Legal Citations Reimagined: IIITH Team Wins Global Recognition For AI- Driven Precedent Retrieval Study
Their study has won the best paper award at the 37th International Conference on Legal Knowledge and Information Systems (JURIX 2024), held in the Czech Republic.
Hyderabad: A research team from IIIT-Hyderabad said it had developed a new method to improve how legal precedents are found using artificial intelligence, by focusing not just on citations themselves but the words around them.
Their study has won the best paper award at the 37th International Conference on Legal Knowledge and Information Systems (JURIX 2024), held in the Czech Republic. The research was led by Gaurang Patil, a second-year MS by Research student, under the guidance of Prof. P. Krishna Reddy.
The IIITH team proposed that the context—phrases and sentences around a citation—can help machines better understand the cited judgement’s purpose.
“Anchor text was once used in web pages to make internet search better. We applied the same logic to legal documents,” said Gaurang. Their experiments, using Supreme Court judgments from the ‘Indian Kanoon’ platform, confirmed that contextual citation data made a real difference. It improved document representation, a critical factor in how AI retrieves and ranks judgements.
Prof. Reddy, who accepted the award, said this opened up new possibilities. “We can now look at using this approach to not just find judgements, but also generate better summaries of them. That’s where we want to go next.”
The work was supported by the iHub Anubhuti-IIITD Foundation and marks a step forward in the digital transformation of legal research, where more context-driven AI models can make legal resources more accessible and efficient for lawyers, judges, and researchers.