This paper provides a comprehensive survey of the application of BERT-based models in information retrieval (IR). It reviews various techniques and categories of BERT-based approaches, including handling long documents, integrating semantic information, balancing effectiveness and efficiency, predicting term weights, query expansion, and document expansion. The authors also compare BERT-based models with large language models (LLMs) like ChatGPT, highlighting the advantages of BERT in terms of computational efficiency and privacy. The survey covers the latest advancements in IR, including the use of BERT for ad-hoc IR tasks and the challenges associated with these applications. Additionally, it provides links to resources such as datasets and toolkits for BERT-based IR systems. The paper concludes by summarizing the key findings and suggesting future research directions in the field of BERT-based IR.This paper provides a comprehensive survey of the application of BERT-based models in information retrieval (IR). It reviews various techniques and categories of BERT-based approaches, including handling long documents, integrating semantic information, balancing effectiveness and efficiency, predicting term weights, query expansion, and document expansion. The authors also compare BERT-based models with large language models (LLMs) like ChatGPT, highlighting the advantages of BERT in terms of computational efficiency and privacy. The survey covers the latest advancements in IR, including the use of BERT for ad-hoc IR tasks and the challenges associated with these applications. Additionally, it provides links to resources such as datasets and toolkits for BERT-based IR systems. The paper concludes by summarizing the key findings and suggesting future research directions in the field of BERT-based IR.