VisAlign: Dataset for Measuring the Alignment between AI and Humans in Visual Perception
Jiyoung Lee, Seungho Kim, Seunghyun Won, Joonseok Lee, Marzyeh Ghassemi, James Thorne, Jaeseok Choi, O-Kil Kwon
Abstract
AI alignment refers to models acting towards human-intended goals, preferences, or ethical principles. Analyzing the similarity between models and humans can be a proxy measure for ensuring AI safety. In this paper, we focus on the models' visual perception alignment with humans, further referred to as AI-human visual alignment. Specifically, we propose a new dataset for measuring AI-human visual alignment in terms of image classification. In order to evaluate AI-human visual alignment, a dataset should encompass samples with various scenarios and have gold human perception labels. Our dataset consists of three groups of samples, namely Must-Act (i.e., Must-Classify), Must-Abstain, and Uncertain, based on the quantity and clarity of visual information in an image and further divided into eight categories. All samples have a gold human perception label; even Uncertain (e.g., severely blurry) sample labels were obtained via crowd-sourcing. The validity of our dataset is verified by sampling theory, statistical theories related to survey design, and experts in the related fields. Using our dataset, we analyze the visual alignment and reliability of five popular visual perception models and seven abstention methods. Our code and data is available at https://github.com/jiyounglee-0523/VisAlign.
BibTeX
@inproceedings{
lee2023visalign,
title={VisAlign: Dataset for Measuring the Alignment between {AI} and Humans in Visual Perception},
author={Jiyoung Lee and Seungho Kim and Seunghyun Won and Joonseok Lee and Marzyeh Ghassemi and James Thorne and Jaeseok Choi and O-Kil Kwon and Edward Choi},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
year={2023},
url={https://openreview.net/forum?id=Wz2BJNQlyI}
}