基于带有LLM相关性判断的高斯过程回归的自然语言推荐多模态物品评分
Multimodal Item Scoring for Natural Language Recommendation via Gaussian Process Regression with LLM Relevance Judgments
AI总结：
针对自然语言推荐中单模态评分函数的局限性，提出GPR-LLM方法，利用高斯过程回归结合LLM相关性判断建模多模态评分函数，在最小标注预算下显著优于基线方法。
摘要：
自然语言推荐（NLRec）根据用户发出的NL请求与NL物品描述段落之间的相关性生成物品建议。现有NLRec方法常使用密集检索（DR），通过聚合用户请求嵌入与相关段落嵌入的内积来计算物品相关性得分。然而，DR将请求视为唯一的相关性标签，从而导致以查询嵌入为中心的单模态评分函数，这通常是查询相关性的弱代理。为更好捕捉复杂NLRec数据可能引发的相关性评分函数的潜在多模态分布，我们提出GPR-LLM，该方法对候选段落子集使用带有LLM相关性判断的高斯过程回归（GPR）。在四个NLRec数据集和两个LLM骨干上的实验表明，使用RBF核的GPR-LLM能够建模多模态相关性评分函数，持续优于更简单的单模态核（内积、余弦相似度），以及包括DR、交叉编码器和逐点基于LLM的相关性评分在内的基线方法，提升达65%。总体而言，GPR-LLM在最小LLM标注预算内为NLRec提供了一种高效有效的方法。
英文摘要：
Natural Language Recommendation (NLRec) generates item suggestions based on the relevance between user-issued NL requests and NL item description passages. Existing NLRec approaches often use Dense Retrieval (DR) to compute item relevance scores from aggregation of inner products between user request embeddings and relevant passage embeddings. However, DR views the request as the sole relevance label, thus leading to a unimodal scoring function centered on the query embedding that is often a weak proxy for query relevance. To better capture the potential multimodal distribution of the relevance scoring function that may arise from complex NLRec data, we propose GPR-LLM that uses Gaussian Process Regression (GPR) with LLM relevance judgments for a subset of candidate passages. Experiments on four NLRec datasets and two LLM backbones demonstrate that GPR-LLM with an RBF kernel, capable of modeling multimodal relevance scoring functions, consistently outperforms simpler unimodal kernels (dot product, cosine similarity), as well as baseline methods including DR, cross-encoder, and pointwise LLM-based relevance scoring by up to 65%. Overall, GPR-LLM provides an efficient and effective approach to NLRec within a minimal LLM labeling budget.
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。
