunzhi Zhang, Stanford;

(8) Deepak Narayanan, Microsoft;

(9) Hannah Benita Teufel, Aleph Alpha;

(10) Marco Bellagente, Aleph Alpha;

(11) Minguk Kang, POSTECH;

(12) Taesung Park, Adobe;

(13) Jure Leskovec, Stanford;

(14) Jun-Yan Zhu, CMU;

(15) Li Fei-Fei, Stanford;

(16) Jiajun Wu, Stanford;

(17) Stefano Ermon, Stanford;

(18) Percy Liang, Stanford.

Abstract and 1 Introduction

2 Core framework

3 Aspects

4 Scenarios

5 Metrics

6 Models

7 Experiments and results

8 Related work

9 Conclusion

10 Limitations

Author contributions, Acknowledgments and References

A Datasheet

B Scenario details

C Metric details

D Model details

E Human evaluation procedure

9 Conclusion

We introduced Holistic Evaluation of Text-to-Image Models (HEIM), a new benchmark to assess 12 important aspects in text-to-image generation, including alignment, quality, aesthetics, originality, reasoning, knowledge, bias, toxicity, fairness, robustness, multilinguality, and efficiency. Our evaluation of 26 recent text-to-image models reveals that different models excel in different aspects, opening up research avenues to study whether and how to develop models that excel across multiple aspects. To enhance transparency and reproducibility, we release our evaluation pipeline, along with the generated images and human evaluation results. We encourage the community to consider the different aspects when developing text-to-image models.

This paper is available on arxiv under CC BY 4.0 DEED license.