CASE: Confusion-Aware Semantic Enhancement for Multi-Object Text-to-Image Generation

CASE: Confusion-Aware Semantic Enhancement for Multi-Object Text-to-Image Generation | 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 Research Article CASE: Confusion-Aware Semantic Enhancement for Multi-Object Text-to-Image Generation Xiaopeng Cao, Tian Zhang, Hailong Ning, Chunyang Zhao, Yizhuo Dong, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9109704/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Diffusion models have significantly improved the visual quality of text-to-image (T2I) synthesis.However, representative models such as Stable Diffusion still suffer from object omission in multi-object scenarios.First, geometric relationships among objects in the embedding space may induce semantic confusion, which hinders the model’s ability to distinguish semantically similar objects.Second, the model often overemphasizes certain objects, which leads to imbalanced attention allocation across multiple objects.To address these issues, we propose a Confusion-Aware Semantic Enhancement (CASE) approach for T2I generation.To mitigate semantic confusion in the embedding space, we design a Confusion-Aware Embedding Decoupling (CAED) mechanism to enhance the semantic separability of geometrically proximate objects.By explicitly enlarging inter-object embedding distances, CAED strengthens the model’s ability to capture the structural semantics of multi-object prompts.To address imbalanced attention allocation in multi-object scenarios, we propose a Confusion-Aware Attention Separation (CAAS) mechanism that enhances object discriminability during denoising and encourages more stable attention distributions.Extensive experiments on multiple T2I benchmarks and different versions of Stable Diffusion demonstrate consistent improvements across a wide range of evaluation metrics. Text-to-Image Generation Diffusion Models Multi-Object Prompts Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 05 May, 2026 Reviews received at journal 01 May, 2026 Reviews received at journal 18 Apr, 2026 Reviewers agreed at journal 24 Mar, 2026 Reviewers agreed at journal 24 Mar, 2026 Reviewers agreed at journal 23 Mar, 2026 Reviewers invited by journal 23 Mar, 2026 Editor assigned by journal 19 Mar, 2026 Submission checks completed at journal 17 Mar, 2026 First submitted to journal 12 Mar, 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-9109704","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":611685761,"identity":"2cb6a66a-9e3b-4de6-b00c-92ae47740e19","order_by":0,"name":"Xiaopeng Cao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvklEQVRIiWNgGAWjYHACxsd/KiTk2NjbDxCthdmA54yNMR/PmQSitbBJ8LakJc6TcDAgTj3/jNwDEpINh9PbJBgSGH5UbCOsReLMuQQDwx2Hc9ukGw8w9py5TViLAXuPQULiGaAWmQMJzIxtxGhh5jE4cLDtcDqbRIIBkVrYewwbG9vSEojXAvRLMjPDGRvDNmAgHyTKL8AQO/6boUJCXr69/eCDHxVEaGFg4EEwDxCjHlXLKBgFo2AUjAKsAABypzvbfBTcDAAAAABJRU5ErkJggg==","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":true,"prefix":"","firstName":"Xiaopeng","middleName":"","lastName":"Cao","suffix":""},{"id":611685764,"identity":"7cc7ef16-9220-4558-a5b5-bbb6c4b8df2f","order_by":1,"name":"Tian Zhang","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Tian","middleName":"","lastName":"Zhang","suffix":""},{"id":611685769,"identity":"5490f3f2-791c-4a07-8cbf-1270ed4c2d9f","order_by":2,"name":"Hailong Ning","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Hailong","middleName":"","lastName":"Ning","suffix":""},{"id":611685777,"identity":"b7ea880c-0fc4-4912-b1c2-341e42e72a84","order_by":3,"name":"Chunyang Zhao","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Chunyang","middleName":"","lastName":"Zhao","suffix":""},{"id":611685784,"identity":"38f8285c-c79c-4dbd-b32c-95efe0c1bde4","order_by":4,"name":"Yizhuo Dong","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Yizhuo","middleName":"","lastName":"Dong","suffix":""},{"id":611685786,"identity":"6479698c-4bda-4e3a-bec1-507fb21c70ac","order_by":5,"name":"Haojie Li","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Haojie","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2026-03-13 03:23:41","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9109704/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9109704/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105565074,"identity":"3c153566-08d1-4ac5-82d5-a1b0eaad817e","added_by":"auto","created_at":"2026-03-27 12:51:47","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4073203,"visible":true,"origin":"","legend":"","description":"","filename":"ztsnarticletemplate.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9109704/v1_covered_e0b69c3e-4918-4318-ba2e-dca7d898e6d3.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"CASE: Confusion-Aware Semantic Enhancement for Multi-Object Text-to-Image Generation","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":"international-journal-of-machine-learning-and-cybernetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jmlc","sideBox":"Learn more about [International Journal of Machine Learning and Cybernetics](http://actavetscand.biomedcentral.com/)","snPcode":"13042","submissionUrl":"https://submission.nature.com/new-submission/13042/3","title":"International Journal of Machine Learning and Cybernetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Text-to-Image Generation, Diffusion Models, Multi-Object Prompts","lastPublishedDoi":"10.21203/rs.3.rs-9109704/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9109704/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDiffusion models have significantly improved the visual quality of text-to-image (T2I) synthesis.However, representative models such as Stable Diffusion still suffer from object omission in multi-object scenarios.First, geometric relationships among objects in the embedding space may induce semantic confusion, which hinders the model\u0026rsquo;s ability to distinguish semantically similar objects.Second, the model often overemphasizes certain objects, which leads to imbalanced attention allocation across multiple objects.To address these issues, we propose a Confusion-Aware Semantic Enhancement (CASE) approach for T2I generation.To mitigate semantic confusion in the embedding space, we design a Confusion-Aware Embedding Decoupling (CAED) mechanism to enhance the semantic separability of geometrically proximate objects.By explicitly enlarging inter-object embedding distances, CAED strengthens the model\u0026rsquo;s ability to capture the structural semantics of multi-object prompts.To address imbalanced attention allocation in multi-object scenarios, we propose a Confusion-Aware Attention Separation (CAAS) mechanism that enhances object discriminability during denoising and encourages more stable attention distributions.Extensive experiments on multiple T2I benchmarks and different versions of Stable Diffusion demonstrate consistent improvements across a wide range of evaluation metrics.\u003c/p\u003e","manuscriptTitle":"CASE: Confusion-Aware Semantic Enhancement for Multi-Object Text-to-Image Generation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-25 15:19:37","doi":"10.21203/rs.3.rs-9109704/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"201821981329256958703665496784880877786","date":"2026-05-05T12:47:54+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-01T17:24:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-18T13:30:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"202570317010249734721688627303809595585","date":"2026-03-25T02:31:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"126038202741333958977626493395124171922","date":"2026-03-24T05:01:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"56202950397980905463678111510577127773","date":"2026-03-24T02:37:12+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-24T02:31:22+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-20T02:57:47+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-17T13:31:47+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Machine Learning and Cybernetics","date":"2026-03-13T03:19:14+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-machine-learning-and-cybernetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jmlc","sideBox":"Learn more about [International Journal of Machine Learning and Cybernetics](http://actavetscand.biomedcentral.com/)","snPcode":"13042","submissionUrl":"https://submission.nature.com/new-submission/13042/3","title":"International Journal of Machine Learning and Cybernetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"42f5880c-0805-423b-93ad-dada23cbde54","owner":[],"postedDate":"March 25th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"201821981329256958703665496784880877786","date":"2026-05-05T12:47:54+00:00","index":16,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-01T17:24:26+00:00","index":14,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-25T15:19:37+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-25 15:19:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9109704","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9109704","identity":"rs-9109704","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}