Loom: LLM-Powered Naturally Embedded Recommendation

Abstract

News publishers usually present recommendations in fixed locations outside the main reading flow, where engagement tends to be low, and maintaining high-quality recommendations at scale is expensive. Although research has improved what is recommended, the way recommendations are presented has changed little. This demonstration presents Loom, a modular pipeline that embeds recommendations inline within the article text. Given an article and, optionally, a reader profile inferred from previous interactions, Loom first retrieves candidate articles and then uses an LLM to determine where recommendations should be inserted and how they should be phrased in context. Recommendations are rendered as clearly marked, clickable inline spans that preserve the tone and flow of the surrounding text. The system is recommendation-model agnostic through a stable retrieval API and continuously collects interaction signals, including clicks, reading time, and survey responses, creating feedback signals that can be used to improve future recommendation and personalization without additional human curation. Users can browse articles with live insertions, toggle personalization, and inspect the generated placements.