Skip to content
Book Open access

MoR: An Adaptive Retrieval Allocation Balancing Long-Term Interest and Short-Term Evidence

Sep 2026 · Proceedings of the 20th ACM Conference on Recommender Systems · pp. 1307-1313 · 0 citations · 9 references

Abstract

Modern recommender systems often rely on multiple retrieval sources to generate candidate items, yet determining optimal quota allocation across these sources remains challenging. We present Mixture of Retrieval (MoR) bandit, a novel framework that dynamically optimizes retrieval allocation by balancing long-term user interests with short-term behavioral signals. Unlike traditional static approaches, our method employs a modified Thompson Sampling algorithm that combines a user’s affinity scores to third-party video channels with their recent behavioral signals, using exponential decay to prevent over-reliance on stale signals. The framework enhances personalization while promoting exploration of potentially valuable but under-exposed content channels. In a large-scale A/B test on Prime Video, MoR bandit achieved a significant 4% increase in third-party channel subscriptions and substantial improvements in user engagement metrics. At the time of writing this paper, MoR has been fully launched in production worldwide serving hundreds of millions of customers. Our approach provides a generalizable solution for multi-source retrieval optimization in large-scale recommender systems.

Read PDF

Similar papers

Book Open access Nov 2026

Personalized Quality-Aware Reranking for Long-Term User Retention in E-commerce

Modern recommendation systems are predominantly optimized for short-term engagement signals such as Click-Through Rate (CTR) and Conversion Rate (CVR), often overlooking post-purchase user satisfaction. As a result, ranking systems can repeatedly promote products that drive immediate interactions but lead to poor user...

Siddhant Doshi, Krish Rupapara, Bhavuk Singhal et al. · 0 citations
Book Open access Sep 2026

UNIQUE: A Unified Retrieval and Ranking System for Large-Scale Feed Recommendation

Results show that UNIQUE provides a stable, efficient, and production-ready framework for unified retrieval and ranking in industrial recommendation, and a balanced quantization mechanism is further introduced to mitigate codebook imbalance and improve long-tail representation.

Zhuang-Chen-Ying-Ying Liu, Yong-Kang Fu, Zuo-Dong Yang et al. · 0 citations
#artificial intelligence Book Open access Oct 2026

LLM-based User-Product Relevance Evaluation for Dynamic Product Advertising

This work proposes an agentic, human-aligned prompting framework that generates multi-category relevance guidelines and distills teacher-model judgments into an efficient student model for scalable evaluation of DPA retrieval systems, enabling faster and more reliable DPA retrieval iteration before A/B testing.

Wen Zhang, Xiao Bai, Hsiang-Wei Chao et al. · 0 citations
Book Open access Nov 2026

Contextual Ranker in Production: Session-Aware Similar Item Ranking for Lower-Funnel E-Commerce

This talk presents Contextual Ranker, a production ranking system for eBay's ''Similar Items'' module on product detail pages. The problem sits at the lower funnel of e-commerce, where users are evaluating whether to purchase the current listing that they are viewing (the seed item ) or switch to an alternative. In thi...

Yu-Chen Guo, Prudhvi Raj Dachapally · 0 citations
Book Open access Sep 2026

A Redundancy Reduction Approach for Controllable Sequential Recommendations

This work proposes a decorrelation-regularized training framework that augments next-item prediction with an auxiliary redundancy-reduction term, and instantiate it with BT-SR, which uses the Barlow Twins objective to form label-consistent positive pairs without synthetic corruptions.

Veronika Ivanova, Marina Munkhoeva, Ivan Razvorotnev et al. · 0 citations
Sep 2026

Evaluating Brand Retrieval and Ranking in Large Language Model Recommendations

Large language models (LLMs) are increasingly used for product recommendation, but evaluating their recommendations presents challenges that differ from conventional information retrieval and recommender systems. LLMs can generate recommendations without an explicit candidate set, and repeated responses to the same que...

E. Malthouse, Kun-Yu Lee, Jing Yang et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.