RAT: Retrieval-Augmented Transformer for Click-Through Rate Prediction

arXiv:2404.02249v1 Announce Type: cross
Abstract: Predicting click-through rates (CTR) is a fundamental task for Web applications, where a key issue is to devise effective models for feature interactions. Current methodologies predominantly concentrate on modeling feature interactions within an individual sample, while overlooking the potential cross-sample relationships that can serve as a reference context to enhance the prediction. To make up for such deficiency, this paper develops a Retrieval-Augmented Transformer (RAT), aiming to acquire fine-grained feature interactions within and across samples. …

arxiv click cs.ai cs.ir cs.lg cs.si prediction rat rate retrieval retrieval-augmented through transformer type