Improving Domain-Specific Retrieval by NLI Fine-Tuning

TL;DR

This paper explores how fine-tuning sentence encoders with NLI data enhances retrieval and ranking performance across English and Polish, demonstrating improvements in both monolingual and multilingual models.

Contribution

It shows that NLI fine-tuning significantly boosts retrieval and ranking tasks in multiple languages, with analysis of embedding properties explaining the improvements.

Findings

NLI fine-tuning improves retrieval and ranking performance.

Both monolingual and multilingual models benefit from NLI data.

Embedding uniformity and alignment are affected by NLI fine-tuning.

Abstract

The aim of this article is to investigate the fine-tuning potential of natural language inference (NLI) data to improve information retrieval and ranking. We demonstrate this for both English and Polish languages, using data from one of the largest Polish e-commerce sites and selected open-domain datasets. We employ both monolingual and multilingual sentence encoders fine-tuned by a supervised method utilizing contrastive loss and NLI data. Our results point to the fact that NLI fine-tuning increases the performance of the models in both tasks and both languages, with the potential to improve mono- and multilingual models. Finally, we investigate uniformity and alignment of the embeddings to explain the effect of NLI-based fine-tuning for an out-of-domain use-case.