MLDP: A Multimodal Learning Framework for Robust and Accurate Information Extraction from Tobacco Box Labels

MLDP: A Multimodal Learning Framework for Robust and Accurate Information Extraction from Tobacco Box Labels | 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 MLDP: A Multimodal Learning Framework for Robust and Accurate Information Extraction from Tobacco Box Labels Songling Huang, Shuai Zhang, Minghua Han, Jianping Yang, Yuyan Bai This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7596494/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract With the deepening of the smart re-baking factory and the accelerating digital transformation of the agricultural industry, tobacco re-baking, as the core link of raw materials for the cigarette process, is facing the urgent need to move from the traditional manual way to the intelligent automated system in terms of cigarette product quality traceability and information management, since the traditional way of relying on manual auditing has been difficult to meet the modern tobacco industry's high standards of efficiency, accuracy and traceability. In this paper, a novel multimodal learning framework, MLDP (Multimodal Learning for Document Processing), is proposed to address the bottlenecks in efficiency, accuracy, and robustness of traditional cigarette box label information extraction methods. The framework integrates the advantages of text modal and image modal, and integrates various DeBERTa variants such as BiLSTM-DeBERTa, Multi-Sample Dropout-DeBERTa, Distil-DeBERTa, etc., and achieves a balance between the model diversity and computational efficiency through the Optuna-driven weighted integration strategy. A balance between model diversity and computational efficiency is achieved; for image modality processing, PaddleOCR is used for character recognition, and a text correction mechanism based on Levenshtein distance is introduced to reduce the digit recognition error. In order to enhance the cross-modal synergy, MLDP designs a dynamic weighted fusion modal decision-making mechanism, and constructs a modal error correction matrix to optimize the information extraction results, so as to improve the robustness and accuracy in complex scenarios. The experiments are conducted on the multimodal dataset of cigarette box labels provided by the re-baking factory, and the results show that the MLDP framework improves the average F1 score by 4.34% compared to the text-only model (MLDP-T) and 5.73% compared to the image-only model (MLDP-V), and achieves a 1.18%~1.55% improvement in recognizing the non-textual elements in the DocBank complex document task, which verifies its strong generalization capability. Furthermore, comparative experiments with state-of-the-art models including LayoutLMv3, UDOP, and Donut demonstrate the superiority of MLDP, achieving an average F1-score of 96.51% and outperforming these strong baselines by 1.3% to 3.7%. The results validate the effectiveness of the dynamic multimodal fusion strategy, especially in handling noisy documents and recognizing numeric fields. In addition, this paper also constructed and open-sourced an industrial-grade cigarette box label multimodal dataset containing 9,719 graphic-aligned samples, which fills the gap of high-quality benchmark data in this field. Physical sciences/Engineering Physical sciences/Mathematics and computing PaddleOCR DeBERTa MLDP Bimodal information extraction Rebaking tobacco box labeling Full Text Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.zip Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 22 Feb, 2026 Reviews received at journal 15 Feb, 2026 Reviewers agreed at journal 04 Feb, 2026 Reviews received at journal 29 Jan, 2026 Reviewers agreed at journal 27 Jan, 2026 Reviewers agreed at journal 27 Jan, 2026 Reviewers agreed at journal 15 Nov, 2025 Reviewers invited by journal 12 Oct, 2025 Editor assigned by journal 12 Oct, 2025 Editor invited by journal 26 Sep, 2025 Submission checks completed at journal 18 Sep, 2025 First submitted to journal 18 Sep, 2025 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-7596494","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":533504969,"identity":"826cbc9d-9988-4848-af39-b6dc4a2911f9","order_by":0,"name":"Songling Huang","email":"","orcid":"","institution":"Yunnan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Songling","middleName":"","lastName":"Huang","suffix":""},{"id":533504970,"identity":"b2b5eb5a-876f-4f53-aaef-9dd6d31ee55f","order_by":1,"name":"Shuai Zhang","email":"","orcid":"","institution":"Yunnan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Shuai","middleName":"","lastName":"Zhang","suffix":""},{"id":533504971,"identity":"4a6434fc-89e0-44d3-8ec6-f12dc44739d0","order_by":2,"name":"Minghua Han","email":"","orcid":"","institution":"Yunnan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Minghua","middleName":"","lastName":"Han","suffix":""},{"id":533504972,"identity":"d0b41d4b-68ca-421e-9446-b8da985113a5","order_by":3,"name":"Jianping Yang","email":"","orcid":"","institution":"Yunnan Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Jianping","middleName":"","lastName":"Yang","suffix":""},{"id":533504973,"identity":"f5afcf60-b4ad-472f-9974-ec6a94568cfa","order_by":4,"name":"Yuyan Bai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuUlEQVRIiWNgGAWjYFACHgaGBAYGOTb25gOkaTHm4zmWQIIWIEicJ5GjQJwGg9u9Bz88bDuc3saQw8Dwo2IbEVrunEuWSGw7nNvGcPYAY8+Z20RouZFjANHC2JfAzNhGnBbjH0At6WzMPAZEazED2ZLAxkasFskbeWkWCefSDdt42BIOEuUXvhu5h2/+KLOWl5//+OCDHxVEaFE4ACQY2ZrBnAOE1QOBfAOI/FNHlOJRMApGwSgYoQAA37c/xMmI3W0AAAAASUVORK5CYII=","orcid":"","institution":"Yunnan Agricultural University","correspondingAuthor":true,"prefix":"","firstName":"Yuyan","middleName":"","lastName":"Bai","suffix":""}],"badges":[],"createdAt":"2025-09-12 04:53:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7596494/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7596494/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94397309,"identity":"a472d636-0549-4ea5-900e-5621b9110f24","added_by":"auto","created_at":"2025-10-27 13:56:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1352289,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReportsSubmission.docx","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/fa347d33bdc06c5cdc7f67ec.docx"},{"id":94396427,"identity":"28bfd65b-cb83-4e5a-a700-e9342053d3a8","added_by":"auto","created_at":"2025-10-27 13:55:59","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8358,"visible":true,"origin":"","legend":"","description":"","filename":"9df177541b974bcb8d9f9df72ce938b0.json","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/d2a58e2a9f6439218617efc9.json"},{"id":94398080,"identity":"cd081c10-22ea-4cfc-b464-5c5f1a1fbf93","added_by":"auto","created_at":"2025-10-27 13:56:58","extension":"zip","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":178310003,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.zip","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/ed01db96845a657df4416ea3.zip"},{"id":94398074,"identity":"fc6ce92c-72f6-40ff-90aa-030604abafb5","added_by":"auto","created_at":"2025-10-27 13:56:58","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":132167,"visible":true,"origin":"","legend":"","description":"","filename":"9df177541b974bcb8d9f9df72ce938b01enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/97f87ee0a189441821cc02b3.xml"},{"id":94397596,"identity":"2f7459ce-1769-4064-8b8e-be45337155f5","added_by":"auto","created_at":"2025-10-27 13:56:43","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87838,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/117a1aafcff7c500b671cbad.jpeg"},{"id":94397606,"identity":"3d74241c-990d-4c8f-8efd-9027294c1183","added_by":"auto","created_at":"2025-10-27 13:56:43","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":188164,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/a3f4a02fa8592f087762303b.png"},{"id":94397310,"identity":"587a45bd-2679-4eb2-9877-b5831ae006e5","added_by":"auto","created_at":"2025-10-27 13:56:37","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38279,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/a0b539eb97ad622a7321ca33.png"},{"id":94397322,"identity":"89446340-9aa4-4e78-a608-819baf2e4b47","added_by":"auto","created_at":"2025-10-27 13:56:37","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":367942,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/f2366b593c4e4028f3a8b0e4.png"},{"id":94397897,"identity":"6865d948-d1df-425a-b564-eff461ec67f8","added_by":"auto","created_at":"2025-10-27 13:56:53","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":85021,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/b1bac0b667eaa067b45c6f59.png"},{"id":94397312,"identity":"4119be53-781e-4602-8e90-e8e89a9a1040","added_by":"auto","created_at":"2025-10-27 13:56:37","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":90189,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/945411f59c977e3c99a30a83.jpeg"},{"id":94397303,"identity":"eda9419b-1433-4dfa-9d0d-cad0fbb10f43","added_by":"auto","created_at":"2025-10-27 13:56:36","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":122490,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/3dd6a473162b6a6c646153d6.png"},{"id":94395891,"identity":"35df7ed1-d7ae-49be-bd28-bd5cd6caf43f","added_by":"auto","created_at":"2025-10-27 13:55:43","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":200202,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/9bb709ce260a518cacf42690.png"},{"id":94397072,"identity":"ae4d1825-d1be-42b1-861c-2f38a854d89b","added_by":"auto","created_at":"2025-10-27 13:56:26","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":91246,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/0ab366a53670c8d6459042e1.png"},{"id":94398011,"identity":"8f844041-c3c4-41a6-bff6-7e2edc065862","added_by":"auto","created_at":"2025-10-27 13:56:56","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32998,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/addbd605da986ad46d3d4501.png"},{"id":94395903,"identity":"15ad3376-d89f-4de7-a260-cf337a278339","added_by":"auto","created_at":"2025-10-27 13:55:44","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21389,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/89cad7ef3472f216a0eae3dc.png"},{"id":94398485,"identity":"9e0d05ea-b07d-4419-b76b-ef78d8b1a158","added_by":"auto","created_at":"2025-10-27 13:57:06","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16481,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/30d4b55d42f9ee167828f3cb.png"},{"id":94398058,"identity":"7e43c6a6-f97d-4222-8ed8-aaf76f87f08c","added_by":"auto","created_at":"2025-10-27 13:56:57","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":50454,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/d949da05c4ba8c2d4e91540f.png"},{"id":94398464,"identity":"1dca0f29-def0-41df-9fc7-434628eb8d53","added_by":"auto","created_at":"2025-10-27 13:57:06","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19326,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/37456520935ec5f9b964d330.png"},{"id":94398399,"identity":"2afd1e40-7dad-48ee-9970-16ff50605e70","added_by":"auto","created_at":"2025-10-27 13:57:04","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":73105,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/afea1e3b6746b544224f3bc9.png"},{"id":94397154,"identity":"8f46ba1e-5b54-4ce7-acc5-569dd0daa4e6","added_by":"auto","created_at":"2025-10-27 13:56:29","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26609,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/f3526a554d1f94f0da403376.png"},{"id":94397005,"identity":"f021e945-8f86-459c-9a55-49825884d6da","added_by":"auto","created_at":"2025-10-27 13:56:23","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19111,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/6d23dfeba5cfa1433bd29a4d.png"},{"id":94396197,"identity":"42ce7664-b6d5-4921-99eb-fd9c04e9da9a","added_by":"auto","created_at":"2025-10-27 13:55:52","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24786,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/7d76b1d74952f2529e40f825.png"},{"id":94396195,"identity":"064f2713-84a3-4199-8763-6f7cbeebaa16","added_by":"auto","created_at":"2025-10-27 13:55:52","extension":"xml","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":128624,"visible":true,"origin":"","legend":"","description":"","filename":"9df177541b974bcb8d9f9df72ce938b01structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/8d7868aaebca60bcfb46916a.xml"},{"id":94397157,"identity":"08f26dc6-afd0-428a-99ed-66118b49f941","added_by":"auto","created_at":"2025-10-27 13:56:30","extension":"html","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":143465,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/fa9ea370ee7a96efb8af2fa7.html"},{"id":94459433,"identity":"c51ef0e1-9cbb-4378-b766-f9582afcc837","added_by":"auto","created_at":"2025-10-27 14:53:01","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1103894,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReportsSubmission.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1_covered_8ae3a349-8e91-4c39-b8ba-aa87ae56218f.pdf"},{"id":94398492,"identity":"bd1739f5-59fc-490f-99b3-d975007d84e4","added_by":"auto","created_at":"2025-10-27 13:57:07","extension":"zip","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":178310003,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.zip","url":"https://assets-eu.researchsquare.com/files/rs-7596494/v1/ed49e60040889e48de0599a2.zip"}],"financialInterests":"No competing interests reported.","formattedTitle":"MLDP: A Multimodal Learning Framework for Robust and Accurate Information Extraction from Tobacco Box Labels","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":"PaddleOCR, DeBERTa, MLDP, Bimodal information extraction, Rebaking tobacco box labeling","lastPublishedDoi":"10.21203/rs.3.rs-7596494/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7596494/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWith the deepening of the smart re-baking factory and the accelerating digital transformation of the agricultural industry, tobacco re-baking, as the core link of raw materials for the cigarette process, is facing the urgent need to move from the traditional manual way to the intelligent automated system in terms of cigarette product quality traceability and information management, since the traditional way of relying on manual auditing has been difficult to meet the modern tobacco industry's high standards of efficiency, accuracy and traceability. In this paper, a novel multimodal learning framework, MLDP (Multimodal Learning for Document Processing), is proposed to address the bottlenecks in efficiency, accuracy, and robustness of traditional cigarette box label information extraction methods. The framework integrates the advantages of text modal and image modal, and integrates various DeBERTa variants such as BiLSTM-DeBERTa, Multi-Sample Dropout-DeBERTa, Distil-DeBERTa, etc., and achieves a balance between the model diversity and computational efficiency through the Optuna-driven weighted integration strategy. A balance between model diversity and computational efficiency is achieved; for image modality processing, PaddleOCR is used for character recognition, and a text correction mechanism based on Levenshtein distance is introduced to reduce the digit recognition error. In order to enhance the cross-modal synergy, MLDP designs a dynamic weighted fusion modal decision-making mechanism, and constructs a modal error correction matrix to optimize the information extraction results, so as to improve the robustness and accuracy in complex scenarios. The experiments are conducted on the multimodal dataset of cigarette box labels provided by the re-baking factory, and the results show that the MLDP framework improves the average F1 score by 4.34% compared to the text-only model (MLDP-T) and 5.73% compared to the image-only model (MLDP-V), and achieves a 1.18%~1.55% improvement in recognizing the non-textual elements in the DocBank complex document task, which verifies its strong generalization capability. Furthermore, comparative experiments with state-of-the-art models including LayoutLMv3, UDOP, and Donut demonstrate the superiority of MLDP, achieving an average F1-score of 96.51% and outperforming these strong baselines by 1.3% to 3.7%. The results validate the effectiveness of the dynamic multimodal fusion strategy, especially in handling noisy documents and recognizing numeric fields. In addition, this paper also constructed and open-sourced an industrial-grade cigarette box label multimodal dataset containing 9,719 graphic-aligned samples, which fills the gap of high-quality benchmark data in this field.\u003c/p\u003e","manuscriptTitle":"MLDP: A Multimodal Learning Framework for Robust and Accurate Information Extraction from Tobacco Box Labels","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-25 23:10:54","doi":"10.21203/rs.3.rs-7596494/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-23T04:39:03+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-15T15:30:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"137239922388821205971745356315128565812","date":"2026-02-04T06:36:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-29T08:25:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"101326079323803198777200233894279617662","date":"2026-01-28T03:18:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"212678055182318319484859655490026768505","date":"2026-01-27T12:27:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"141608433062850552834444955934803440138","date":"2025-11-15T16:40:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-12T15:40:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-12T15:39:27+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-09-26T12:56:30+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-18T06:23:08+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-09-18T05:57:04+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":"9d09d351-ab32-4273-83a3-62ce38a4a62b","owner":[],"postedDate":"October 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":56723009,"name":"Physical sciences/Engineering"},{"id":56723010,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-16T06:54:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-25 23:10:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7596494","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7596494","identity":"rs-7596494","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}