1. Social Signals Predict Contemporary Art Prices Better Than Visual Features, Particularly in Emerging Markets
Our recent publication, “Social Signals Predict Contemporary Art Prices Better Than Visual Features, Particularly in Emerging Markets,” explores the determinants of art valuation in global markets. Utilizing a comprehensive dataset of contemporary art auctions, our research reveals that social factors such as artist reputation and market conditions significantly outperform visual characteristics in predicting art prices. This study highlights the profound influence of social constructs on art valuation, particularly in emerging markets where expert knowledge is limited.
2. Labor Space: Unifying Representation of the Labor Market via Large Language Models
Our groundbreaking paper, “Labor Space: Unifying Representation of the Labor Market via Large Language Models,” introduces a novel vector-space embedding of the labor market. Authored by Seongwoon Kim, an esteemed alumni who graduated this spring, along with Yong-Yeol Ahn and Jaehyuk Park, this study leverages Google’s BERT model to create a comprehensive framework that maps the intricate relationships among various labor market entities, including skills, occupations, industries, and firms. This unified representation facilitates a deeper understanding of the labor market, enabling more effective analysis and strategic decision-making.
