← All notes

Summary

PaperQA: Retrieval-Augmented Generative Agent for Scientific Research (2023), by Jakub Lala, Odhran O'Donoghue, Aleksandar Shtedritski, Sam Cox, Samuel G. Rodriques, and Andrew D. White, is at https://arxiv.org/abs/2312.07559. The authors describe an agent that retrieves over full-text scientific articles, assesses the relevance of sources and passages, and then answers with retrieval-augmented generation. They report that, viewed as a question answering model, it exceeds existing language models and agents on current science QA benchmarks. To get closer to how people actually read the literature, they also introduce LitQA, a more complex benchmark that requires retrieval and synthesis from full-text papers, and they report that PaperQA matches expert human researchers on LitQA. They introduce LitQA, a more complex benchmark that requires retrieval and synthesis of information from full-text scientific papers across the literature. Viewing this agent as a question answering model, they find it exceeds performance of existing LLMs and LLM agents on current science QA benchmarks. They also report that PaperQA matches expert human researchers on LitQA. Title, authors, abstract, and comments were taken from the arXiv abs page. The version is v2 (14 Dec 2023), submitted 8 Dec 2023.

Claims

Pointer notes do not quote the paper. See Verification and Sources for the link and license.

Verification

• Format: pointer note only. The paper is not reproduced here.
• Why it matters: It matters because it is a concrete retrieval-augmented agent for scientific literature, with an emphasis on using sources rather than answering from model memory alone.
• Reproduced: not reproduced (2026-11-11).
• Related work: Related notes already in the library include Self-RAG and RAG basics.
• Limits: This is a pointer note only. Benchmark comparisons, including the LitQA human comparison, are the authors' reported results from the v2 paper (14 Dec 2023).
• License: arXiv default (non-exclusive distribution license) on the abs page (2026-11-11; http://arxiv.org/licenses/nonexclusive-distrib/1.0/). Submitted 8 Dec 2023 (v1); last revised 14 Dec 2023 (v2). Sources: [S1] https://arxiv.org/abs/2312.07559, v2 (14 Dec 2023). checked_on: 2026-11-11. Paper license: arXiv non-exclusive distribution; this note links to it and does not reproduce it. The note itself is CC BY 4.0.

When to apply

Use when looking at citation-backed literature agents, or at LitQA as a full-text scientific QA setting.

Facts

• id: F1 claim: 'PaperQA is a retrieval-augmented agent for answering questions over the scientific literature, including retrieval over full-text articles and relevance assessment of sources and passages.' confidence: medium date: 2026-11-11 sources: [S1] license: CC-BY-4.0
• id: F2 claim: 'The authors introduce LitQA, a benchmark that requires retrieval and synthesis from full-text scientific papers.' confidence: medium date: 2026-11-11 sources: [S1] license: CC-BY-4.0
• id: F3 claim: 'They report that PaperQA matches expert human researchers on LitQA.' confidence: medium date: 2026-11-11 sources: [S1] license: CC-BY-4.0

Sources

• [S1] arXiv 2312.07559, https://arxiv.org/abs/2312.07559. Paper license: arXiv default (non-exclusive distribution license). v2 (14 Dec 2023). Accessed 2026-11-11. This note links to the paper and does not reproduce it.

License: this note is CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Pointer by iinventedlight (AI-assisted). Paper license: arXiv non-exclusive distribution; this note links to it and does not reproduce it.

@misc{iil_note_wgng1826,
  author = {{'iinventedlight (AI-assisted)'}},
  title = {{PaperQA: Retrieval-Augmented Generative Agent for Scientific Research}},
  howpublished = {iinventedlight Library},
  year = {2026},
  date = {2026-11-11},
  url = {https://iinventedlight.com/library/notes/2312-07559-paperqa/},
  note = {Permanent ID iil:note:wgng1826. License: CC-BY-4.0.},
}

Markdown

Spotted a mistake? Use the MCP feedback tool or email hello@iinventedlight.com.

For language models and agents: start at /llms.txt | /agents.md | /.well-known/agent-library.json. This page in Markdown: /library/notes/2312-07559-paperqa.md.