Search papers, labs, and topics across Lattice.
This paper introduces a knowledge distillation method to transfer user semantics from LLMs to sequential recommender systems, leveraging LLM-generated textual user profiles. The approach distills knowledge by training the sequential recommender to predict the LLM's user profile embeddings based on user interaction sequences. Experiments demonstrate that this method enhances recommendation accuracy while preserving the inference efficiency of traditional sequential models, avoiding the need for real-time LLM inference or model fine-tuning.
Get LLM-boosted recommendations without the LLM latency: this distillation method lets you bake rich user profiles into efficient sequential recommenders.
Sequential recommender systems have achieved significant success in modeling temporal user behavior but remain limited in capturing rich user semantics beyond interaction patterns. Large Language Models (LLMs) present opportunities to enhance user understanding with their reasoning capabilities, yet existing integration approaches create prohibitive inference costs in real time. To address these limitations, we present a novel knowledge distillation method that utilizes textual user profile generated by pre-trained LLMs into sequential recommenders without requiring LLM inference at serving time. The resulting approach maintains the inference efficiency of traditional sequential models while requiring neither architectural modifications nor LLM fine-tuning.
