cover

Holistic Evaluation of Text-to-Image Models: Human evaluation procedure

13 Oct 2024

This article outlines the human evaluation process for AI-generated images using Amazon Mechanical Turk and compliance with Human Subjects.

cover

A Deep Dive Into Stable Diffusion and Other Leading Text-to-Image Models

13 Oct 2024

Explore the latest advancements in text-to-image models like Stable Diffusion, DALL-E, and Dreamlike Diffusion.

cover

From Birdwatching to Fairness in Image Generation Models

12 Oct 2024

Discover how various benchmarks, including MS-COCO and DrawBench, are used to evaluate AI image generation models.

cover

Human vs. Machine: Evaluating AI-Generated Images Through Human and Automated Metrics

12 Oct 2024

Explore how human annotators assess AI-generated images based on alignment, quality, aesthetics, and originality.

cover

Holistic Evaluation of Text-to-Image Models: Datasheet

12 Oct 2024

Discover the HEIM benchmark—a new tool designed to holistically evaluate text-to-image models across 12 crucial aspects like quality, bias, and more.

cover

Holistic Evaluation of Text-to-Image Models: Author contributions, Acknowledgments and References

12 Oct 2024

Meet the authors whose contributions formed a holistic evaluation of text-to-image models.

cover

Limitations in AI Model Evaluation: Bias, Efficiency, and Human Judgment

12 Oct 2024

Explore the limits of current AI model evaluations.

cover

New Dimensions in Text-to-Image Model Evaluation

12 Oct 2024

Explore an holistic evaluation of image generation models sets new benchmarks for quality, ethics, aesthetics, and societal impact.

cover

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

12 Oct 2024

Discover HEIM, the groundbreaking benchmark assessing text-to-image models across 12 key aspects like quality, fairness, originality, and more.