让大模型与推荐系统互相学习，提升推荐精度且不增加计算开销
Bidirectional Knowledge Distillation for Enhancing Sequential Recommendation with Large Language Models
- 双向蒸馏：大模型与推荐模型交替优化，动态共享知识
- 在多个数据集上显著提升推荐准确率，推理成本不变
- 无需额外参数，适合实际部署的高效融合方案
大型语言模型（LLMs）在理解与生成语义模式方面表现出色，是序列推荐任务的有力候选者。然而，当与传统推荐模型（CRMs）结合时，常面临推理成本高和知识传递方式静态的问题。本文提出一种新型双向蒸馏框架LLMD4Rec，促进以大模型为中心与基于CRM的推荐系统间的动态、双向知识交互。不同于传统的单向蒸馏，LLMD4Rec通过交替优化两个模型，增强CRM的语义理解能力，并使大模型融入用户-物品交互的协同信号。通过样本自适应加权与输出分布对齐，该方法在不引入额外参数的前提下实现有效知识迁移。在真实数据集上的大量实验表明，LLMD4Rec在多个基准上显著提升推荐精度，且不增加推理成本。该方法为融合大模型与传统推荐模型的优势提供了可扩展、高效的解决方案。
原文摘要 · Abstract (English)
Large language models (LLMs) have demonstrated exceptional performance in understanding and generating semantic patterns, making them promising candidates for sequential recommendation tasks. However, when combined with conventional recommendation models (CRMs), LLMs often face challenges related to high inference costs and static knowledge transfer methods. In this paper, we propose a novel mutual distillation framework, LLMD4Rec, that fosters dynamic and bidirectional knowledge exchange between LLM-centric and CRM-based recommendation systems. Unlike traditional unidirectional distillation methods, LLMD4Rec enables iterative optimization by alternately refining both models, enhancing the semantic understanding of CRMs and enriching LLMs with collaborative signals from user-item interactions. By leveraging sample-wise adaptive weighting and aligning output distributions, our approach eliminates the need for additional parameters while ensuring effective knowledge transfer. Extensive experiments on real-world datasets demonstrate that LLMD4Rec significantly improves recommendation accuracy across multiple benchmarks without increasing inference costs. This method provides a scalable and efficient solution for combining the strengths of both LLMs and CRMs in sequential recommendation systems.
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。