Improving LLM-powered Recommendations with Personalized Information

Jiahao Liu, Xueshuo Yan, Dongsheng Li, Guangping Zhang, Hansu Gu +4 more
2/19/2025
cs.IRcs.AI

Official code: jhliu0807/CoT-Rec (11 stars, last commit 2025-02-20).

Abstract

Due to the lack of explicit reasoning modeling, existing LLM-powered recommendations fail to leverage LLMs' reasoning capabilities effectively. In this paper, we propose a pipeline called CoT-Rec, which integrates two key Chain-of-Thought (CoT) processes -- user preference analysis and item perception analysis -- into LLM-powered recommendations, thereby enhancing the utilization of LLMs' reasoning abilities. CoT-Rec consists of two stages: (1) personalized information extraction, where user preferences and item perception are extracted, and (2) personalized information utilization, where this information is incorporated into the LLM-powered recommendation process. Experimental results demonstrate that CoT-Rec shows potential for improving LLM-powered recommendations. The implementation is publicly available at https://github.com/jhliu0807/CoT-Rec.

View on arXivView PDF

Code Implementations(1)

jhliu0807/CoT-RecOfficial100%
110Python, Jupyter NotebookFeb 18, 20251 years ago

Similar Papers

ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoning

86% match1 repo72024

A Survey on LLM-powered Agents for Recommender Systems

82% match1 repo12025

GOT4Rec: Graph of Thoughts for Sequential Recommendation

82% match2024

Explainable Active Learning for Preference Elicitation

82% match2023

DRDT: Dynamic Reflection with Divergent Thinking for LLM-based Sequential Recommendation

82% match2023

HLLM: Enhancing Sequential Recommendations via Hierarchical Large Language Models for Item and User Modeling

81% match1 repo6252024

Cite this paper

@article{liu2025improving,
  title  = {Improving LLM-powered Recommendations with Personalized Information},
  author = {Jiahao Liu and Xueshuo Yan and Dongsheng Li and Guangping Zhang and Hansu Gu and Peng Zhang and Tun Lu and Li Shang and Ning Gu},
  year   = {2025},
  eprint = {2502.13845},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2502.13845v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2502.13845.svg)](https://codeofpaper.com/paper/2502.13845)

Discussion