发表机构
Chongqing University of Technology; Peking University; Chongqing University（重庆理工大学; 北京大学; 重庆大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有方法仅从单一视角蒸馏LLM偏好知识的问题，提出P3Rec框架，联合提取并内化先验与后验偏好推理，利用兴趣熵自适应校准用户表示，在多个公开数据集上验证了有效性。
AI 中文摘要
大语言模型（LLMs）展现出强大的语义理解和偏好推理能力，为推荐系统中的用户建模提供了新的机遇。现有的LLM-as-Enhancer方法通常将LLM衍生的偏好知识提炼到轻量级推荐器中，以避免昂贵的在线LLM推理。然而，它们往往仅从单一视角构建蒸馏知识。先验偏好捕捉用户稳定且一致的兴趣，但对当前决策提供的指导有限，而后验偏好则揭示与目标相关的细粒度兴趣，但可能过度依赖目标线索。为解决这些局限性，我们提出了P$^3$Rec，一个联合提取并内化互补的先验和后验偏好推理知识的框架。具体而言，P$^3$Rec首先从用户侧推导出与目标无关的先验偏好和以目标为条件的后验偏好，同时从项目语义和前序交互中进一步提取以项目为中心的偏好表示。然后，它通过先验偏好吸收和后验引导的偏好蒸馏，逐步将先验和后验知识内化到行为表示中。由于由此产生的综合偏好表示可能并非总是提供同等决定性的检索方向，P$^3$Rec进一步利用兴趣熵刻画历史兴趣分散度，并在对比检索优化之前自适应地校准用户表示。通过这种方式，P$^3$Rec在保持高效推荐的同时实现了更完整的偏好推理。在多个公开数据集上的大量实验证明了其有效性。
英文摘要
Large language models (LLMs) exhibit strong semantic understanding and preference reasoning capabilities, offering new opportunities for user modeling in recommender systems. Existing LLM-as-Enhancer methods typically distill LLM-derived preference knowledge into lightweight recommenders to avoid costly online LLM inference. However, they often construct distillation knowledge from only one perspective. Prior preference captures users' stable and consistent interests but provides limited guidance for the current decision, whereas posterior preference reveals target-relevant fine-grained interests but may rely excessively on target clues. To address these limitations, we propose P$^3$Rec, a framework that jointly extracts and internalizes complementary prior and posterior preference reasoning knowledge. Specifically, P$^3$Rec first derives target-agnostic prior preferences and target-conditioned posterior preferences from the user side, while further extracting item-centric preference representations from item semantics and predecessor interactions. It then progressively internalizes prior and posterior knowledge into behavioral representations through prior preference absorption and posterior-guided preference distillation. Since the resulting comprehensive preference representation may not always provide an equally decisive retrieval direction, P$^3$Rec further characterizes historical interest dispersion with interest entropy and adaptively calibrates the user representation before contrastive retrieval optimization. In this way, P$^3$Rec achieves more complete preference reasoning while preserving efficient recommendation. Extensive experiments on multiple public datasets demonstrate its effectiveness.
