# ItemRAG: Item-Based Retrieval-Augmented Generation for LLM-Based Recommendation

arXiv: https://arxiv.org/abs/2511.15141 · HTML: https://mlsanity.com/paper/2511.15141/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in LLM reasoning for recommendation, one of 17 directions in LLM reasoning.
- Builds on 17 of its 29 references that are in the map.
- First posted on arXiv 19 Nov 2025.

## Placed in

- LLM reasoning / [LLM reasoning for recommendation](https://mlsanity.com/lr/llm-reasoning-for-recommendation/)
- Retrieval-augmented generation / [RAG for recommendation](https://mlsanity.com/rag/rag-for-recommendation/)

## Builds on (most cited first, 17 of 17)

- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation](https://mlsanity.com/paper/2002.02126/) (2020)
- [BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer](https://mlsanity.com/paper/1904.06690/) (2019)
- [A Survey on Large Language Models for Recommendation](https://mlsanity.com/paper/2305.19860/) (2023)
- [Bridging Language and Items for Retrieval and Recommendation: Benchmarking LLMs as Semantic Encoders](https://mlsanity.com/paper/2403.03952/) (2024)
- [Knowledge Graph Retrieval-Augmented Generation for LLM-based Recommendation](https://mlsanity.com/paper/2501.02226/) (2025)
- [Large Language Models meet Collaborative Filtering: An Efficient All-round LLM-based Recommender System](https://mlsanity.com/paper/2404.11343/) (2024)
- [CoRAL: Collaborative Retrieval-Augmented Large Language Models Improve Long-tail Recommendation](https://mlsanity.com/paper/2403.06447/) (2024)
- [Graph Retrieval-Augmented LLM for Conversational Recommendation Systems](https://mlsanity.com/paper/2503.06430/) (2025)
- [Process-Supervised LLM Recommenders via Flow-guided Tuning](https://mlsanity.com/paper/2503.07377/) (2025)
- [Unleashing the Power of Large Language Model for Denoising Recommendation](https://mlsanity.com/paper/2502.09058/) (2025)
- [Multi-Grained Patch Training for Efficient LLM-based Recommendation](https://mlsanity.com/paper/2501.15087/) (2025)
- [The Whole is Better than the Sum: Using Aggregated Demonstrations in In-Context Learning for Sequential Recommendation](https://mlsanity.com/paper/2403.10135/) (2024)
- [Multi-Behavior Recommender Systems: A Survey](https://mlsanity.com/paper/2503.06963/) (2025)
- [Revisiting Prompt Engineering: A Comprehensive Evaluation for LLM-based Personalized Recommendation](https://mlsanity.com/paper/2507.13525/) (2025)
- [KERAG_R: Knowledge-Enhanced Retrieval-Augmented Generation for Recommendation](https://mlsanity.com/paper/2507.05863/) (2025)

## Built on by (most cited first, 2 of 2)

- [Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems](https://mlsanity.com/paper/2607.04433/) (2026)
- [Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders](https://mlsanity.com/paper/2604.07825/) (2026)

## Abstract (arXiv)

Recently, large language models (LLMs) have been widely used as recommender systems, owing to their reasoning capability and effectiveness in handling cold-start items. A common approach prompts an LLM with a target user's purchase history to recommend items from a candidate set, often enhanced with retrieval-augmented generation (RAG). Most existing RAG approaches retrieve purchase histories of users similar to the target user; however, these histories often contain noisy or weakly relevant information and provide little or no useful information for candidate items. To address these limitations, we propose ItemRAG, a novel RAG approach that shifts focus from coarse user-history retrieval to fine-grained item-level retrieval. ItemRAG augments the description of each item in the target user's history or the candidate set by retrieving items relevant to each. To retrieve items not merely semantically similar but informative for recommendation, ItemRAG leverages co-purchase information alongside semantic information. Especially, through their careful combination, ItemRAG prioritizes more informative retrievals and also benefits cold-start items. Through extensive experiments, we demonstrate that ItemRAG consistently outperforms existing RAG approaches under both standard and cold-start item recommendation settings. Supplementary materials, code, and datasets are provided at https://github.com/kswoo97/ItemRAG.