提升大模型对用户-物品高阶交互的感知能力，改进推荐效果
Enhancing High-order Interaction Awareness in LLM-based Recommender Model
- 通过增强全词嵌入，让大模型更好理解图构建的交互关系
- 在直接和序列推荐任务上超越当前最优方法
- 适合关注大模型推荐中长程依赖与知识融合的研究者
大语言模型（LLMs）在推荐任务中展现出强大的推理能力，通过将推荐转化为文本生成任务实现。然而，现有方法或忽略、或低效建模用户-物品的高阶交互。为此，本文提出增强型基于大模型的推荐系统（ELMRec）。通过增强全词嵌入，显著提升大模型对图结构化交互的理解能力，无需图预训练。该发现为通过全词嵌入将丰富知识图谱融入大模型推荐提供了新思路。我们还发现大模型常依据用户早期交互而非近期行为进行推荐，并提出重排序解决方案。实验表明，ELMRec在直接推荐和序列推荐任务上均优于现有最先进方法。
原文摘要 · Abstract (English)
Large language models (LLMs) have demonstrated prominent reasoning capabilities in recommendation tasks by transforming them into text-generation tasks. However, existing approaches either disregard or ineffectively model the user-item high-order interactions. To this end, this paper presents an enhanced LLM-based recommender (ELMRec). We enhance whole-word embeddings to substantially enhance LLMs' interpretation of graph-constructed interactions for recommendations, without requiring graph pre-training. This finding may inspire endeavors to incorporate rich knowledge graphs into LLM-based recommenders via whole-word embedding. We also found that LLMs often recommend items based on users' earlier interactions rather than recent ones, and present a reranking solution. Our ELMRec outperforms state-of-the-art (SOTA) methods in both direct and sequential recommendations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。