LegalAI Research via a Technological Perspective

This talk reviews the trajectory of LegalAI research from task-specific models to LLM-based systems. In its earlier stage, the work focused on automatic trials under a “one model, one task” paradigm, addressing judicial questioning, court’s view generation, and judgment prediction with attention to interpretability and fairness. More recently, the research has shifted toward large legal models, represented by Zhihai-Luwen, which are constructed through continued pretraining, supervised fine-tuning, and retrieval-augmented generation (RAG). These models are further applied through large–small model collaboration and counter-questioning strategies in judicial scenarios. This evolution highlights the shift from narrow automation to general-purpose legal intelligence.

Yiquan Wu is an Assistant Professor (ZJU-100 Young Professor, Doctoral Supervisor) at Zhejiang University. He received his Ph.D. and B.S. degrees from Zhejiang University. His research interests encompass LegalAI, Large Language Models, Natural Language Processing, and Causal Inference. He has published over 20 papers in prestigious conferences and journals such as ACL, EMNLP, NAACL, SIGIR, AAAI, KDD, and WWW.

Moderator: Benjamin Chen, Associate Professor & Director of the Law and Technology Centre, The University of Hong Kong Faculty of Law

To register, please go to https://hkuems1.hku.hk/hkuems/ec_regform.aspx?guest=Y&UEID=102549.  A paper will be circulated in advance and attendees will be expected to have read the paper before the seminar.  

We are applying for a CPD point with the Law Society of Hong Kong.