Automated LLM based Extraction of Standardized Synthesis Procedures: an All-Domain, Zero-Shot Approach

Automated LLM based Extraction of Standardized Synthesis Procedures: an All-Domain, Zero-Shot Approach | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Automated LLM based Extraction of Standardized Synthesis Procedures: an All-Domain, Zero-Shot Approach Pedro Mendes, Daniel Costa, Matteo Manica, Teodoro Laino, Filipa Ribeiro This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7860460/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Millions of synthesis procedures have been published, but their meta-analysis (e.g., to identify successful synthesis patterns) is troublesome due to the variety of reporting structures. Mapping unstructured language to a precise sequence of actions requires an understanding of domain-specific jargon, a challenge usually addressed by fine-tuning models on labeled data. Herein, we present a simple, training-free workflow for laboratory action extraction that works across fields. It encodes domain knowledge through explicit action sets and uses powerful, readily available Large Language Models (LLMs). Applied to zeolite synthesis, our approach outperforms existing open-source tools and delivers context-aware results with open, locally runnable LLMs. It also matches the performance of state-of-the-art, field-specific models in their own domains, highlighting the generalization ability of current LLMs. With our methodology and open algorithms, chemists can evaluate actions sets, screen LLMs for their specific needs, and accurately digitize laboratory procedures with minimal effort. Scientific community and society/Scientific community/Research data Physical sciences/Chemistry/Cheminformatics Physical sciences/Chemistry/Chemistry publishing Full Text Additional Declarations There is NO Competing Interest. Supplementary Files CostaetalLLM4synthesisESI.pdf Electronic Supplementary Information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7860460","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":537497351,"identity":"9693171d-245d-4a14-b1d0-c75fd50a835b","order_by":0,"name":"Pedro Mendes","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-2768-353X","institution":"Instituto Superior Técnico","correspondingAuthor":true,"prefix":"","firstName":"Pedro","middleName":"","lastName":"Mendes","suffix":""},{"id":537497352,"identity":"e862ab50-0301-41a1-be7a-5b0d29126cf4","order_by":1,"name":"Daniel Costa","email":"","orcid":"","institution":"Instituto Superior Técnico","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Costa","suffix":""},{"id":537497353,"identity":"8ea9f793-3a54-422e-977c-39f71e988c29","order_by":2,"name":"Matteo Manica","email":"","orcid":"","institution":"IBM Research Europe","correspondingAuthor":false,"prefix":"","firstName":"Matteo","middleName":"","lastName":"Manica","suffix":""},{"id":537497354,"identity":"fe1cbc35-4e58-46ea-b3e0-254621874b0b","order_by":3,"name":"Teodoro Laino","email":"","orcid":"https://orcid.org/0000-0001-8717-0456","institution":"IBM Research Europe","correspondingAuthor":false,"prefix":"","firstName":"Teodoro","middleName":"","lastName":"Laino","suffix":""},{"id":537497355,"identity":"01453312-08fc-4737-92d9-621f00e9443e","order_by":4,"name":"Filipa Ribeiro","email":"","orcid":"","institution":"Instituto Superior Técnico","correspondingAuthor":false,"prefix":"","firstName":"Filipa","middleName":"","lastName":"Ribeiro","suffix":""}],"badges":[],"createdAt":"2025-10-14 16:00:56","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7860460/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7860460/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96243408,"identity":"e5014a48-ceee-466d-9871-0efc20edca55","added_by":"auto","created_at":"2025-11-19 07:16:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1952721,"visible":true,"origin":"","legend":"Article File","description":"","filename":"CostaetalLLM4synthesismanuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7860460/v1_covered_380661d8-287c-4785-85e7-47a5ab04dc8e.pdf"},{"id":95908876,"identity":"75891e8a-ed37-43fb-b214-1713e680c7a6","added_by":"auto","created_at":"2025-11-14 10:00:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1391778,"visible":true,"origin":"","legend":"Electronic Supplementary Information","description":"","filename":"CostaetalLLM4synthesisESI.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7860460/v1/d593e0ff442ccb9c57072bc6.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Automated LLM based Extraction of Standardized Synthesis Procedures: an All-Domain, Zero-Shot Approach","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7860460/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7860460/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Millions of synthesis procedures have been published, but their meta-analysis (e.g., to identify successful synthesis patterns) is troublesome due to the variety of reporting structures. Mapping unstructured language to a precise sequence of actions requires an understanding of domain-specific jargon, a challenge usually addressed by fine-tuning models on labeled data. Herein, we present a simple, training-free workflow for laboratory action extraction that works across fields. It encodes domain knowledge through explicit action sets and uses powerful, readily available Large Language Models (LLMs). \r\nApplied to zeolite synthesis, our approach outperforms existing open-source tools and delivers context-aware results with open, locally runnable LLMs. It also matches the performance of state-of-the-art, field-specific models in their own domains, highlighting the generalization ability of current LLMs. With our methodology and open algorithms, chemists can evaluate actions sets, screen LLMs for their specific needs, and accurately digitize laboratory procedures with minimal effort.","manuscriptTitle":"Automated LLM based Extraction of Standardized Synthesis Procedures: an All-Domain, Zero-Shot Approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-14 10:00:04","doi":"10.21203/rs.3.rs-7860460/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"96385eb5-1505-42c1-a2f9-0cc38def27bb","owner":[],"postedDate":"November 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":57162749,"name":"Scientific community and society/Scientific community/Research data"},{"id":57162750,"name":"Physical sciences/Chemistry/Cheminformatics"},{"id":57162751,"name":"Physical sciences/Chemistry/Chemistry publishing"}],"tags":[],"updatedAt":"2025-11-14T10:00:04+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-14 10:00:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7860460","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7860460","identity":"rs-7860460","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}