Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{187220,
        author = {SUMEDH KHEDEKAR and ABHIJEET JOSHI and PRATIK JOSHI and YASH BADGUJAR and DR. MR. YOGESHCHANDRA PURANIK},
        title = {Legal Perplexity–AI-Based Legal Document Query Assistant},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {4933-4939},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187220},
        abstract = {Recent Advances in large language models (LLMS) have spurred interest in automated legal question answering, but purely parametric LLMs often hallucinate or cite nonexistent authorities. Retrieval Augmented Generation (RAG) mitigates this by integrating factual legal text into generation process. We present Legal Perplexity, a RAG-based system for answering user queries about constitutional and legal texts. The backend is implemented with python and FastAPI and it uses a dense-vector retrieval pipeline to ground answers in constitutional law documents. Document chunks are embedded via Transformer encoders and indexed in a vector store (e.g FAISS) for nearest neighbour search. At query time, user questions are embedded and matched against the database; the top relevant passages are combined with the query and passed to language model (OpenAI GPT or Hugging Face) to produce a grounded answer.},
        keywords = {Legal AI, question, answering, retrieval-augmented generation (RAG), FastAPI, embeddings, Vector search, constitutional law, language models.},
        month = {November},
        }