Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Nikita Severin, Aleksei Ziablitsev, Yulia Savelyeva, Valeriy Tashchilin, Ivan Bulychev, Mikhail Yushkov, Artem Kushneruk, Amaliya Zaryvnykh, Dmitrii Kiselev, Andrey Savchenko, Ilya Makarov

TL;DR

LLM-KT is a flexible framework designed to enhance collaborative filtering (CF) models by seamlessly integrating LLM (Large Language Model)-generated features into an intermediate layer of any CF model, allowing the model to reconstruct and leverage the embeddings internally.

Abstract

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 model-agnostic 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.

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

TL;DR

LLM-KT is a flexible framework designed to enhance collaborative filtering (CF) models by seamlessly integrating LLM (Large Language Model)-generated features into an intermediate layer of any CF model, allowing the model to reconstruct and leverage the embeddings internally.

Abstract

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 model-agnostic 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.

Paper Structure

This paper contains 11 sections, 2 equations, 1 figure, 3 tables.

Table of Contents

  1. Introduction
  2. Proposed Method
  3. Proposed Knowledge Transfer
  4. Training process
  5. LLM-KT Framework
  6. General Pipeline
  7. Framework Features
  8. Internal Structure
  9. Experimental Setup
  10. Experimental Results
  11. Conclusion

Figures (1)

  • Figure 1: Proposed LLM-KT framework. A user can set up an experimental config, selecting the components and specifying the whole pipeline by declaring the sequence of predefined commands that will be applied. The framework supports various of them, including setting loss functions at different stages, selecting the layer number to conduct knowledge transfer into, freezing weights, and more.