VisAlign is a dataset for measuring AI-human visual alignment in terms of image classification, a fundamental task in machine perception. In order to evaluate AI-Human visual alignment, a dataset should encompass samples with various scenarios that may arise in the real world and have gold human perception labels. The 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.
Paper | Code | Results | Date | Stars |
---|