TrustTrade: Human-Inspired Selective Consensus Reduces Decision Uncertainty in LLM Trading Agents

TrustTrade: Human-Inspired Selective Consensus Reduces Decision Uncertainty in LLM Trading Agents | 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 TrustTrade: Human-Inspired Selective Consensus Reduces Decision Uncertainty in LLM Trading Agents Minghan LI, Rachel Gonsalves, Weiyue Li, Sunghoon Yoon, Mengyu Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9334812/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Large language models (LLMs) are increasingly deployed as autonomous agents in financial trading. However, they often exhibit a hazardous behavioral bias that we term uniform trust, whereby retrieved information is implicitly assumed to be factual and heterogeneous sources are treated as equally informative. This assumption stands in sharp contrast to human decision-making, which relies on selective filtering, cross-validation, and experience-driven weighting of information sources. As a result, LLM-based trading systems are particularly vulnerable to multi-source noise and misinformation, amplifying factual hallucinations and leading to unstable risk–return performance. To bridge this behavioral gap, we introduce TrustTrade (Trust-Rectified Unified Selective Trader), a multi-agent selective consensus framework inspired by human epistemic heuristics. TrustTrade replaces uniform trust with cross-agent consistency by aggregating information from multiple independent LLM agents and dynamically weighting signals based on their semantic and numerical agreement. Consistent signals are prioritized, while divergent, weakly grounded, or temporally inconsistent inputs are selectively discounted. To further stabilize decision-making, TrustTrade incorporates deterministic temporal signals as reproducible anchors and a reflective memory mechanism that adapts risk preferences at test time without additional training. Together, these components suppress noise amplification and hallucination-driven volatility, yielding more stable and risk-aware trading behavior. Across controlled backtesting in high-noise market environments (2024 Q1 and 2026 Q1), the proposed TrustTrade calibrates LLM trading behavior from extreme risk–return regimes toward a human-aligned, mid-risk and mid-return profile. Physical sciences/Mathematics and computing Biological sciences/Neuroscience Biological sciences/Psychology Social science/Psychology large language models autonomous trading agents multi-agent consensus factual hallucination decision uncertainty Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 14 May, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviewers agreed at journal 22 Apr, 2026 Reviewers invited by journal 21 Apr, 2026 Editor assigned by journal 21 Apr, 2026 Submission checks completed at journal 16 Apr, 2026 First submitted to journal 06 Apr, 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-9334812","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":629767947,"identity":"4b61a250-8794-4b0a-bf4b-708dac99fee6","order_by":0,"name":"Minghan LI","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Minghan","middleName":"","lastName":"LI","suffix":""},{"id":629767949,"identity":"8586a9c4-76fe-4763-ab2a-36b6cb139833","order_by":1,"name":"Rachel Gonsalves","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Rachel","middleName":"","lastName":"Gonsalves","suffix":""},{"id":629767950,"identity":"eeea3ee7-e2f3-4cda-add4-916f2e3ea857","order_by":2,"name":"Weiyue Li","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Weiyue","middleName":"","lastName":"Li","suffix":""},{"id":629767951,"identity":"ad6a8cdc-27fe-496f-8967-47495892c49a","order_by":3,"name":"Sunghoon Yoon","email":"","orcid":"","institution":"Daegu Gyeongbuk Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Sunghoon","middleName":"","lastName":"Yoon","suffix":""},{"id":629767953,"identity":"5b892829-f409-48d0-9657-5e035171be72","order_by":4,"name":"Mengyu Wang","email":"data:image/png;base64,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","orcid":"","institution":"Harvard University","correspondingAuthor":true,"prefix":"","firstName":"Mengyu","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2026-04-06 14:23:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9334812/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9334812/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108182023,"identity":"0903395e-9f2a-4eb4-bb31-06bf0baa44bb","added_by":"auto","created_at":"2026-04-30 08:59:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":10456105,"visible":true,"origin":"","legend":"","description":"","filename":"TrustTradeMinghanLi.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9334812/v1_covered_7ada5b81-b2e8-44de-8903-74382ae56991.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"TrustTrade: Human-Inspired Selective Consensus Reduces Decision Uncertainty in LLM Trading Agents","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":"npj-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [npj Artificial Intelligence](https://www.nature.com/npjai)","snPcode":"443878","submissionUrl":"https://submission.springernature.com/new-submission/443878/3","title":"npj Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"large language models, autonomous trading agents, multi-agent consensus, factual hallucination, decision uncertainty","lastPublishedDoi":"10.21203/rs.3.rs-9334812/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9334812/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Large language models (LLMs) are increasingly deployed as autonomous agents in financial trading. However, they often exhibit a hazardous behavioral bias that we term uniform trust, whereby retrieved information is implicitly assumed to be factual and heterogeneous sources are treated as equally informative. This assumption stands in sharp contrast to human decision-making, which relies on selective filtering, cross-validation, and experience-driven weighting of information sources. As a result, LLM-based trading systems are particularly vulnerable to multi-source noise and misinformation, amplifying factual hallucinations and leading to unstable risk–return performance. To bridge this behavioral gap, we introduce TrustTrade (Trust-Rectified Unified Selective Trader), a multi-agent selective consensus framework inspired by human epistemic heuristics. TrustTrade replaces uniform trust with cross-agent consistency by aggregating information from multiple independent LLM agents and dynamically weighting signals based on their semantic and numerical agreement. Consistent signals are prioritized, while divergent, weakly grounded, or temporally inconsistent inputs are selectively discounted. To further stabilize decision-making, TrustTrade incorporates deterministic temporal signals as reproducible anchors and a reflective memory mechanism that adapts risk preferences at test time without additional training. Together, these components suppress noise amplification and hallucination-driven volatility, yielding more stable and risk-aware trading behavior. Across controlled backtesting in high-noise market environments (2024 Q1 and 2026 Q1), the proposed TrustTrade calibrates LLM trading behavior from extreme risk–return regimes toward a human-aligned, mid-risk and mid-return profile.","manuscriptTitle":"TrustTrade: Human-Inspired Selective Consensus Reduces Decision Uncertainty in LLM Trading Agents","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-29 05:16:54","doi":"10.21203/rs.3.rs-9334812/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-14T09:40:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"316866873608614146959309804995698023052","date":"2026-04-24T01:09:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"233834506926391826593174322211431870549","date":"2026-04-23T15:20:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"261319949421982058469350290902448997292","date":"2026-04-23T13:56:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"174504069279909086565877241908936065013","date":"2026-04-22T05:59:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-21T13:20:36+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-21T12:36:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-17T03:12:08+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Artificial Intelligence","date":"2026-04-06T14:12:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-artificial-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [npj Artificial Intelligence](https://www.nature.com/npjai)","snPcode":"443878","submissionUrl":"https://submission.springernature.com/new-submission/443878/3","title":"npj Artificial Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7e3bbd8e-c233-4053-bd28-26f2b2b7f355","owner":[],"postedDate":"April 29th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-14T09:40:29+00:00","index":57,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67029315,"name":"Physical sciences/Mathematics and computing"},{"id":67029316,"name":"Biological sciences/Neuroscience"},{"id":67029317,"name":"Biological sciences/Psychology"},{"id":67029318,"name":"Social science/Psychology"}],"tags":[],"updatedAt":"2026-04-29T05:16:54+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-29 05:16:54","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9334812","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9334812","identity":"rs-9334812","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}