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.

翻译：大型语言模型（LLM）在自然语言处理任务中展现出卓越性能。然而，在推荐系统领域，由于用户行为数据与自然语言之间存在固有结构差异，LLM难以有效建模用户偏好与物品之间的关联。尽管基于提示的方法能够生成推荐结果，但其对推荐任务的理解不足导致性能受限。为弥补这一差距，我们构建了一个全面的指令微调数据集ITDR，涵盖用户-物品交互和用户-物品理解两大根任务下的七个子任务。该数据集整合了13个公开推荐数据集的数据，通过人工设计的标准化模板构建而成，包含约20万个实例。实验结果表明，ITDR显著提升了GLM-4、Qwen2.5、Qwen2.5-Instruct和LLaMA-3.2等主流开源LLM在推荐任务上的性能。此外，我们分析了任务间的相关性，并探究了任务描述与数据规模对指令微调效果的影响。最后，我们与具有海量参数的闭源LLM进行了对比实验。我们的微调数据集ITDR、微调后的大型推荐模型、所有LoRA模块及完整实验结果均已公开于https://github.com/hellolzk/ITDR。