What’s the problem?
Large academic conferences can attract thousands of attendees and hundreds of presentations, making program construction increasingly complex. Organisers must group related papers into coherent sessions, manage multiple tracks and presentation rooms, minimise presenter conflicts and create schedules that attendees can easily navigate.
As research communities continue to grow and diversify, creating high-quality conference programs becomes increasingly difficult, placing significant administrative demands on conference organisers and volunteers.
What’s UQ’s innovation?
TaxoConf was developed by Professor Zhifeng Bao and Daomin Ji from UQ’s School of Electrical Engineering and Computer Science. The AI-powered conference program management system automates the construction of technical programs for large academic conferences.
The platform analyses accepted papers and presentation abstracts, automatically generates a topic taxonomy, and organises contributions into thematically coherent oral and poster sessions. It also accounts for practical scheduling constraints such as presenter conflicts, session sizes, room capacities and parallel tracks. Rather than replacing organisers, TaxoConf supports a human-in-the-loop workflow that allows conference chairs to review, refine and modify automatically generated programs.
The system can generate conference-ready schedules and program materials in a fraction of the time required by traditional approaches.
What’s the impact?
TaxoConf has already been adopted by major international conferences including VLDB 2026, SIGIR 2026 and IJCNN 2026, with further adoption planned for ICDE 2027, SIGIR 2027 and DASFAA 2027.
These conferences attract thousands of researchers and industry participants from around the world, demonstrating the practical value of the technology in managing large-scale scientific events. At VLDB 2026, which attracted more than 2,000 attendees, TaxoConf helped organise more than 150 sessions across six parallel tracks covering research papers, industry sessions, demonstrations, tutorials, panels and workshops.
By automating a task that traditionally requires weeks of manual effort, TaxoConf helps organisers create more coherent and accessible conference programs while reducing administrative workload. The technology demonstrates how AI can support the management of increasingly large and complex bodies of scientific knowledge and improve the experience for organisers, presenters and attendees alike.
Led by
- Professor Zhifeng Bao
School of Electrical Engineering and Computer Science, The University of Queensland
Project Team
- Daomin Ji
School of Electrical Engineering and Computer Science, The University of Queensland
Collaborator
- Dr. Junhao Gan
School of Computing and Information Systems, The University of Melbourne
| AI Research Strength | Data-Centric AI Scalable & Sustainable AI |
|---|---|
| Industry Portfolio | Advancing AI |
| Key Partners | VLDB · SIGIR · IJCNN · ICDE · DASFAA conference communities |
| Key Publications | Ji et al. (2026). Taxonomy Maintenance in the Wild over Evolving Scholarly Data: Reliability, Efficiency, and Cost-Effectiveness. Proceedings of the ACM SIGMOD International Conference on Management of Data (SIGMOD 2027). arXiv:2607.09149. Ji et al. (2026). TaxoConf: Taxonomy-Guided Automatic Conference Program Organization. arXiv preprint. arXiv:2609.06604. |
| Supporting Link | taxoconf.com |
Published 15 September 2026