Legal AI Summarizer & Citation
View on GitHubGenerative AI
Legal AI Summarizer & Citation
A legal-document summarization prototype whose checked-in Flask application extracts PDF text with OCR and generates summaries with BART; citation search is described as in development.
PythonFlaskTransformersBARTOCR
Key Highlights
- The current application accepts PDF uploads, performs OCR, and summarizes extracted text with facebook/bart-large-cnn.
- The repository description says an advanced legal citation search system is currently in development.
Tech Stack
PythonFlaskTransformersBARTOCR
Overview
The public repository README is brief and describes a legal assistant concept. The checked-in Flask app currently implements PDF upload, OCR text extraction and BART-based summarization; the repository description says citation search is in development. The larger RAG, corpus-scale and evaluation claims previously shown on this portfolio are not supported by the checked-in README or source and have been removed.
How It Works
- Accept a PDF upload through a Flask application.
- Convert pages to images and extract text through OCR.
- Generate a summary using the facebook/bart-large-cnn model.
Tech Stack
- Python, Flask, pdf2image, pytesseract and Hugging Face Transformers (BART).