Abstract
Textual entailment classification is one of the hardest tasks for the Natural Language Processing community. In particular, working on entailment with legal statutes comes with an increased difficulty, for example in terms of different abstraction levels, terminology and required domain knowledge to solve this task. In course of the COLIEE competition, we develop three approaches to classify entailment. The first approach combines Sentence-BERT embeddings with a graph neural network, while the second approach uses the domain-specific model LEGAL-BERT, further trained on the competition’s retrieval task and fine-tuned for entailment classification. The third approach involves embedding syntactic parse trees with the KERMIT encoder and using them with a BERT model. In this work, we discuss the potential of the latter technique and why of all our submissions, the LEGAL-BERT runs may have outperformed the graph-based approach.
Highlights
In this work, we develop three approaches for legal textual entailment classification on the English version of the Japanese Civil Code
Textual entailment classification requires capabilities which are normally attributed to humans who can acquire a deep knowledge of the legal domain to understand and interpret legal texts to reason about their relationship and lawfulness
With the advent of deep learning, there are many models which are tested on natural language inference tasks, and the same development exists in the Competition on Legal Information Extraction/Entailment (COLIEE) competition
Summary
We develop three approaches for legal textual entailment classification on the English version of the Japanese Civil Code. Textual entailment classification requires capabilities which are normally attributed to humans who can acquire a deep knowledge of the legal domain to understand and interpret legal texts to reason about their relationship and lawfulness. Such reasoning capabilities are yet to be developed on a machine, for example as a decision support in specific legal cases. With the advent of deep learning, there are many models which are tested on natural language inference tasks, and the same development exists in the COLIEE competition Their decision making is hard to understand for a human, deep learning approaches have consistently achieved good results in the past years on this task. They are often outperforming more explainable methods and because they are not trusted in the legal domain, the research and detailed analysis of their strengths and weaknesses is important to understand future research directions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
