Short Academic CV

Research Profile

My research focuses on trustworthy agentic AI, large language model (LLM) safety and evaluation, and structured natural language processing (NLP) under limited or evolving supervision. I develop data-centric methods for building controllable, inspectable, and auditable AI systems, with applications in high-stakes domains such as clinical decision support, financial services, and public-sector AI.

Education

• PhD in Computer Science, Monash University, Australia, 2023 Thesis: Semantic Parsing in Limited Resource Conditions Supervisors: Prof. Gholamreza Haffari, Dr. Lizhen Qu, Prof. Philip R. Cohen

• Master of Computing (Advanced) in Artificial Intelligence, Australian National University, Australia, 2015 Thesis: Representation Learning for Weakly Supervised Relation Extraction

• Bachelor of Engineering in Electronic Information Engineering, Wuhan University of Science and Technology, China, 2013

Academic and Industry Appointments

• Lecturer, School of Computing Technologies, RMIT University, Melbourne, 2024–present
• Research Fellow, Faculty of Information Technology, Monash University, 2023–2024
• Research Scientist, Openstream.ai, 2023–2024
• Software Development Engineer, Microsoft Search Technology Center Asia, 2017–2018
• Engineer, Hong Kong Applied Science and Technology Research Institute, 2016–2017
• Visiting Student, National ICT Australia, 2015–2016

Selected Funding and Awards

• Wellcome Trust AI4You(th) project, co-applicant, on safe clinical LLMs for youth mental health; approximately AUD 6M total project funding, in partnership with Google Health.
• CSIRO Data61 Next Generation AI Graduate Program, co-principal investigator, on AI for next-generation food and waste systems.
• EMNLP 2025 Outstanding Paper Award, senior author, for DiscoSG: Towards Discourse-Level Text Scene Graph Parsing through Iterative Graph Refinement; one of seven such awards among 3,200+ accepted papers; first confirmed Australia-affiliated EMNLP Outstanding Paper.
• Monash Faculty of Information Technology Research Scholarship and International Postgraduate Research Scholarship.

Teaching

• RMIT University
  • Lecturer and course coordinator, ISYS1079 / ISYS3476: Managing Semi-structured and Unstructured Data
  • Lecturer, COSC2610: Web Programming Studio
  • Lecturer, COSC2502: C++ Programming Studio
• Monash University
  • Administrative tutor and guest lecturer, FIT5149: Applied Data Analysis
  • Tutor, FIT5125: IT Research Methods

Supervision

My supervision is centred on Trustworthy LLMs, Domain-grounded AI, and Structured NLP. See the Supervision page for current supervision areas and students.

Academic Service

• Senior Area Chair, EMNLP 2026
• Area Chair, NeurIPS 2026, ACL Rolling Review, and NLPCC 2026
• Program Chair, PersonaLLM Workshop at NeurIPS 2025; Publication Chair, ALTA Workshop 2026; Shared Task Organizer, ALTA 2024
• Reviewer for ACL, EMNLP, NAACL, EACL, ICLR, NeurIPS, AAAI, IJCAI, and related NLP/AI venues

Software, Open Source, and Patents

• FactualSceneGraph: toolkit for faithful and consistent textual scene-graph parsing, connecting FACTUAL (Findings of ACL 2023) with DiscoSG (EMNLP 2025 Outstanding Paper Award); .
• StarCoder 2 / The Stack v2: contributor to the BigCode open-science code LLM initiative; The Stack v2 spans 619 programming languages and supports StarCoder2 models trained on 3.3-4.3T tokens. Paper
• SCAR: ACL 2025 data selection method and toolkit for efficient instruction tuning; selecting as little as 0.7% of the full dataset can match or surpass full-data fine-tuning in reported benchmarks. Paper
• US Patent 12,548,554: active learning based multilingual semantic parser.
• US Patent Application 18/756,077: programmer-interpreter approach for LLM post-editing.

Languages

• English
• Mandarin Chinese