Overview
Background
Shuai Wang is a Research Fellow at ielab, The University of Queensland, working on AI-powered search. He builds systems that find information and answer questions, using large language models and retrieval-augmented generation (RAG), and he focuses on making those systems faster and cheaper to run. His broader research contributions span federated search optimization and improving model efficiency in IR and retrieval-augmented generation (RAG) applications. His work has been published at premier venues including SIGIR, ECIR, WSDM, and EMNLP. He has served on program committees for SIGIR, ECIR, ICTIR, and TOIS. A lot of his work comes back to one practical question: how do we get good search and reliable answers without depending on expensive, closed commercial AI?
Shuai has published 25+ papers at venues such as SIGIR, WSDM, ECIR, and EMNLP. He coordinates and teaches INFS7410 (Information Retrieval and Web Search) at UQ.
Shuai completed his PhD on automating medical systematic reviews using neural retrieval systems and generative models (thesis: AI-driven Automated Systematic Reviews). His doctoral work encompassed automatic MeSH term suggestion, screening prioritization, seed-driven retrieval methods, and automatic Boolean query formulation.
Education
- PhD, The University of Queensland (2021–2025)
- Master's degree, The University of Queensland (2020–2021)
- Bachelor's degree, The University of Western Australia (2017–2019)
Availability
- Dr Shuai Wang is:
- Available for supervision
Qualifications
- Masters (Coursework) of Software Engineering, The University of Queensland
- Doctor of Philosophy of Information Retrieval and Web Search, The University of Queensland
Research interests
-
Efficient, Effective and Adaptive Retrieval
Search systems should return the right results without wasting computation. This area studies dense and sparse retrieval that stays accurate while controlling cost, and adaptive models that scale their effort to the difficulty of each query. Topics include multi-representation embeddings, model efficiency, and the trade-offs between effectiveness, speed, and memory in large-scale search.
-
Efficient Retrieval-Augmented Generation (RAG)
RAG lets language models answer questions using retrieved evidence, but it is costly to run at scale. This area focuses on making RAG cheaper and faster through context compression, reduced redundant computation, and better memory use, so that reliable, grounded question answering can run on local hardware rather than depending only on large commercial APIs.
-
Effective Search in the Agent Era
Search looks very different when the user is an AI agent rather than a person, so the way retrieval works needs rethinking. Agents issue many queries, chain steps together, and act on what they find, which calls for search designed specifically for them. This area asks how to build retrieval that is effective, efficient, and reliable for agentic systems.
-
AI for Systematic Reviews and Clinical Evidence
Finding and screening medical evidence is slow and labour-intensive. This area applies retrieval and language models to evidence synthesis, including query formulation, term suggestion, and screening automation, helping clinicians and researchers work faster while keeping the high recall and transparency that high-stakes reviews demand.
Works
Search Professor Shuai Wang’s works on UQ eSpace
2022
Conference Publication
Neural rankers for effective screening prioritisation in medical systematic review literature search
Wang, Shuai, Scells, Harrisen, Koopman, Bevan and Zuccon, Guido (2022). Neural rankers for effective screening prioritisation in medical systematic review literature search. 26th Australasian Document Computing Symposium, Adelaide, SA, Australia, 15-16 December 2022. New York, NY, United States: ACM. doi: 10.1145/3572960.3572980
2022
Journal Article
Automated MeSH term suggestion for effective query formulation in systematic reviews literature search
Wang, Shuai, Scells, Harrisen, Koopman, Bevan and Zuccon, Guido (2022). Automated MeSH term suggestion for effective query formulation in systematic reviews literature search. Intelligent Systems with Applications, 16 200141, 1-14. doi: 10.1016/j.iswa.2022.200141
2022
Conference Publication
From little things big things grow: a collection with seed studies for medical systematic review literature search
Wang, Shuai, Scells, Harrisen, Clark, Justin, Koopman, Bevan and Zuccon, Guido (2022). From little things big things grow: a collection with seed studies for medical systematic review literature search. SIGIR '22: The 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, Madrid, Spain, 11-15 July 2022. New York, United States: Association for Computing Machinery. doi: 10.1145/3477495.3531748
2022
Conference Publication
To Interpolate or not to Interpolate: PRF, dense and sparse retrievers
Li, Hang, Wang, Shuai, Zhuang, Shengyao, Mourad, Ahmed, Ma, Xueguang, Lin, Jimmy and Zuccon, Guido (2022). To Interpolate or not to Interpolate: PRF, dense and sparse retrievers. SIGIR '22: The 45th International ACM SIGIR Conference on Research and Development in Information Retrieval, Madrid, Spain, 11-15 July 2022. New York, United States: Association for Computing Machinery. doi: 10.1145/3477495.3531884
2022
Conference Publication
Seed-driven document ranking for systematic reviews: a reproducibility study
Wang, Shuai, Scells, Harrisen, Mourad, Ahmed and Zuccon, Guido (2022). Seed-driven document ranking for systematic reviews: a reproducibility study. 44th European Conference on IR Research, ECIR 2022, Stavanger, Norway, 10–14 April 2022. Cham, Switzerland: Springer. doi: 10.1007/978-3-030-99736-6_46
2021
Conference Publication
MeSH Term Suggestion for Systematic Review Literature Search
Wang, Shuai, Li, Hang, Scells, Harrisen, Locke, Daniel and Zuccon, Guido (2021). MeSH Term Suggestion for Systematic Review Literature Search. Australasian Document Computing Symposium, Melbourne, VIC, Australia, 9 December 2021. New York, NY, United States: Association for Computing Machinery. doi: 10.1145/3503516.3503530
2021
Conference Publication
BERT-based Dense Retrievers Require Interpolation with BM25 for Effective Passage Retrieval
Wang, Shuai, Zhuang, Shengyao and Zuccon, Guido (2021). BERT-based Dense Retrievers Require Interpolation with BM25 for Effective Passage Retrieval. The ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR), Canada, 11 July 2021. New York, NY, USA: Association for Computing Machinery, Inc. doi: 10.1145/3471158.3472233
Funding
Past funding
Supervision
Availability
- Dr Shuai Wang is:
- Available for supervision
Looking for a supervisor? Read our advice on how to choose a supervisor.
Media
Enquiries
For media enquiries about Dr Shuai Wang's areas of expertise, story ideas and help finding experts, contact our Media team: