When a user uploads a file (CSV, Excel, or PDF), LibScience automatically extracts the bibliographic references and analyzes them using a parser that identifies the key elements of each reference: title, authors, year, and DOI. This structured data is then cross-referenced with trusted external sources such as PubMed, Crossref, or OpenAlex to locate the corresponding real reference. A scoring algorithm then compares the input data with the data retrieved from these reliable sources, normalizing formats to avoid false mismatches, and calculates an overall confidence score based on the title, authors, and DOI. This score allows users to instantly identify correct, questionable, or unfindable references.
LibScience maintains a 99.9% accuracy rate in detecting reference errors, missing citations, and formatting issues. Our AI is trained on millions of academic papers and continuously improves with every manuscript processed. We also provide detailed reports so you can verify our findings.
Absolutely. We use enterprise-grade encryption for all data in transit and at rest. LibScience is GDPR compliant. Your manuscripts are never used to train our models without explicit permission, and you maintain full control over your data with options to delete it at any time.
We support over 10,000 citation styles including all major formats like APA, MLA, Chicago, IEEE, Harvard, Vancouver, and many more. You can also create custom citation rules specific to your journal’s requirements.
Yes. LibScience can be used as a reference quality-control step before a manuscript is submitted, reviewed, or published. Upload your document or bibliography, check DOIs, PMIDs, metadata, and retracted sources, then correct any issues before editorial approval. Our team supports you with a simple and fast setup, without disrupting your usual workflow.
Blockchain is a technology for storing and transmitting information that is transparent, secure, and operates without a central authority. It enables the creation of decentralized and immutable records.
LibScience uses the Polygon blockchain, a “layer 2” of the Ethereum blockchain, to ensure transparency, traceability, and security for scientific publications. Smart contracts, written in Solidity, manage interactions and transactions in an automated and secure way.
A smart contract is a computer program that executes automatically when predefined conditions are met. On LibScience, smart contracts manage publications, access rights, and transactions securely and transparently.
Using blockchain for scientific publishing ensures the integrity and traceability of publications, reduces publishing costs, and promotes open and collaborative science.
A “layer 2” is a solution built on top of an existing blockchain (such as Ethereum) to improve efficiency and reduce transaction costs. Polygon is a “layer 2” that makes Ethereum more efficient and cost-effective.
Blockchain guarantees publication security through its decentralized and immutable nature. Each transaction is verified and permanently recorded, making data inviolable and transparent.
Blockchain offers researchers a secure and transparent platform to publish their work, retain intellectual property, and collaborate effectively with other researchers.
To start using blockchain on LibScience, simply create an account on our platform and follow the instructions to publish your work. We also provide educational resources and technical support to guide you through each step.
Commencez à vérifier vos références scientifiques avec LibScience et publiez en toute confiance.
They support us
This project was funded by the government as part of the All Deeptech 2 Intervention Guarantee Fund.
This project was funded by the government as part of the All Deeptech 2 Intervention Guarantee Fund.