Retrieval Augmented Spelling Correction for E-Commerce Applications

TL;DR

This paper introduces a retrieval augmented generation approach for e-commerce spelling correction, effectively distinguishing new brand names from misspellings by integrating catalog retrieval with a fine-tuned language model.

Contribution

It presents a novel retrieval augmented spelling correction method that enhances accuracy by incorporating product catalog data into a language model's context.

Findings

Improved spelling correction accuracy with RAG framework

Retrieving product catalog data enhances model performance

Fine-tuning LLM with retrieved context yields better results

Abstract

The rapid introduction of new brand names into everyday language poses a unique challenge for e-commerce spelling correction services, which must distinguish genuine misspellings from novel brand names that use unconventional spelling. We seek to address this challenge via Retrieval Augmented Generation (RAG). On this approach, product names are retrieved from a catalog and incorporated into the context used by a large language model (LLM) that has been fine-tuned to do contextual spelling correction. Through quantitative evaluation and qualitative error analyses, we find improvements in spelling correction utilizing the RAG framework beyond a stand-alone LLM. We also demonstrate the value of additional finetuning of the LLM to incorporate retrieved context.