Human Evaluation of Text-to-Image Models on a Multi-Task Benchmark
This provides a structured evaluation framework for researchers and developers in AI and vision-language understanding, though it is incremental as it builds on existing benchmarking efforts.
The authors tackled the problem of evaluating text-to-image models by introducing a new multi-task benchmark with human ratings, finding that models like Stable Diffusion and DALL-E 2 were compared across 3,600 ratings, revealing struggles with complex prompts involving multiple objects and attributes.
We provide a new multi-task benchmark for evaluating text-to-image models. We perform a human evaluation comparing the most common open-source (Stable Diffusion) and commercial (DALL-E 2) models. Twenty computer science AI graduate students evaluated the two models, on three tasks, at three difficulty levels, across ten prompts each, providing 3,600 ratings. Text-to-image generation has seen rapid progress to the point that many recent models have demonstrated their ability to create realistic high-resolution images for various prompts. However, current text-to-image methods and the broader body of research in vision-language understanding still struggle with intricate text prompts that contain many objects with multiple attributes and relationships. We introduce a new text-to-image benchmark that contains a suite of thirty-two tasks over multiple applications that capture a model's ability to handle different features of a text prompt. For example, asking a model to generate a varying number of the same object to measure its ability to count or providing a text prompt with several objects that each have a different attribute to identify its ability to match objects and attributes correctly. Rather than subjectively evaluating text-to-image results on a set of prompts, our new multi-task benchmark consists of challenge tasks at three difficulty levels (easy, medium, and hard) and human ratings for each generated image.