arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.18249cs.IR
收藏

双树LLM增强的隐式协同过滤负采样

Dual-Tree LLM-Enhanced Negative Sampling for Implicit Collaborative Filtering

Beijing Institute of Technology(北京理工大学)

Jiayi Wu, Zhengyu Wu, Xunkai Li, Rong-Hua Li, Guoren Wang

AI总结:

本文提出双树LLM增强负采样方法,通过无文本和无微调的结构化编码提升隐式协同过滤推荐性能。

摘要:

负采样是隐式协同过滤推荐中的关键技术,通过对比观察到的交互与采样的未观察到的交互来实现高效有效的训练。最近,大语言模型(LLM)在推荐系统中展现出潜力;然而,LLM赋能的负采样研究仍较为有限。现有方法严重依赖文本信息和任务特定的微调,限制了实际应用。为此,我们提出了一个无文本和无微调的双树LLM增强负采样方法(DTL-NS)。它包含两个模块:(i)一个离线假负识别模块,利用层次索引树将协同结构和潜在语义信息转换为结构化的物品ID编码,用于LLM推理,从而准确识别假负;(ii)一个多视图硬负采样模块,结合用户-物品偏好分数与物品-物品层次相似性,从这些编码中挖掘高质量的负样本,从而提高推荐模型的判别能力。广泛的实验验证了DTL-NS的有效性。此外,DTL-NS在不同的隐式CF模型、负采样方法和LLM中均表现出广泛的应用性,一致地提升了推荐性能。

英文摘要:

Negative sampling is a pivotal technique in implicit collaborative filtering (CF) recommendation, enabling efficient and effective training by contrasting observed interactions with sampled unobserved ones. Recently, large language models (LLMs) have shown promise in recommender systems; however, research on LLM-empowered negative sampling remains underexplored. Existing methods heavily rely on textual information and task-specific fine-tuning, limiting practical applicability. To this end, we propose a text-free and fine-tuning-free Dual-Tree LLM-enhanced Negative Sampling method (DTL-NS). It consists of two modules: (i) an offline false negative identification module that leverages hierarchical index trees to transform collaborative structural and latent semantic information into structured item-ID encodings for LLM inference, enabling accurate identification of false negatives; and (ii) a multi-view hard negative sampling module that combines user-item preference scores with item-item hierarchical similarities from these encodings to mine high-quality negatives, thus improving the discriminative ability of recommender models. Extensive experiments demonstrate the effectiveness of DTL-NS. Moreover, DTL-NS shows broad applicability across different implicit CF models, negative sampling methods, and LLMs, consistently enhancing recommendation performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑