AI RESEARCH

UniPinRec: Unifying Generative Retrieval and Ranking at Pinterest Scale

arXiv CS.LG

ArXi:2606.00422v1 Announce Type: cross Modern recommendation systems predominantly train retrieval and ranking as separate models despite both increasingly relying on large transformers encoding the same user behavior data, duplicating parameters, compute, and serving cost. Prior work unifies the model architecture but not the full pipeline: input formats