Selective routing improves novelty in personalized recommendations
Routing Between Generative and Collaborative User Profiles: A Serving-Time Gate for Controllable Novelty
Information Retrieval
Summary
Making recommendations more interesting without losing accuracy is tricky. The authors studied how to use smart user profiles made by large language models (LLMs) only when it really helps. They built a system that decides in real time if a user should see recommendations based on an LLM profile or a usual collaborative method. This system boosts the chance of showing new content without making recommendations worse overall.
What this means in practice
- •For streaming service engineers: Optimize movie, TV, and sports recommendations to include more novel choices without losing overall relevance by selectively using generated user profiles.$Commercial implications: Enables streaming platforms to sell more engaging personalized recommendations that balance novelty and accuracy using selective generative profiling.
- •For e-commerce recommendation teams: Implement routing systems to selectively apply generative user profiles, improving discovery of novel products while maintaining purchase likelihood.
Tested on one dataset.
Authors
Milad Sabouri, Neeraj Sharma, Sardar Hamidian, Shaghayegh Agah
Abstract
Large language models (LLMs) enable rich semantic user profiles for recommendation, but such profiles are more expensive to generate and are not necessarily desirable to deploy uniformly. We study whether LLM-generated profiles can instead be invoked selectively within a production recommendation pipeline. Using a real-world streaming dataset covering movies, TV shows, and sports content, we train a serving-time routing gate that assigns each user to either a collaborative sequential recommendation model or a recommendation model driven by an LLM-generated profile. The gate uses only serving-time features and learns to identify users for whom profile-based routing can increase Novelty@10 while preserving ranking relevance. A routing threshold controls how aggressively users are sent to the generative model, exposing a tunable novelty--relevance trade-off. At an overall NDCG-loss budget of 5\%, the learned gate increases Novelty@10 by 6.5\% while routing 12.5\% of users, outperforming simple heuristic and random routing policies at comparable relevance cost. These results show that LLM-generated user profiles can serve as a controllable complement to collaborative recommendation, while results with non-generative semantic profiles indicate that the benefit stems from selective routing rather than LLM generation alone.