A Discourse-Aware Graph–Text Framework for Coherent and Faithful Summarization of Long Legal Documents

A Discourse-Aware Graph–Text Framework for Coherent and Faithful Summarization of Long Legal Documents | 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 A Discourse-Aware Graph–Text Framework for Coherent and Faithful Summarization of Long Legal Documents Amitha S, Kayarvizhy N, Samraat Dabolay, Shlok Shivaram Iyer, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8767898/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Summarization systems often fail to maintain coherence and ability to track facts when it comes to long documents. This can be particularly seen in legal and governmental reports since there is a need to reason across multiple sections and there are also a lot of underlying factual contexts. This work proposes a hybrid approach which consists of a framework that is aware of all the facts using it to chunk the data and uses hierarchical attention and fact based graph to capture relations. It does this across sentences, chunks, and even from tuples extracted from the facts. We also propose a decoding mechanism that is aware of the citations and uses that to align the generated statement with the evidence that supports it. This significantly improves interpretability and also reduces unsupported content. The framework is evaluated on the GovReport dataset and it has achieved good performance like 52.80 ROUGE-1, 21.50 ROUGE-2, and 24.40 ROUGE-L against other baseline long document summarizers. It does this with improved factual grounding evidenced by a 93.9% fact similarity score. It also averages 11 sentences supported by evidence for each summary. These results confirm that the proposed framework leads to a more faithful summaries which is coherent and also is supported strongly by the facts. Biological sciences/Computational biology and bioinformatics Physical sciences/Mathematics and computing Long-Document Summarization Discourse Chunking Hierarchical Attention Graph-Based Reasoning Citation Alignment Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted 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-8767898","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":597134781,"identity":"3173165f-51d7-41d1-89ec-18dfe748eb78","order_by":0,"name":"Amitha S","email":"","orcid":"","institution":"BMS College of Engineering","correspondingAuthor":false,"prefix":"","firstName":"Amitha","middleName":"","lastName":"S","suffix":""},{"id":597134784,"identity":"2efe6151-01df-474d-acfa-4824c60bda2a","order_by":1,"name":"Kayarvizhy N","email":"data:image/png;base64,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","orcid":"","institution":"BMS College of Engineering Bengaluru","correspondingAuthor":true,"prefix":"","firstName":"Kayarvizhy","middleName":"","lastName":"N","suffix":""},{"id":597134788,"identity":"9530bcc8-44f9-462e-8a01-9f54cc6bd6dd","order_by":2,"name":"Samraat Dabolay","email":"","orcid":"","institution":"BMS College of Engineering Bengaluru","correspondingAuthor":false,"prefix":"","firstName":"Samraat","middleName":"","lastName":"Dabolay","suffix":""},{"id":597134790,"identity":"e10a6648-e289-46a3-a8a3-e2b9eb44e67f","order_by":3,"name":"Shlok Shivaram Iyer","email":"","orcid":"","institution":"BMS College of Engineering Bengaluru","correspondingAuthor":false,"prefix":"","firstName":"Shlok","middleName":"Shivaram","lastName":"Iyer","suffix":""},{"id":597134791,"identity":"31b49194-d1ea-4438-b340-335543a58724","order_by":4,"name":"Sneha N Shastri","email":"","orcid":"","institution":"BMS College of Engineering Bengaluru","correspondingAuthor":false,"prefix":"","firstName":"Sneha","middleName":"N","lastName":"Shastri","suffix":""},{"id":597134792,"identity":"c3d52e16-5052-475e-bbe5-dd6398852967","order_by":5,"name":"Varsha P","email":"","orcid":"","institution":"BMS College of Engineering Bengaluru","correspondingAuthor":false,"prefix":"","firstName":"Varsha","middleName":"","lastName":"P","suffix":""}],"badges":[],"createdAt":"2026-02-02 17:53:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8767898/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8767898/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105442661,"identity":"f7802151-cee6-4d60-b6f5-a6581425d0b0","added_by":"auto","created_at":"2026-03-26 06:27:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":737632,"visible":true,"origin":"","legend":"","description":"","filename":"ADiscourseAwareGraphTextFrameworkforCoherentandFaithfulSummarizationofLongLegalDocuments.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8767898/v1_covered_7696a3a2-f13f-4bee-b98c-3da92a1a21d1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Discourse-Aware Graph–Text Framework for Coherent and Faithful Summarization of Long Legal Documents","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Long-Document Summarization, Discourse Chunking, Hierarchical Attention, Graph-Based Reasoning, Citation Alignment","lastPublishedDoi":"10.21203/rs.3.rs-8767898/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8767898/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSummarization systems often fail to maintain coherence and ability to track facts when it comes to long documents. This can be particularly seen in legal and governmental reports since there is a need to reason across multiple sections and there are also a lot of underlying factual contexts. This work proposes a hybrid approach which consists of a framework that is aware of all the facts using it to chunk the data and uses hierarchical attention and fact based graph to capture relations. It does this across sentences, chunks, and even from tuples extracted from the facts. We also propose a decoding mechanism that is aware of the citations and uses that to align the generated statement with the evidence that supports it. This significantly improves interpretability and also reduces unsupported content.\u003c/p\u003e \u003cp\u003eThe framework is evaluated on the GovReport dataset and it has achieved good performance like 52.80 ROUGE-1, 21.50 ROUGE-2, and 24.40 ROUGE-L against other baseline long document summarizers. It does this with improved factual grounding evidenced by a 93.9% fact similarity score. It also averages 11 sentences supported by evidence for each summary. These results confirm that the proposed framework leads to a more faithful summaries which is coherent and also is supported strongly by the facts.\u003c/p\u003e","manuscriptTitle":"A Discourse-Aware Graph–Text Framework for Coherent and Faithful Summarization of Long Legal Documents","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-26 16:18:01","doi":"10.21203/rs.3.rs-8767898/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a9fd387c-83ed-4571-b8a6-356836e7355b","owner":[],"postedDate":"February 26th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":63543603,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":63543604,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-03-26T06:25:49+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-26 16:18:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8767898","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8767898","identity":"rs-8767898","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}