paper-qa
High accuracy RAG for answering questions from scientific documents with citations
What is paper-qa?
High accuracy RAG for answering questions from scientific documents with citations
Overview of paper-qa
What it is
PaperQA2 is a package for retrieval augmented generation (RAG) on PDFs, text files, Microsoft Office documents, and source code files, with a focus on the scientific literature. It is designed to answer questions from scientific documents with high accuracy and provide citations. The project is open source under the Apache-2.0 license.
Key features
- A simple interface to get answers with grounded responses containing in-text citations.
- State-of-the-art implementation including document metadata-awareness in embeddings and LLM-based re-ranking and contextual summarization (RCS).
- Support for agentic RAG, where a language agent can iteratively refine queries and answers.
- Automatic redundant fetching of paper metadata, including citation and journal quality data from multiple providers.
- A usable full-text search engine for a local repository of PDF/text files.
- A robust interface for customization, with default support for all LiteLLM models.
Use cases
PaperQA2 is intended for high-accuracy question answering, summarization, and contradiction detection from scientific literature. It can be used to query a local collection of research papers, automatically retrieve metadata, and generate cited answers. The package is suitable for researchers, analysts, or anyone needing to extract information from scientific documents with verifiable sources.
Getting started
bashpip install paper-qamkdir my_paperscurl -o my_papers/PaperQA2.pdf https://arxiv.org/pdf/2409.13740cd my_paperspqa ask 'What is PaperQA2?'
License and pricing
The license is Apache-2.0. The project is open source and does not state any pricing.
Based on the project's public README and repository metadata.