DocumentIQA: Scientific Document Insight Question/Answer

Introduction

Question/Answering on scientific documents. In our implementation we use Grobid for text extraction instead of the raw PDF2Text converter. Thanks to Grobid we are able to precisely extract abstract and full-text. This is just the beginning and publishing might help gathering more feedback.

NOTE: This project focus on scientific articles. Uploading books or other large document might not work as expected.

Work in progress

https://document-insights.streamlit.app/

OpenAI or HuggingFace API KEY required

Acknolwedgement

This project is developed at the National Institute for Materials Science (NIMS) in Japan.