AI RESEARCH
Editor's Choice: Evaluating Abstract Intent in Image Editing through Atomic Entity Analysis
arXiv CS.CV
•
ArXi:2605.14842v1 Announce Type: new Humans naturally communicate through abstract concepts like "mood". However, current image editing benchmarks focus primarily on explicit, literal commands, leaving abstract instructions largely underexplored. In this work, we first formalize the definition and taxonomy of abstract image editing. To measure instruction-following in this challenging domain, we