SelfDR: Self-Distillation from Reasoning for LLM-Based Recommendation

2026-09-02

AI Summary

为解决LLM推荐系统中推理成本高问题，提出SelfDR框架，通过自蒸馏方法直接生成推荐，提高效率和准确性。

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 systems. To address this challenge, we propose SelfDR, a Self-Distillation from Reasoning framework for LLM-based Recommendation. SelfDR distills an LLM's own reasoning-enhanced predictions to produce recommendations directly, improving recommendation effectiveness while maintaining inference efficiency. All components in the framework are built on the same base LLM, without relying on any external models. Specifically, the teacher recommender is constructed by training a reasoner with downstream performance as the reward, enabling it to generate targeted rationales that are later incorporated into the teacher's input. A student recommender for direct recommendation, with the same underlying model, then learns from the teacher through self-distillation with a dynamic weighting strategy. Extensive experiments on three public datasets validate the effectiveness, rationality, and efficiency of SelfDR. Codes are available at https://github.com/JiangDeccc/SelfDistillation.

Problem

• Large Language Models
• Recommendation
• Reasoning
• Computational Costs
• Inference Efficiency

Innovation

• Self-Distillation
• Reasoning
• Efficiency
• Recommendation