Keywords
Abstract
Information retrieval systems are examined from traditional lexical matching to modern neural models that are able to capture the semantic relationship between the query and the documents. A novel framework is proposed by integrating the query understanding component based on the BERT model and the rerank approach guided by the language model, aiming at enhancing the search engine effectiveness. Experiments conducted on the MSMARCO benchmark show significant improvements of 47.0% relative NDCG@10 values compared to the baseline BM25 model and an absolute 15.9% boost over the baseline BERT model. The query understanding model attains an accuracy of 94.3% on the intent classification task while being computationally efficient enough to be put into practice.
References
[1] Breuer, T., Frihat, S., Fuhr, N., Lewandowski, D., Schaer, P., & Schenkel, R. (2025). Large language models for information retrieval: Challenges and chances. Data Science. Advance online publication.
[2] Choi, J., Jung, E., Suh, J., & Rhee, W. (2021, July). Improving bi-encoder document ranking models with two rankers and multi-teacher distillation. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 2222–2226). Virtual Event, Canada.
[3] Craswell, N., Mitra, B., Yilmaz, E., Campos, D., & Lin, J. (2021, July). Ms marco: Benchmarking ranking models in the large-data regime. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 1566–1576). Virtual Event, Canada.
[4] Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019, June). BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) (pp. 4171–4186). Minneapolis, Minnesota, USA. https://doi.org/10.18653/v1/N19-1423
[5] Dong, C., Yao, S., Jiao, P., Yang, J., Jin, Y., Huang, Z., Zhou, X., Ou, D., Tang, H., & Zheng, B. (2025). TaoSR1: The thinking model for e-commerce relevance search. ArXiv.
[6] Guo, J., Fan, Y., Pang, L., Yang, L., Ai, Q., Zamani, H., Wu, C., Croft, W. B., & Cheng, X. (2020). A deep look into neural ranking models for information retrieval. Information Processing & Management, 57(6), 102067. https://doi.org/10.1016/j.ipm.2019.102067
[7] Lawrie, D., MacAvaney, S., Mayfield, J., McNamee, P., Oard, D. W., Soldaini, L., & Yang, E. (2025). Overview of the TREC 2024 NeuCLIR track. ArXiv
[8] Li, J., Zeng, W., Cheng, S., Ma, Y., Tang, J., Wang, S., & Yin, D. (2023, July). Graph enhanced BERT for query understanding. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 2160–2164). Taipei, Taiwan.
[9] Padaki, R., Dai, Z., & Callan, J. (2020, April). Rethinking query expansion for BERT reranking. In European Conference on Information Retrieval (pp. 297–304). Lisbon, Portugal. Springer.
[10] Santhanam, K., Khattab, O., Saad-Falcon, J., Potts, C., & Zaharia, M. (2022, July). ColBERTv2: Effective and efficient retrieval via lightweight late interaction. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (pp. 3715–3734). Seattle, Washington, USA. https://doi.org/10.18653/v1/2022.naacl-main.272
[11] Wang, J., Huang, J. X., Tu, X., Wang, J., Huang, A. J., Laskar, M. T. R., & Bhuiyan, A. (2024). Utilizing BERT for information retrieval: Survey, applications, resources, and challenges. ACM Computing Surveys, 56(7), 1–33.
[12] Zhan, J., Mao, J., Liu, Y., Zhang, M., & Ma, S. (2020, July). An analysis of BERT in document ranking. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 1941–1944). Virtual Event, China.
[13] Zhu, Y., Yuan, H., Wang, S., Liu, J., Liu, W., Deng, C., Chen, H., Liu, Z., Dou, Z., & Wen, J.-R. (2025). Large language models for information retrieval: A survey. ACM Transactions on Information Systems, 44(1), 1–54.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2026 Bangyi Yang