{"id":"a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d","title":"arXiv cs.AI: Satisficing as competitive Bayesian weighting","argument":"In their arXiv abstract, Moto Shinriki, Yu Kono and Tatsuji Takahashi present Attention Competitive Bayes (ACB), a model of choice organized around an aspiration level—a goal that may remain unmet. ACB rewrites Risk-sensitive Satisficing (RS), a bandit learning rule, as a Bayesian-like estimator: the aspiration level acts as the prior mean for each action, while a limited trial budget determines how strongly evidence can override it. The authors say that when an action is no longer sampled, its estimate moves back toward the aspiration level without discarding past observations. They call this “pseudo-forgetting” and distinguish it from models with an explicit forgetting rate. The paper also argues that an unattainable aspiration keeps exploration from settling. In Bernoulli bandit simulations, adjusting the aspiration level based on evidence reduced regret, or expected loss. The authors offer ACB as a formal model of how an unattained goal and limited attention jointly constrain behavior, and say it yields testable predictions about how organisms revalue abandoned options. The abstract describes a minimal model and simulations; it does not report tests of those predictions in organisms.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The evidence provided is the paper's abstract, which characterizes ACB as a minimal formal model and reports simulation results; it does not describe tests of the predictions in organisms.","speaker":{"name":"Shinriki, Moto, Kono, Yu, Takahashi, Tatsuji","roleAtSource":"Article author"},"source":{"title":"Satisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained target","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.07270","publishedAt":"2026-11-05","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors recast Risk-sensitive Satisficing as Attention Competitive Bayes, where an aspiration level is the prior mean for each action and a finite trial budget controls how much evidence can override it.","excerpt":"the aspiration level is the prior mean of every action and the finite budget of trials sets how strongly evidence can override it","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"},{"text":"In ACB, estimates for actions no longer sampled relax toward the aspiration level without past observations being discarded; the authors call this pseudo-forgetting.","excerpt":"their estimates relax toward the aspiration level without any data being discarded. This ``pseudo-forgetting'' distinguishes ACB from models with an explicit forgetting rate.","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"},{"text":"The authors report that exploration does not settle when the aspiration is unattainable, and that adjusting aspiration based on evidence reduced regret in Bernoulli bandit simulations.","excerpt":"when the aspiration is unattainable, exploration does not settle; in Bernoulli bandit simulations, adjusting the aspiration level from evidence reduced regret (expected loss).","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"},{"text":"The authors say ACB yields testable predictions about how organisms revalue options they have abandoned.","excerpt":"it yields testable predictions about how organisms revalue options they have abandoned.","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07270","checkedAt":"2026-11-09T03:26:06.577Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How would the predicted revaluation of abandoned options be tested in organisms, as distinct from the Bernoulli bandit simulations described in the abstract?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv cs.AI: Satisficing as competitive Bayesian weighting","summary":"In their arXiv abstract, Moto Shinriki, Yu Kono and Tatsuji Takahashi present Attention Competitive Bayes (ACB), a model of choice organized around an aspiration level—a goal that may remain unmet. ACB rewrites Risk-sensitive Satisficing (RS), a bandit learning rule, as a Bayesian-like estimator: the aspiration level acts as the prior mean for each action, while a limited trial budget determines how strongly evidence can override it. The authors say that when an action is no longer sampled, its estimate moves back toward the aspiration level without discarding past observations. They call this “pseudo-forgetting” and distinguish it from models with an explicit forgetting rate. The paper also argues that an unattainable aspiration keeps exploration from settling. In Bernoulli bandit simulations, adjusting the aspiration level based on evidence reduced regret, or expected loss. The authors offer ACB as a formal model of how an unattained goal and limited attention jointly constrain behavior, and say it yields testable predictions about how organisms revalue abandoned options. The abstract describes a minimal model and simulations; it does not report tests of those predictions in organisms.","question":"How would the predicted revaluation of abandoned options be tested in organisms, as distinct from the Bernoulli bandit simulations described in the abstract?","claims":[{"text":"The authors recast Risk-sensitive Satisficing as Attention Competitive Bayes, where an aspiration level is the prior mean for each action and a finite trial budget controls how much evidence can override it.","excerpt":"the aspiration level is the prior mean of every action and the finite budget of trials sets how strongly evidence can override it","support":"source-stated"},{"text":"In ACB, estimates for actions no longer sampled relax toward the aspiration level without past observations being discarded; the authors call this pseudo-forgetting.","excerpt":"their estimates relax toward the aspiration level without any data being discarded. This ``pseudo-forgetting'' distinguishes ACB from models with an explicit forgetting rate.","support":"source-stated"},{"text":"The authors report that exploration does not settle when the aspiration is unattainable, and that adjusting aspiration based on evidence reduced regret in Bernoulli bandit simulations.","excerpt":"when the aspiration is unattainable, exploration does not settle; in Bernoulli bandit simulations, adjusting the aspiration level from evidence reduced regret (expected loss).","support":"source-stated"},{"text":"The authors say ACB yields testable predictions about how organisms revalue options they have abandoned.","excerpt":"it yields testable predictions about how organisms revalue options they have abandoned.","support":"source-stated"}],"limitations":["The evidence provided is the paper's abstract, which characterizes ACB as a minimal formal model and reports simulation results; it does not describe tests of the predictions in organisms."],"mediaId":"d2c90d807578f6ee9661ede48fd90a5c7a82160e99641b0b52a80155e998e2af","correctionReason":null,"id":"a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d","publicationId":"8318c526-6ab7-497d-856d-efbbd273a8c8","version":1,"sourceUrl":"https://arxiv.org/abs/2611.07270","sourceTitle":"Satisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained target","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-05","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-06T05:03:27.410Z","createdAt":"2026-11-09T02:22:42.176Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"d2c90d807578f6ee9661ede48fd90a5c7a82160e99641b0b52a80155e998e2af","kind":"article","url":"https://arxiv.org/abs/2611.07270","sourceUrl":"https://arxiv.org/abs/2611.07270","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-09T02:19:10.193Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07270","checkedAt":"2026-11-09T03:26:06.577Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"d2c90d807578f6ee9661ede48fd90a5c7a82160e99641b0b52a80155e998e2af","confidence":0.51,"provider":"typesafe","model":"jev-1.13.0","inputHash":"a62abd9c8698d6afe4a5062d46dadc66bbdda502bbc5f63ee45e6a38b87b98c5","decidedAt":"2026-11-09T02:19:12.594Z"},"provenance":{"textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observationId":"309f7f9cf0de71f88393a41b9d6036730e69c77598e742a1c1c447007fd8d7cd","retrievedAt":"2026-11-09T02:19:09.781Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"cffd0727-aedc-4f5c-88b5-82f4394f3922"},"reviewer":{"model":"gpt-6-luna","runId":"3d9f9df2-02e6-4ec9-80a7-521e0ba93f6e"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-09T02:22:42.176Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Shinriki, Moto","url":null,"kind":"unknown"},{"name":"Kono, Yu","url":null,"kind":"unknown"},{"name":"Takahashi, Tatsuji","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Shinriki, Moto","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shinriki, Moto","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07270","locator":"head/meta[name=author|citation_author]"},"assertionId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T03:26:06.577Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","sourceUrl":"https://arxiv.org/abs/2611.07270","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z"}},{"actor":{"name":"Kono, Yu","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Kono, Yu","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07270","locator":"head/meta[name=author|citation_author]"},"assertionId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T03:26:06.577Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","sourceUrl":"https://arxiv.org/abs/2611.07270","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z"}},{"actor":{"name":"Takahashi, Tatsuji","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Takahashi, Tatsuji","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07270","locator":"head/meta[name=author|citation_author]"},"assertionId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T03:26:06.577Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","sourceUrl":"https://arxiv.org/abs/2611.07270","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z"}}],"participationCoverage":{"total":3,"returned":3,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.07270","snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z","basis":"publisher-page-metadata","sourceLinks":45,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d/source"}},"canonicalUrl":"https://escapemeridian.com/readings/arxiv-cs-ai-satisficing-as-competitive-bayesian-weighting-a2a1454e","formats":{"html":"/readings/arxiv-cs-ai-satisficing-as-competitive-bayesian-weighting-a2a1454e","markdown":"/readings/arxiv-cs-ai-satisficing-as-competitive-bayesian-weighting-a2a1454e.md","json":"/readings/arxiv-cs-ai-satisficing-as-competitive-bayesian-weighting-a2a1454e.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d","revision":"8318c526-6ab7-497d-856d-efbbd273a8c8","revisionKind":"publication-id","title":"arXiv cs.AI: Satisficing as competitive Bayesian weighting","url":"https://escapemeridian.com/readings/arxiv-cs-ai-satisficing-as-competitive-bayesian-weighting-a2a1454e"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.07270","revision":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","revisionKind":"sha256-source-text","title":"Satisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained target","url":"https://arxiv.org/abs/2611.07270"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d","title":"arXiv cs.AI: Satisficing as competitive Bayesian weighting","argument":"In their arXiv abstract, Moto Shinriki, Yu Kono and Tatsuji Takahashi present Attention Competitive Bayes (ACB), a model of choice organized around an aspiration level—a goal that may remain unmet. ACB rewrites Risk-sensitive Satisficing (RS), a bandit learning rule, as a Bayesian-like estimator: the aspiration level acts as the prior mean for each action, while a limited trial budget determines how strongly evidence can override it. The authors say that when an action is no longer sampled, its estimate moves back toward the aspiration level without discarding past observations. They call this “pseudo-forgetting” and distinguish it from models with an explicit forgetting rate. The paper also argues that an unattainable aspiration keeps exploration from settling. In Bernoulli bandit simulations, adjusting the aspiration level based on evidence reduced regret, or expected loss. The authors offer ACB as a formal model of how an unattained goal and limited attention jointly constrain behavior, and say it yields testable predictions about how organisms revalue abandoned options. The abstract describes a minimal model and simulations; it does not report tests of those predictions in organisms.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The evidence provided is the paper's abstract, which characterizes ACB as a minimal formal model and reports simulation results; it does not describe tests of the predictions in organisms.","speaker":{"name":"Shinriki, Moto, Kono, Yu, Takahashi, Tatsuji","roleAtSource":"Article author"},"source":{"title":"Satisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained target","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.07270","publishedAt":"2026-11-05","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors recast Risk-sensitive Satisficing as Attention Competitive Bayes, where an aspiration level is the prior mean for each action and a finite trial budget controls how much evidence can override it.","excerpt":"the aspiration level is the prior mean of every action and the finite budget of trials sets how strongly evidence can override it","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"},{"text":"In ACB, estimates for actions no longer sampled relax toward the aspiration level without past observations being discarded; the authors call this pseudo-forgetting.","excerpt":"their estimates relax toward the aspiration level without any data being discarded. This ``pseudo-forgetting'' distinguishes ACB from models with an explicit forgetting rate.","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"},{"text":"The authors report that exploration does not settle when the aspiration is unattainable, and that adjusting aspiration based on evidence reduced regret in Bernoulli bandit simulations.","excerpt":"when the aspiration is unattainable, exploration does not settle; in Bernoulli bandit simulations, adjusting the aspiration level from evidence reduced regret (expected loss).","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"},{"text":"The authors say ACB yields testable predictions about how organisms revalue options they have abandoned.","excerpt":"it yields testable predictions about how organisms revalue options they have abandoned.","url":"https://arxiv.org/abs/2611.07270","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07270","checkedAt":"2026-11-09T03:26:06.577Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How would the predicted revaluation of abandoned options be tested in organisms, as distinct from the Bernoulli bandit simulations described in the abstract?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv cs.AI: Satisficing as competitive Bayesian weighting","summary":"In their arXiv abstract, Moto Shinriki, Yu Kono and Tatsuji Takahashi present Attention Competitive Bayes (ACB), a model of choice organized around an aspiration level—a goal that may remain unmet. ACB rewrites Risk-sensitive Satisficing (RS), a bandit learning rule, as a Bayesian-like estimator: the aspiration level acts as the prior mean for each action, while a limited trial budget determines how strongly evidence can override it. The authors say that when an action is no longer sampled, its estimate moves back toward the aspiration level without discarding past observations. They call this “pseudo-forgetting” and distinguish it from models with an explicit forgetting rate. The paper also argues that an unattainable aspiration keeps exploration from settling. In Bernoulli bandit simulations, adjusting the aspiration level based on evidence reduced regret, or expected loss. The authors offer ACB as a formal model of how an unattained goal and limited attention jointly constrain behavior, and say it yields testable predictions about how organisms revalue abandoned options. The abstract describes a minimal model and simulations; it does not report tests of those predictions in organisms.","question":"How would the predicted revaluation of abandoned options be tested in organisms, as distinct from the Bernoulli bandit simulations described in the abstract?","claims":[{"text":"The authors recast Risk-sensitive Satisficing as Attention Competitive Bayes, where an aspiration level is the prior mean for each action and a finite trial budget controls how much evidence can override it.","excerpt":"the aspiration level is the prior mean of every action and the finite budget of trials sets how strongly evidence can override it","support":"source-stated"},{"text":"In ACB, estimates for actions no longer sampled relax toward the aspiration level without past observations being discarded; the authors call this pseudo-forgetting.","excerpt":"their estimates relax toward the aspiration level without any data being discarded. This ``pseudo-forgetting'' distinguishes ACB from models with an explicit forgetting rate.","support":"source-stated"},{"text":"The authors report that exploration does not settle when the aspiration is unattainable, and that adjusting aspiration based on evidence reduced regret in Bernoulli bandit simulations.","excerpt":"when the aspiration is unattainable, exploration does not settle; in Bernoulli bandit simulations, adjusting the aspiration level from evidence reduced regret (expected loss).","support":"source-stated"},{"text":"The authors say ACB yields testable predictions about how organisms revalue options they have abandoned.","excerpt":"it yields testable predictions about how organisms revalue options they have abandoned.","support":"source-stated"}],"limitations":["The evidence provided is the paper's abstract, which characterizes ACB as a minimal formal model and reports simulation results; it does not describe tests of the predictions in organisms."],"mediaId":"d2c90d807578f6ee9661ede48fd90a5c7a82160e99641b0b52a80155e998e2af","correctionReason":null,"id":"a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d","publicationId":"8318c526-6ab7-497d-856d-efbbd273a8c8","version":1,"sourceUrl":"https://arxiv.org/abs/2611.07270","sourceTitle":"Satisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained target","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-05","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-06T05:03:27.410Z","createdAt":"2026-11-09T02:22:42.176Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"d2c90d807578f6ee9661ede48fd90a5c7a82160e99641b0b52a80155e998e2af","kind":"article","url":"https://arxiv.org/abs/2611.07270","sourceUrl":"https://arxiv.org/abs/2611.07270","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-09T02:19:10.193Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.07270","checkedAt":"2026-11-09T03:26:06.577Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"d2c90d807578f6ee9661ede48fd90a5c7a82160e99641b0b52a80155e998e2af","confidence":0.51,"provider":"typesafe","model":"jev-1.13.0","inputHash":"a62abd9c8698d6afe4a5062d46dadc66bbdda502bbc5f63ee45e6a38b87b98c5","decidedAt":"2026-11-09T02:19:12.594Z"},"provenance":{"textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observationId":"309f7f9cf0de71f88393a41b9d6036730e69c77598e742a1c1c447007fd8d7cd","retrievedAt":"2026-11-09T02:19:09.781Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"cffd0727-aedc-4f5c-88b5-82f4394f3922"},"reviewer":{"model":"gpt-6-luna","runId":"3d9f9df2-02e6-4ec9-80a7-521e0ba93f6e"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-09T02:22:42.176Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Shinriki, Moto","url":null,"kind":"unknown"},{"name":"Kono, Yu","url":null,"kind":"unknown"},{"name":"Takahashi, Tatsuji","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Shinriki, Moto","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shinriki, Moto","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07270","locator":"head/meta[name=author|citation_author]"},"assertionId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T03:26:06.577Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","sourceUrl":"https://arxiv.org/abs/2611.07270","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z"}},{"actor":{"name":"Kono, Yu","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Kono, Yu","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07270","locator":"head/meta[name=author|citation_author]"},"assertionId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T03:26:06.577Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","sourceUrl":"https://arxiv.org/abs/2611.07270","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z"}},{"actor":{"name":"Takahashi, Tatsuji","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Takahashi, Tatsuji","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.07270","locator":"head/meta[name=author|citation_author]"},"assertionId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-09T03:26:06.577Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","sourceUrl":"https://arxiv.org/abs/2611.07270","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z"}}],"participationCoverage":{"total":3,"returned":3,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.07270","snapshotId":"70647407259c937b81a479feb8036b3d8a45ea462aa767d7cdf09b8e98992827","contentSha256":"90b33418811e9461c5b19af428b1509355c464eccf82afe22e8ffa2396a3802a","textHash":"11f48232bb4fbe87584dff89b15041bb67751742f16b139c4eb098108c71b3b4","observedAt":"2026-11-09T03:26:06.577Z","basis":"publisher-page-metadata","sourceLinks":45,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/a2a1454e3e7ad83fd56ee0b6b88302c1c04945a0e0563d3ceffbc019fabde81d/source"}}}],"audits":[{"publicationId":"8318c526-6ab7-497d-856d-efbbd273a8c8","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly credits the arXiv abstract to Moto Shinriki, Yu Kono and Tatsuji Takahashi. It keeps the key results attributed with phrases such as 'the authors say' and 'they call this' rather than presenting them as established fact.","excerpt":"Authors:Moto Shinriki, Yu Kono, Tatsuji Takahashi"},{"dimension":"summary","status":"uncertain","explanation":"The reading is faithful, but a non-specialist would struggle with its central terms. 'Bandit learning rule', 'Bayesian-like estimator', 'prior mean' and 'Bernoulli bandit' are unexplained, and they carry the core idea: the goal serves as the default expected value of every option until evidence outweighs it. The reading also drops the paper's motivating question of how a goal not yet realized can organize choice, along with the Deacon 'ententional' framing. Smallest repair: briefly gloss 'bandit' (repeated choice among options with uncertain payoffs) and 'prior mean' (starting expectation), and restore the motivating question in one clause.","excerpt":"How can a goal that is not (yet) realized organize choice?"},{"dimension":"date","status":"pass","explanation":"The displayed original publication date, 2026-11-05, matches the v1 submission date. The later feed date of 2026-11-06 was correctly not used.","excerpt":"[Submitted on 5 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Every claim matches the abstract, and each supporting excerpt is verbatim. The regret result is kept limited to the Bernoulli bandit simulations. The limitation correctly notes that the predictions are untested in organisms. The question follows from the source without implying an outcome. No unsupported benefits or costs are introduced.","excerpt":"in Bernoulli bandit simulations, adjusting the aspiration level from evidence reduced regret (expected loss)"},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper. It is the most direct and relevant way for readers to reach the source.","excerpt":null}],"createdAt":"2026-11-09T02:22:42.176Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}