LLM-KT: A Versatile Framework for Knowledge Transfer from Large Language Models to Collaborative Filtering

ICDM . 2024.

Северин Н. Н., Булычев И. Д., Yushkov M., Kushneruk A., Кулаева А. А., Kiselev D., Savchenko A., Makarov I., Ziablitsev A., Savelyeva Y., Tashchilin V.

We present LLM-KT, a flexible framework designed to enhance collaborative filtering (CF) models by seamlessly integrating LLM (Large Language Model)-generated features. Unlike existing methods that rely on passing LLM-generated features as direct inputs, our framework injects these features into an intermediate layer of any CF model, allowing the model to reconstruct and leverage the embeddings internally. This modelagnostic approach works with a wide range of CF models without requiring architectural changes, making it adaptable to various recommendation scenarios. Our framework is built for easy integration and modification, providing researchers and developers with a powerful tool for extending CF model capabilities through efficient knowledge transfer. We demonstrate its effectiveness through experiments on the MovieLens and Amazon datasets, where it consistently improves baseline CF models. Experimental studies showed that LLM-KT is competitive with the state-of-the-art methods in context-aware settings but can be applied to a broader range of CF models than current approaches. Index Terms—Large Language Model (LLM), recommender systems, knowledge transfer, RecBole framework

Keywords: knowledge transfer, Collaborative filtering, Large Language Models, LLM-KT

Publication based on the results of: Models and methods for natural language processing, recommendation systems and data mining (2024)