Spaces:
Sleeping
Sleeping
| title: DocuCite Agent | |
| emoji: 👁 | |
| colorFrom: pink | |
| colorTo: red | |
| sdk: gradio | |
| sdk_version: 5.33.1 | |
| app_file: app.py | |
| pinned: false | |
| license: mit | |
| Demo: | |
| https://github.com/user-attachments/assets/24056190-9413-4bb3-9918-fcd739b690a5 | |
| Team Members | |
| Monit KORN | |
| Setthika SUN | |
| <a name="mcp-server-track"></a> | |
| DocuCite-AI | |
| This repository contains is a Gradio-based Agentic Retrieval-Augmented Generation application using LangChain and LangGraph. It answers questions based on the PDF file you upload, the tool automatically retrieving and citing the page and paragraph where it found the supporting information. | |
| Features | |
| 1. PDF Upload & Embedding: Upload any PDF, automatically chunk, clean, and embed its contents into a vector store (Chroma). | |
| 2. Agentic Retrieval: Uses LangGraph to orchestrate an agentic workflow that decides when to call a retrieval tool or generate a direct answer. | |
| 3. Citation-Aware Tool: Custom document_search tool returns results prefixed with [Page X, Paragraph Y]. | |
| 4. Interactive UI: Built with Gradio for easy deployment and use. | |
| Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference | |