Human Centered AI for Indian Legal Text Analytics

Legal research is a crucial task in the practice of law. It requires intensehuman effort and intellectual prudence to research a legal case and preparearguments. Recent boom in generative AI has not translated to proportionaterise in impactful legal applications, because of low trustworthiness and andthe scarcity of specialized datasets for training Large Language Models (LLMs).This position paper explores the potential of LLMs within Legal Text Analytics(LTA), highlighting specific areas where the integration of human expertise cansignificantly enhance their performance to match that of experts. We introducea novel dataset and describe a human centered, compound AI system thatprincipally incorporates human inputs for performing LTA tasks with LLMs.