ITDR: An Instruction Tuning Dataset for Enhancing Large Language Models in Recommendations

TL;DR

This paper introduces ITDR, a large instruction tuning dataset designed to improve the performance of large language models in recommendation tasks by bridging the gap between user behavior data and natural language understanding.

Contribution

The paper presents a comprehensive recommendation instruction tuning dataset, ITDR, built from multiple datasets, to enhance LLMs' effectiveness in recommendation scenarios.

Findings

ITDR significantly improves LLM performance on recommendation tasks.

Analysis of task correlations and data scale impacts.

Comparison with closed-source LLMs demonstrates effectiveness.

Abstract

Large language models (LLMs) have demonstrated outstanding performance in natural language processing tasks. However, in the field of recommender systems, due to the inherent structural discrepancy between user behavior data and natural language, LLMs struggle to effectively model the associations between user preferences and items. Although prompt-based methods can generate recommendation results, their inadequate understanding of recommendation tasks leads to constrained performance. To address this gap, we construct a comprehensive instruction tuning dataset, ITDR, which encompasses seven subtasks across two root tasks: user-item interaction and user-item understanding. The dataset integrates data from 13 public recommendation datasets and is built using manually crafted standardized templates, comprising approximately 200,000 instances. Experimental results demonstrate that ITDR significantly enhances the performance of mainstream open-source LLMs such as GLM-4, Qwen2.5, Qwen2.5-Instruct and LLaMA-3.2 on recommendation tasks. Furthermore, we analyze the correlations between tasks and explore the impact of task descriptions and data scale on instruction tuning effectiveness. Finally, we perform comparative experiments against closed-source LLMs with massive parameters. Our tuning dataset ITDR, the fine-tuned large recommendation models, all LoRA modules, and the complete experimental results are available at https://github.com/hellolzk/ITDR.