BERT-QE: Contextualized Query Expansion for Document Re-ranking
This addresses query-document mismatch in information retrieval, offering a domain-specific improvement for search systems.
The paper tackled the problem of query expansion introducing non-relevant information by proposing BERT-QE, a model that uses BERT to select relevant document chunks for expansion, resulting in significant performance improvements over BERT-Large models on TREC Robust04 and GOV2 test collections.
Query expansion aims to mitigate the mismatch between the language used in a query and in a document. However, query expansion methods can suffer from introducing non-relevant information when expanding the query. To bridge this gap, inspired by recent advances in applying contextualized models like BERT to the document retrieval task, this paper proposes a novel query expansion model that leverages the strength of the BERT model to select relevant document chunks for expansion. In evaluation on the standard TREC Robust04 and GOV2 test collections, the proposed BERT-QE model significantly outperforms BERT-Large models.