Cluster Language Model for Improved E-Commerce Retrieval and Ranking: Leveraging Query Similarity and Fine-Tuning for Personalized Results

Abstract: This paper proposes a novel method to improve the accuracy of product search in e-commerce by utilizing a cluster LLM. The method aims to address the limitations of the bi-encoder architecture while maintaining a minimal additional training burden. The approach involves labeling top products for each query, generating semantically similar query clusters using the K-Means clustering algorithm, and fine-tuning a global LLM into cluster LLMs on individual clusters. The parameters of each cluster LLM are fine-tuned to learn local manifolds in the feature space efficiently, capturing the nuances of various query types within each cluster. The inference is performed by assigning a new query to its respective cluster and utilizing the corresponding cluster LLM for retrieval. The proposed method results in more accurate and personalized retrieval results, offering a superior alternative to the popular bi-encoder based retrieval models in semantic search.