Hyderabad: A team of IIIT-Hyderabad said it has developed a new way to improve how to use artificial intelligence to discover legal precedents, not only the citation itself but the words around it.
Their research won the Best Paper Award at the 37th International Conference on Legal Knowledge and Information Systems (Jurix 2024) held in the Czech Republic. The study was directed by graduate second-year MS Gaurang Patil under the guidance of Professor P. Krishna Reddy.
The IIITH team proposed that the environment (words and sentences surrounding the citation) can help machines better understand the purpose of citation judgments.
“Anchor text used to make internet search better in web pages. We applied the same logic to legal documents,” Gaurang said. Their experiment used a Supreme Court decision, which was conducted from the “Indian Kanoon” platform, confirming that contextual citations do differ. It improves document representation, which is a key factor in how AI retrieves and ranks.
Professor Reddy, who accepted the award, said this opened up new possibilities. “Now, we can consider using this approach not only to find judgments, but also to produce a better summary. That’s where we’re going next.”
This work is supported by the IHUB ANUBHUTI-IIITD Foundation and marks a step forward in the digital transformation of legal research, in which more context-driven AI models can make legal resources more accessible and effective for lawyers, judges and researchers.
