Bringing legal knowledge to the public by constructing a legal question bank using large-scale pre-trained language model
Mingruo Yuan, Ben Kao, Tien-Hsuan Wu, Michael M. K. Cheung, Henry W. H. Chan, Anne S. Y. Cheung, Felix W. H. Chan, Yongxi Chen
TL;DR
This study addresses the public access gap to legal knowledge by proposing a three‑step framework: (1) translate primary legal sources into lay explanations (CLIC-pages), (2) assemble a scalable Legal Question Bank (LQB) whose answers reside in CLIC-pages, and (3) implement an interactive CLIC Recommender (CRec) to map layperson descriptions to relevant legal knowledge. Using GPT‑3 prompting with multiple page partitioning strategies, the authors generate machine-generated questions (MGQs) and compare them to human-composed questions (HCQs), finding MGQs to be more scalable and diverse, while HCQs are more precise. The Hybrid partitioning approach achieves the best overall metrics, delivering high quantity, high precision, broad coverage (up to 93% of paragraph content) and rich diversity, with MGQs approaching or surpassing manual question coverage in many cases. The CRec component demonstrates practical utility by ranking and presenting the most relevant MGQs and linked CLIC pages for a user’s scenario, suggesting a viable path to making legal information navigable and comprehensible for the general public. Overall, the work offers a scalable, data-driven pipeline for closing the legal knowledge gap and enables public-facing access to structured legal knowledge via explainable, context-aware questions and recommendations.
Abstract
Access to legal information is fundamental to access to justice. Yet accessibility refers not only to making legal documents available to the public, but also rendering legal information comprehensible to them. A vexing problem in bringing legal information to the public is how to turn formal legal documents such as legislation and judgments, which are often highly technical, to easily navigable and comprehensible knowledge to those without legal education. In this study, we formulate a three-step approach for bringing legal knowledge to laypersons, tackling the issues of navigability and comprehensibility. First, we translate selected sections of the law into snippets (called CLIC-pages), each being a small piece of article that focuses on explaining certain technical legal concept in layperson's terms. Second, we construct a Legal Question Bank (LQB), which is a collection of legal questions whose answers can be found in the CLIC-pages. Third, we design an interactive CLIC Recommender (CRec). Given a user's verbal description of a legal situation that requires a legal solution, CRec interprets the user's input and shortlists questions from the question bank that are most likely relevant to the given legal situation and recommends their corresponding CLIC pages where relevant legal knowledge can be found. In this paper we focus on the technical aspects of creating an LQB. We show how large-scale pre-trained language models, such as GPT-3, can be used to generate legal questions. We compare machine-generated questions (MGQs) against human-composed questions (HCQs) and find that MGQs are more scalable, cost-effective, and more diversified, while HCQs are more precise. We also show a prototype of CRec and illustrate through an example how our 3-step approach effectively brings relevant legal knowledge to the public.