cs.IRNov 19, 2026

From Sessions to Serving: Integrating LLM-Based Embeddings into Retrieval for Production Content Recommendation

Authors: Zhixiong Hu, Meng Li, Maggie Zhang, Deborshi Goswami, David Bailey, James Bradley

Organizations: Autodesk, San Francisco, CA, USA

Abstract

Personalized content recommendation on large marketplace websites must reconcile heterogeneous candidate content (e.g., customer stories, product pages, industry pages, and learning content) with sparse, fast-changing user behavior, placing increasing demands on the retrieval stage to represent user interest well, not just the downstream ranking stage. We present LSER (LLM Session Embedding Retrieval), a production content recommendation system for autodesk.com that derives session- and history-level embeddings from a pretrained large language model (LLM) and adds them as auxiliary user-side features in the retrieval-stage user tower alongside existing learned sequence encoders. To keep user-side embeddings fresh under real-time behavior, we build a nearline serving path that asynchronously refreshes sequential and LLM-derived user vectors for our main content recommendation server. We validate the incremental value of the LLM features through an ablation study across our full multi-content-type catalog. Our results indicate that integrating LLM-derived embeddings as additional retrieval signals, together with production-grade nearline serving, is technically feasible and can improve model performances.

Figures & tables

Explore similar work

CardsList
  1. Enhancing High-order Interaction Awareness in LLM-based Recommender Model

    Sep 30, 2024Xinfeng Wang, Jin Cui, Fumiyo Fukumoto +1Sequential RecommendationFeature Interaction Modeling

  2. L2Rec: Towards Dual-View Understanding of LLMs for Personalized Recommendation

    May 26, 2026Pingjun Pan, Tingting Zhou, Peiyao Lu +3Mixture-of-Experts Language ModelsLLM Personalization