Transformer Based Language Models for Similar Text Retrieval and Ranking

TL;DR

This paper introduces transformer-based methods for similar text retrieval and ranking that do not rely on initial bag-of-words approaches, enabling accurate results even with no shared non-stopwords.

Contribution

The paper presents novel transformer-based techniques that directly retrieve and rank similar texts without initial bag-of-words filtering, improving accuracy in cases with no common words.

Findings

Effective retrieval without bag-of-words step

Supervised and unsupervised BERT-based methods demonstrated

Improved accuracy in text ranking tasks

Abstract

Most approaches for similar text retrieval and ranking with long natural language queries rely at some level on queries and responses having words in common with each other. Recent applications of transformer-based neural language models to text retrieval and ranking problems have been very promising, but still involve a two-step process in which result candidates are first obtained through bag-of-words-based approaches, and then reranked by a neural transformer. In this paper, we introduce novel approaches for effectively applying neural transformer models to similar text retrieval and ranking without an initial bag-of-words-based step. By eliminating the bag-of-words-based step, our approach is able to accurately retrieve and rank results even when they have no non-stopwords in common with the query. We accomplish this by using bidirectional encoder representations from transformers (BERT) to create vectorized representations of sentence-length texts, along with a vector nearest neighbor search index. We demonstrate both supervised and unsupervised means of using BERT to accomplish this task.