Abstract: Described herein is a technique to facilitate filtering during candidate retrieval stage performed by an information retrieval system that utilizes embedding models. An aNN indexing structure is created for each end-user, and in some instances, each activity type. This allows a single request for candidate content items to invoke a single process to obtain content items that satisfy the filtering criteria (e.g., in this case, the in-network requirement) from the separate per-end-user indexes.

Abstract: Described herein is a technique to facilitate filtering during candidate retrieval stage performed by an information retrieval system that utilizes embedding models. An aNN indexing structure is created for each end-user, and in some instances, each activity type. This allows a single request for candidate content items to invoke a single process to obtain content items that satisfy the filtering criteria (e.g., in this case, the in-network requirement) from the separate per-end-user indexes.

Abstract: Techniques for optimizing machine-learned models based on dwell time of network-transmitted content items are provided. In one technique, impression data and selection data are used train a selection prediction model. For each impression, a dwell time associated with that impression is determined and compared to a skip time. If the dwell time is less than the skip time, then a first training label that indicates that the impression is skipped is associated with the impression. If the dwell time is greater than the skip time, then a second training label that indicates that the impression is not skipped is associated with the impression. These training labels are used to train a skip prediction model. The selection prediction model and the skip prediction model are used in a content item selection event to generate a score for each candidate content item. The scores are used to select a content item.