EMBL AI Librarian: Life-Sciences Knowledge Layer for AI Agents

Source

Source name: arXiv
Source kind: paper
Authors / organization: Luigi Sigillo, Matteo Silvestri, Francesco Tabaro, Rajat Bhatnagar, Syed Irtaza Mubashar, Matt Jeffryes, Daljit Nijjer, Vittorio Perera, Ola Spjuth, Julio Saez-Rodriguez, Melissa Harrison, Fabio Petroni
Venue: arXiv: cs.CL
Published: 2026-07-30
Collected: 2026-07-31
URL: http://arxiv.org/abs/2607.28229v1
PDF: https://arxiv.org/pdf/2607.28229v1
DOI: UNKNOWN

Plain-Language Brief

The paper introduces EMBL AI Librarian, a system that enhances the Europe PMC interface for AI agents to enable them to search and retrieve evidence in natural language. The system was evaluated across four benchmarks and showed improvements in various tasks related to literature synthesis, claim verification, open-domain question answering, and downstream biology tasks.

What the Document Says

The web is accessed increasingly by AI agents.
Life-sciences require knowledge for AI agents, especially with the growth of agentic pipelines.
Europe PMC, a resource with over 40M indexed records, is used to meet this need but was not built for AI agents.
EMBL AI Librarian upgrades Europe PMC interface for AI agents: an agent asks in natural language and receives evidence that answers it.
An LLM orchestrates the whole knowledge retrieval process.
The system was evaluated across four benchmarks: literature synthesis, claim verification, open-domain question answering, and downstream biology tasks such as protocol questions and sequence manipulation.
On ScholarQABench, Librarian improves Citation F1 by more than $16$ points over strong recently published baselines.
Used as the retrieval layer of an existing claim-verification pipeline, it increases agreement with expert consensus.
On the open-form LitQA2 benchmark, a GPT-5.4 agent scores about $8$ points higher when grounded in Librarian than with web search.
The results show that equipping life-science agents with the Librarian knowledge layer improves performance across a range of tasks.
The code is publicly available at https://github.com/petroni-lab/librarian

Key Concepts Mentioned

AI agents
life-sciences
Europe PMC
EMBL AI Librarian
LLM
knowledge layer
literature synthesis
claim verification
open-domain question answering
downstream biology tasks

Claims Made by the Document

Improved performance across a range of tasks when equipping life-science agents with the Librarian knowledge layer.

Evidence Used

Improvements in ScholarQABench, claim-verification pipeline, open-form LitQA2 benchmark

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

The web is increasingly accessed by AI agents rather than humans. Every agent needs knowledge, especially in the life-sciences, where agentic pipelines are growing fast. Access to the literature is a crucial part of that need, and resources such as Europe PMC, with over 40M indexed records, are widely used to meet it. Yet these resources were not built for AI agents: they take keywords and complex syntax and return whole papers, so every agent must learn the syntax, issue several searches, and read full papers to find the evidence it needs. We introduce EMBL AI Librarian, a knowledge layer that upgrades the Europe PMC interface for AI agents: an agent asks in natural language and receives evidence that answers it. A single LLM orchestrates the whole knowledge retrieval process: it plans complementary subqueries executed by the live Europe PMC search engine, then reads the selected papers and locates the relevant evidence. We evaluate Librarian across four benchmarks: literature synthesis, claim verification, open-domain question answering, and downstream biology tasks such as protocol questions and sequence manipulation. On ScholarQABench, Librarian improves Citation F1 by more than $16$ points over strong recently published baselines. Used as the retrieval layer of an existing claim-verification pipeline, it increases agreement with expert consensus; and on the open-form LitQA2 benchmark, a GPT-5.4 agent scores about $8$ points higher when grounded in Librarian than with web search. Overall, our results show that equipping life-science agents with the Librarian knowledge layer improves performance across a range of tasks. We release our code publicly at https://github.com/petroni-lab/librarian