Green AI for Sustainable Question Answering: Carbon-Aware Fine-Tuning and Retrieval-Augmented Generation at Scale

Green AI for Sustainable Question Answering: Carbon-Aware Fine-Tuning and Retrieval-Augmented Generation at Scale | 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 Green AI for Sustainable Question Answering: Carbon-Aware Fine-Tuning and Retrieval-Augmented Generation at Scale Tarunjit Yumnam, Sumegh Tharewal, Amit Kumar Sahu, Timothy Malche This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8765202/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 10 You are reading this latest preprint version Abstract Large Language Models have become a gem of the 21st century in modern Artificial Intelligence applications, however their environmental impact has raised concerns. This research compares the carbon footprints of three approaches for adapting large language models to new domains, Retrieval Augmented Generation (RAG), Full fine-tuning for three epochs (Full fine-tuning) and Parameter-efficient Fine-tuning (PEFT). RAG and LLM Fine-tuning are two knowledge adaptation techniques widely used to build a domain specific question-answering (QA) model and to improve environmental impact and user experience mitigation techniques such as parameter-efficient fine-tuning (PEFT), quantization and retrieval optimization will be discussed in this proposed system. This paper proposed a break-even analysis across varying query volumes and update frequencies to determine the carbon efficiencies of each approach. This proposed system investigates CO2 emissions by fine-tuning two different LLM models namely t5-small and DistilBERT with three different knowledge adaptation methods, namely, LLM full fine-tuning, LLM LoRA fine-tuning, and LLM-RAG model with a SQuAD QA dataset and provided various mitigation strategies to reduce CO₂ emissions without compromising model quality. The proposed system highlights the CO2 emission of t5-small LoRA fine-tuning is lowest among t5-small fine-tuning methods and DistilBERT LoRA fine-tuning is lowest among DistilBERT fine-tuning methods. The t5-small LoRA fine-tuning recorded 37.13% less carbon emission with compare to the DistilBERT LoRA fine-tuning. This research work finds a way to fine-tune the LLMs model with freely available GPUs, while the actual price to buy the A100 and T4 GPU is very costly. Physical sciences/Engineering Physical sciences/Mathematics and computing Sustainable Large Language Model (LLM) Stanford Question Answering Dataset (SQuAD) Low-Rank Adaptation (LoRA) Green AI Metrics Retrieval-Augmented Generation (RAG) Vector Database Optimization. Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 14 May, 2026 Reviews received at journal 02 Mar, 2026 Reviewers agreed at journal 24 Feb, 2026 Reviews received at journal 24 Feb, 2026 Reviewers agreed at journal 24 Feb, 2026 Reviewers invited by journal 24 Feb, 2026 Editor assigned by journal 14 Feb, 2026 Editor invited by journal 11 Feb, 2026 Submission checks completed at journal 04 Feb, 2026 First submitted to journal 04 Feb, 2026 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-8765202","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":596649977,"identity":"3988a1f6-f855-4ad1-8a40-453c0dcecf99","order_by":0,"name":"Tarunjit Yumnam","email":"","orcid":"","institution":"DBS Global University","correspondingAuthor":false,"prefix":"","firstName":"Tarunjit","middleName":"","lastName":"Yumnam","suffix":""},{"id":596649978,"identity":"452c4358-1730-41ee-953b-35208bad73eb","order_by":1,"name":"Sumegh Tharewal","email":"","orcid":"","institution":"DBS Global University","correspondingAuthor":false,"prefix":"","firstName":"Sumegh","middleName":"","lastName":"Tharewal","suffix":""},{"id":596649979,"identity":"600376c8-7e43-4d26-b5a0-dd9af87c1cbb","order_by":2,"name":"Amit Kumar Sahu","email":"","orcid":"","institution":"Amit Kumar Sahu, IT Lead Consultant, EBOS Group Limited","correspondingAuthor":false,"prefix":"","firstName":"Amit","middleName":"Kumar","lastName":"Sahu","suffix":""},{"id":596649980,"identity":"d3fa1f2e-25aa-4252-95fa-94f151070198","order_by":3,"name":"Timothy Malche","email":"data:image/png;base64,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","orcid":"","institution":"Manipal University Jaipur","correspondingAuthor":true,"prefix":"","firstName":"Timothy","middleName":"","lastName":"Malche","suffix":""}],"badges":[],"createdAt":"2026-02-02 12:54:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8765202/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8765202/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104398566,"identity":"ef21f1c7-2832-48a4-9903-526d57ce5e33","added_by":"auto","created_at":"2026-03-11 12:02:57","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":882955,"visible":true,"origin":"","legend":"","description":"","filename":"GreenAIforSustainableQuestionAnsweringCarbonAwareFineTuningandRetrievalAugmentedGenerationatScale.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8765202/v1_covered_75662780-4bcb-4609-8302-ac0eaf63fde8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Green AI for Sustainable Question Answering: Carbon-Aware Fine-Tuning and Retrieval-Augmented Generation at Scale","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Sustainable Large Language Model (LLM), Stanford Question Answering Dataset (SQuAD), Low-Rank Adaptation (LoRA), Green AI Metrics, Retrieval-Augmented Generation (RAG), Vector Database Optimization.","lastPublishedDoi":"10.21203/rs.3.rs-8765202/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8765202/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Large Language Models have become a gem of the 21st century in modern Artificial Intelligence applications, however their environmental impact has raised concerns. This research compares the carbon footprints of three approaches for adapting large language models to new domains, Retrieval Augmented Generation (RAG), Full fine-tuning for three epochs (Full fine-tuning) and Parameter-efficient Fine-tuning (PEFT). RAG and LLM Fine-tuning are two knowledge adaptation techniques widely used to build a domain specific question-answering (QA) model and to improve environmental impact and user experience mitigation techniques such as parameter-efficient fine-tuning (PEFT), quantization and retrieval optimization will be discussed in this proposed system. This paper proposed a break-even analysis across varying query volumes and update frequencies to determine the carbon efficiencies of each approach. This proposed system investigates CO2 emissions by fine-tuning two different LLM models namely t5-small and DistilBERT with three different knowledge adaptation methods, namely, LLM full fine-tuning, LLM LoRA fine-tuning, and LLM-RAG model with a SQuAD QA dataset and provided various mitigation strategies to reduce CO₂ emissions without compromising model quality. The proposed system highlights the CO2 emission of t5-small LoRA fine-tuning is lowest among t5-small fine-tuning methods and DistilBERT LoRA fine-tuning is lowest among DistilBERT fine-tuning methods. The t5-small LoRA fine-tuning recorded 37.13% less carbon emission with compare to the DistilBERT LoRA fine-tuning. This research work finds a way to fine-tune the LLMs model with freely available GPUs, while the actual price to buy the A100 and T4 GPU is very costly.","manuscriptTitle":"Green AI for Sustainable Question Answering: Carbon-Aware Fine-Tuning and Retrieval-Augmented Generation at Scale","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-27 14:28:30","doi":"10.21203/rs.3.rs-8765202/v1","editorialEvents":[{"type":"communityComments","content":1},{"type":"decision","content":"Revision requested","date":"2026-05-14T05:45:36+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-02T05:40:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"19581665526592865087260044748367309544","date":"2026-02-25T04:10:54+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-25T04:10:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"42154360791218199447661283421880785755","date":"2026-02-25T03:56:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-24T12:05:36+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-14T05:57:36+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-11T12:12:50+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-04T12:13:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-02-04T12:01:01+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9a940912-321f-4ea2-9a93-5b895ca541e7","owner":[],"postedDate":"February 27th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-14T05:45:36+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[{"id":63484979,"name":"Physical sciences/Engineering"},{"id":63484980,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-14T05:54:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-27 14:28:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8765202","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8765202","identity":"rs-8765202","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}