A new paper details LSER, a production content recommendation system that integrates Large Language Model (LLM)-based embeddings into the retrieval stage. This system aims to improve how user interests are represented by adding LLM-derived session and history embeddings as auxiliary features. A nearline serving path was developed to keep these embeddings updated in real-time, and an ablation study confirmed the incremental value of these LLM features across a diverse catalog of content. AI
IMPACT Integrates LLM capabilities into retrieval systems, potentially improving personalization and user experience in content platforms.
RANK_REASON Academic paper detailing a new system for content recommendation using LLM embeddings. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
- alphaXiv
- arXiv
- CatalyzeX Code Finder for Papers
- Connected Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Influence Flower
- Large Language Model
- Litmaps
- LLM
- ScienceCast
- scite Smart Citations
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →