User context type affects streaming recommendations for habitual and exploratory listeners
When LLM-Inferred User Context Adds Value in Production Streaming Recommendation
Information Retrieval
Summary
Online streaming platforms use information about what users watch or listen to recommend new content. This paper shows that large language models (LLMs) can create summaries of a user’s past actions to help recommend new content, but these work best only for users who explore new types of content. For people who mostly stick to familiar genres or artists, traditional summary profiles are better. The results suggest that recommendation systems should adapt which user profile style they use based on how users behave.
What this means in practice
- •For streaming platform engineers: Design recommendation systems that switch between profile types depending on whether users are habitual or exploratory to improve user experience.
- •For video game platform developers: Implement adaptive user profiling methods to tailor game recommendations based on player behavior patterns to increase engagement.
Authors
Milad Sabouri, Neeraj Sharma, Sardar Hamidian, Shaghayegh Agah
Abstract
Contextual information in recommender systems is shifting from static, predefined variables toward latent representations inferred from behavior. Large language models support this shift by rendering an unstructured interaction history as a natural-language summary, which yields a thematic user context that can be encoded and used in place of an aggregate profile. The conditions under which such generated profiles outperform aggregate embeddings have received limited characterization mainly at the domain level. We evaluate semantic user-profiling strategies on a production streaming platform, ranking against the full catalog. The evaluation covers a 2*2 design space crossing representation type (aggregate or LLM-generated) with contextual scope (holistic history or attention-fused short-term and long-term contexts). The relative ordering of the two representation types is conditional on the user's consumption regime. Aggregate profiles are consistently stronger under habitual consumption, which characterizes approximately four-fifths of the population, while LLM-generated profiles are stronger for exploratory users whose subsequent interactions diverge semantically from their history. We also observe a popularity-attractor effect in LLM-generated profiles, which modestly raises within-list diversity while substantially lowering catalog coverage and reducing novelty. These results indicate that a context-aware system can select a profiling strategy from the inferred consumption regime rather than applying one representation to all users.