Legal information is often difficult for ordinary people to understand due to the use of complex terminology and lengthy legal documents. In Pakistan, limited access to simplified legal guidance creates challenges for students, researchers, and the general public in understanding their legal rights and responsibilities. To address this issue, we developed LexAI – AI Powered Legal Assistant, an intelligent system designed to simplify Pakistani laws and provide accessible legal information through Artificial Intelligence technologies.LexAI utilizes Natural Language Processing (NLP), Retrieval-Augmented Generation (RAG), and LLaMA-based summarization to analyze user queries, retrieve relevant legal information, and generate simplified responses in understandable language. The system is trained on a legal dataset collected from publicly available Pakistani legal resources and government law websites, including constitution articles, legal acts, ordinances, and legal question-answer data. The dataset was carefully preprocessed and organized to improve retrieval accuracy and response quality.LexAI aims to improve legal awareness, reduce the complexity of legal information, and provide quick access to legal guidance through a user-friendly AI-powered platform. The project demonstrates the practical application of Artificial Intelligence in the legal domain and highlights the potential of AI-driven solutions for improving access to legal knowledge in Pakistan.
Tools: Python,LangChain,FAISS,OpenAI API,Streamlit,Hugging Face Transformers,Tensor Flow,PyTorch
Department: Department of Mathematics
Poster
