DesignPref: Why AI needs your personal design taste
TL;DR
AI can now draft UIs and slides, but design taste is personal. DesignPref is a new dataset that captures individual preferences from professional designers—so models can learn your style, not the crowd’s.
- What it is: 12k pairwise comparisons of AI-generated UIs, rated by 20 pros with multi-level choices plus written rationales.
- Big finding: experts often disagree (Krippendorff’s alpha = 0.25) because they prioritize different aspects like hierarchy, layout, color, and whitespace.
- Why it matters: majority-vote “judge” models gloss over individual taste.
- What works better: personalization. Fine-tuning or RAG with designer-specific annotations predicts each person’s picks better—using up to 20× fewer examples.
- Why it’s new: the first dataset for personalized evaluation of visual design generation.
Paper: https://arxiv.org/abs/2511.20513v1
Authors: Yi-Hao Peng, Jeffrey P. Bigham, Jason Wu
Paper: https://arxiv.org/abs/2511.20513v1
Register: https://www.AiFeta.com
#GenerativeAI #UXDesign #UIDesign #AIDesign #Personalization #HumanCenteredAI #Dataset #DesignResearch
