返回全部动态
SelfDR：基于推理自蒸馏的LLM推荐系统框架
原标题：SelfDR: Self-Distillation from Reasoning for LLM-Based Recommendation
AI 摘要
SelfDR 是一个面向 LLM 推荐系统的自蒸馏框架，通过将模型自身推理增强的预测蒸馏为直接推荐结果，在保持推理效率的同时提升推荐效果。该框架基于同一基础 LLM 构建教师和学生模型，教师通过训练推理器生成高质量理由，学生则通过动态加权策略学习。在三个公开数据集上的实验验证了其有效性、合理性和效率。
以上摘要由 AI 生成，可能存在误差。事实请以原文为准。
正文节选
SelfDR: Self-Distillation from Reasoning for LLM-Based Recommendation Abstract. Large Language Models (LLMs) have recently emerged as powerful backbones for recommendation. To better elicit their capabilities, reasoning has been widely incorporated to help LLMs interpret rich textual signals and improve recommendation accuracy. However, explicitly generating intermediate reasoning traces often incurs substantial computational costs, which limits practical deployment in real-world recommender sys
发布时间：2026-09-04 12:00
抓取时间：2026-09-04 12:14
来源机构：arXiv