Zhanli-Li/DeepRead
[DeepRead] This is the official implementation of the DeepRead paper.
DeepRead helps you extract precise answers from complex PDFs, like financial reports or research papers, by understanding their structure, including images. You input a PDF document and a question, and it gives you a concise, accurate answer. This tool is for anyone who needs to quickly get specific information from long, detailed documents.
Use this if you frequently need to find precise answers or specific data points buried within lengthy, structured PDF documents.
Not ideal if your documents are not PDFs, or if you only need general summaries rather than specific extractions.
Stars
26
Forks
2
Language
Python
License
Apache-2.0
Category
Last pushed
Feb 24, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/rag/Zhanli-Li/DeepRead"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
GrapeCity-AI/gc-qa-rag
A RAG (Retrieval-Augmented Generation) solution Based on Advanced Pre-generated QA Pairs. 基于高级...
UKPLab/PeerQA
Code and Data for PeerQA: A Scientific Question Answering Dataset from Peer Reviews, NAACL 2025
Arfazrll/RAG-DocsInsight-Engine
Retrieval Augmented Generation (RAG) engine for intelligent document analysis. integrating LLM,...
faerber-lab/SQuAI
SQuAI: Scientific Question-Answering with Multi-Agent Retrieval-Augmented Generation (CIKM'25)
Vbj1808/Dokis
Lightweight RAG provenance middleware. Verifies every claim in an LLM response is grounded in a...