"All AI Models Might Be the Same": AI models transform inputs like images and text into high-dimensional vectors called embeddings. Influenced by factors such as architecture, training data, and input modality, these embeddings are model-specific and not directly interoperable. In this talk, we show that—despite these fundamental differences—embeddings across models share significant common geometric structure. We then introduce vec2vec, a method that translates between embedding spaces by learning a universal semantic representation without paired training data or predefined mappings. This shared structure provides empirical support for the Platonic Representation Hypothesis and enables black-box inversion: reconstructing original inputs from vector representations alone. — Rishi D. Jha

"Personalized Recommendations without Inducing Congestion: Mitigating Disparities in the NYC High School Match": At scale, recommendations can be self-defeating: if they steer too many users toward the same items, then even users who were originally predicted to have a high chance of matching to an item may not, due to increased competition. In this talk, I will describe this phenomenon of recommendation-induced congestion, introduce an approach for dealing with it, and discuss the deployment of this approach in an informational intervention for the NYC High School Match. — Erica Chiang