Personalizing Incremental Video Search with Hybrid Text and ID Embeddings

Apple researchers developed a personalization system for incremental video search on Apple TV that combines text-based and ID-based embeddings to improve ranking, particularly for short, ambiguous prefix queries. In a three-week online experiment, the system achieved statistically significant gains of +1.14% tap-through rate and +1.23% conversion rate, with the largest improvements seen for users with longer watch histories.

content type paper /research/ published July 2026 Personalizing Incremental Video Search with Hybrid Text and ID Embeddings AuthorsVivek Kanojiya, Vishalaksh Aggarwal, Daeho Baek, Lyndon Kennedy, Xuetao Yin Personalizing Incremental Video Search with Hybrid Text and ID Embeddings AuthorsVivek Kanojiya, Vishalaksh Aggarwal, Daeho Baek, Lyndon Kennedy, Xuetao Yin Incremental video search requires high-quality ranking after each keystroke, where intent is often underspecified e.g., 1–3 character prefixes . We present a personalization system for Apple TV search that combines complementary semantic and collaborative signals at ranking time. Our approach learns two item embedding spaces: i a text-based multilingual encoder TextEmb fine-tuned on co-engagement triplets via contrastive learning, and ii an ID-based collaborative embedding model IdEmb trained on interaction-derived positives. At serving time, we construct user representations from recent watch history and inject text- and ID-based user–item cosine similarities into a pairwise XGBoost ranker. We evaluate the system with temporally held-out offline datasets and a three-week online controlled experiment. Offline, for sessions with user history, the personalized ranker improves NDCG@10 by 2.99% and MRR by 3.30% over the non-personalized baseline. Crucially, slice analyses show that personalization is most needed in incremental search, where intent is still forming: on ambiguous prefix queries 1–3 characters , NDCG@10 lift is +8.63%, versus only +1.46% on longer, more fully specified queries. Users with longer watch histories benefit more from personalization than newer users: NDCG lift rises from +2.13% for users with 1–5 history items to +4.37% for users with 51–100. This larger lift occurs even though baseline relevance is lower for long-history cohorts NDCG@10 drops from 0.733 to 0.680 , indicating that personalization adds the most value where default ranking underperforms. Online, treatment yields statistically significant gains of +1.14% tap-through rate and +1.23% conversion rate, with a 2.91% improvement in converted-item rank position. We further analyze coverage–precision trade-offs between semantic and collaborative embeddings through ablations isolating each signal, and evaluate embedding quality on a held-out corpus with LLM-judged similarity labels to reduce click/exposure bias. BayesCNS: A Unified Bayesian Approach to Address Cold Start and Non-Stationarity in Search Systems at Scale December 11, 2024 research area Knowledge Bases and Search /research/?domain=Knowledge%20Bases%20and%20Search , research area Methods and Algorithms /research/?domain=Methods%20and%20Algorithms conference AAAI /research/?event=AAAI Information Retrieval IR systems used in search and recommendation platforms frequently employ Learning-to-Rank LTR models to rank items in response to user queries. These models heavily rely on features derived from user interactions, such as clicks and engagement data. This dependence introduces cold start issues for items lacking user engagement and poses challenges in adapting to non-stationary shifts in user behavior over time. We… Consistent Collaborative Filtering via Tensor Decomposition August 16, 2023 research area Knowledge Bases and Search /research/?domain=Knowledge%20Bases%20and%20Search , research area Methods and Algorithms /research/?domain=Methods%20and%20Algorithms Collaborative filtering is the de facto standard for analyzing users’ activities and building recommendation systems for items. In this work we develop Sliced Anti-symmetric Decomposition SAD , a new model for collaborative filtering based on implicit feedback. In contrast to traditional techniques where a latent representation of users user vectors and items item vectors are estimated, SAD introduces one additional latent vector to each…