{"id":"1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c","title":"arXiv cs.AI: Hypnos learns sleep physiology representations via next-token prediction","argument":"Authors Jonathan F. Carter and Lionel Tarassenko present Hypnos, a model trained to predict the next token across eight sensing modalities in more than 20,000 overnight polysomnography recordings. They argue that next-token prediction offers a scalable alternative to masked-reconstruction and contrastive training for physiological signals. The model generates embeddings from continuous streams using any subset of its supported modalities. On evaluated benchmarks, the authors report that Hypnos outperforms existing foundation models on most tasks, matches strong supervised baselines in sleep stage classification while using 100× less labelled data, and surpasses a dedicated ECG foundation model at detecting atrial fibrillation. These results make the work relevant to researchers exploring whether one model can represent signals across different physiological tasks. The reported benchmark results do not by themselves establish performance in clinical use.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The abstract reports results on evaluated benchmarks and held-out test sets but does not detail their composition or establish performance in clinical use. The supplied evidence is the paper's abstract; it does not provide the methods or results needed to assess the reported comparisons in detail.","speaker":{"name":"Carter, Jonathan F., Tarassenko, Lionel","roleAtSource":"Article author"},"source":{"title":"Hypnos: Learning Generalisable Representations of Sleep Physiology via Next-Token Prediction","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2606.09605","publishedAt":"2026-06-08","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors trained Hypnos on eight sensing modalities from more than 20,000 overnight polysomnography recordings, using next-token prediction across modalities.","excerpt":"We develop Hypnos, a multi-modal sleep foundation model trained using eight different sensing modalities (e.g. EEG, ECG, respiratory signals) drawn from over 20{,}000 overnight polysomnography recordings.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"},{"text":"The authors report that Hypnos outperforms existing foundation models on most evaluated benchmarks.","excerpt":"Hypnos significantly outperforms existing foundation models on most benchmarks evaluated.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"},{"text":"In sleep stage classification, the authors report that Hypnos matches strong supervised baselines on held-out test sets while using 100 times less labelled data.","excerpt":"In sleep stage classification, we match the performance of strong supervised baselines on held-out test sets whilst using 100\\times less labelled data.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"},{"text":"The authors report that Hypnos surpasses a dedicated ECG foundation model at detecting atrial fibrillation.","excerpt":"Hypnos even generalises beyond sleep recordings, surpassing a dedicated ECG foundation model at detecting atrial fibrillation.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2606.09605","checkedAt":"2026-11-08T20:50:06.902Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How would Hypnos perform on independent data and in clinical settings beyond the evaluated benchmarks?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv cs.AI: Hypnos learns sleep physiology representations via next-token prediction","summary":"Authors Jonathan F. Carter and Lionel Tarassenko present Hypnos, a model trained to predict the next token across eight sensing modalities in more than 20,000 overnight polysomnography recordings. They argue that next-token prediction offers a scalable alternative to masked-reconstruction and contrastive training for physiological signals. The model generates embeddings from continuous streams using any subset of its supported modalities. On evaluated benchmarks, the authors report that Hypnos outperforms existing foundation models on most tasks, matches strong supervised baselines in sleep stage classification while using 100× less labelled data, and surpasses a dedicated ECG foundation model at detecting atrial fibrillation. These results make the work relevant to researchers exploring whether one model can represent signals across different physiological tasks. The reported benchmark results do not by themselves establish performance in clinical use.","question":"How would Hypnos perform on independent data and in clinical settings beyond the evaluated benchmarks?","claims":[{"text":"The authors trained Hypnos on eight sensing modalities from more than 20,000 overnight polysomnography recordings, using next-token prediction across modalities.","excerpt":"We develop Hypnos, a multi-modal sleep foundation model trained using eight different sensing modalities (e.g. EEG, ECG, respiratory signals) drawn from over 20{,}000 overnight polysomnography recordings.","support":"source-stated"},{"text":"The authors report that Hypnos outperforms existing foundation models on most evaluated benchmarks.","excerpt":"Hypnos significantly outperforms existing foundation models on most benchmarks evaluated.","support":"source-stated"},{"text":"In sleep stage classification, the authors report that Hypnos matches strong supervised baselines on held-out test sets while using 100 times less labelled data.","excerpt":"In sleep stage classification, we match the performance of strong supervised baselines on held-out test sets whilst using 100\\times less labelled data.","support":"source-stated"},{"text":"The authors report that Hypnos surpasses a dedicated ECG foundation model at detecting atrial fibrillation.","excerpt":"Hypnos even generalises beyond sleep recordings, surpassing a dedicated ECG foundation model at detecting atrial fibrillation.","support":"source-stated"}],"limitations":["The abstract reports results on evaluated benchmarks and held-out test sets but does not detail their composition or establish performance in clinical use.","The supplied evidence is the paper's abstract; it does not provide the methods or results needed to assess the reported comparisons in detail."],"mediaId":"da22c5a94ce86fcc3413e56e617e8baaf033a75e53658ba837f8ec4894b6b0a4","correctionReason":null,"id":"1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c","publicationId":"c0ba98e5-0467-4ceb-8c1e-f4ce6d5aec87","version":1,"sourceUrl":"https://arxiv.org/abs/2606.09605","sourceTitle":"Hypnos: Learning Generalisable Representations of Sleep Physiology via Next-Token Prediction","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-06-08","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-06T05:10:42.964Z","createdAt":"2026-11-08T19:46:31.659Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"da22c5a94ce86fcc3413e56e617e8baaf033a75e53658ba837f8ec4894b6b0a4","kind":"article","url":"https://arxiv.org/abs/2606.09605","sourceUrl":"https://arxiv.org/abs/2606.09605","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-08T19:39:20.249Z","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/2606.09605","checkedAt":"2026-11-08T20:50:06.902Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"da22c5a94ce86fcc3413e56e617e8baaf033a75e53658ba837f8ec4894b6b0a4","confidence":0.89,"provider":"typesafe","model":"jev-1.13.0","inputHash":"46586b3af805b14b5a70af50207eaa826d7254ca7d136472dbd95d11d142d38f","decidedAt":"2026-11-08T19:39:21.273Z"},"provenance":{"textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observationId":"b50cde0f19cf9255372843e7ffc8378df108785a898d940e47ef9521ae7c6149","retrievedAt":"2026-11-08T19:39:20.141Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"7ae0643b-86a6-4df5-8c06-0ff28170fbc8"},"reviewer":{"model":"gpt-6-luna","runId":"d2137151-4a4d-4ca8-848f-8b60a9491b86"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-08T19:46:31.659Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Carter, Jonathan F.","url":null,"kind":"unknown"},{"name":"Tarassenko, Lionel","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Carter, Jonathan F.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Carter, Jonathan F.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2606.09605","locator":"head/meta[name=author|citation_author]"},"assertionId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T20:50:06.902Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a","sourceUrl":"https://arxiv.org/abs/2606.09605","contentSha256":"abe5bec00ec02300209fa081e2dc7c1e393283a4e6a4fc72c1fc7102af8acec2","textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observedAt":"2026-11-08T20:50:06.902Z"}},{"actor":{"name":"Tarassenko, Lionel","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Tarassenko, Lionel","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2606.09605","locator":"head/meta[name=author|citation_author]"},"assertionId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T20:50:06.902Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a","sourceUrl":"https://arxiv.org/abs/2606.09605","contentSha256":"abe5bec00ec02300209fa081e2dc7c1e393283a4e6a4fc72c1fc7102af8acec2","textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observedAt":"2026-11-08T20:50:06.902Z"}}],"participationCoverage":{"total":2,"returned":2,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2606.09605","snapshotId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a","contentSha256":"abe5bec00ec02300209fa081e2dc7c1e393283a4e6a4fc72c1fc7102af8acec2","textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observedAt":"2026-11-08T20:50:06.902Z","basis":"publisher-page-metadata","sourceLinks":43,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c/source"}},"canonicalUrl":"https://escapemeridian.com/readings/arxiv-cs-ai-hypnos-learns-sleep-physiology-representations-via-1df556cb","formats":{"html":"/readings/arxiv-cs-ai-hypnos-learns-sleep-physiology-representations-via-1df556cb","markdown":"/readings/arxiv-cs-ai-hypnos-learns-sleep-physiology-representations-via-1df556cb.md","json":"/readings/arxiv-cs-ai-hypnos-learns-sleep-physiology-representations-via-1df556cb.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":[{"id":"term-embedding-vector","name":"Embedding vector","explanation":"A list of numbers that represents a piece of text, an image or another item so software can compare meanings.","referenceUrl":"https://developers.google.com/machine-learning/glossary#embedding-vector"},{"id":"term-benchmark","name":"Benchmark","explanation":"A defined set of tests used to compare systems. Results depend on the tasks and measures selected.","referenceUrl":"https://arxiv.org/abs/2211.09110"},{"id":"term-foundation-model","name":"Foundation model","explanation":"A large model trained on broad data that can be adapted to many different tasks.","referenceUrl":"https://developers.google.com/machine-learning/glossary#foundation-model"},{"id":"term-held-out-set","name":"Held-out set","explanation":"Data kept aside during training and used only to test how well a model works on examples it has not seen.","referenceUrl":"https://developers.google.com/machine-learning/glossary#holdout-data"}],"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":"1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c","revision":"c0ba98e5-0467-4ceb-8c1e-f4ce6d5aec87","revisionKind":"publication-id","title":"arXiv cs.AI: Hypnos learns sleep physiology representations via next-token prediction","url":"https://escapemeridian.com/readings/arxiv-cs-ai-hypnos-learns-sleep-physiology-representations-via-1df556cb"},"source":{"type":"source","id":"https://arxiv.org/abs/2606.09605","revision":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","revisionKind":"sha256-source-text","title":"Hypnos: Learning Generalisable Representations of Sleep Physiology via Next-Token Prediction","url":"https://arxiv.org/abs/2606.09605"},"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":"1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c","title":"arXiv cs.AI: Hypnos learns sleep physiology representations via next-token prediction","argument":"Authors Jonathan F. Carter and Lionel Tarassenko present Hypnos, a model trained to predict the next token across eight sensing modalities in more than 20,000 overnight polysomnography recordings. They argue that next-token prediction offers a scalable alternative to masked-reconstruction and contrastive training for physiological signals. The model generates embeddings from continuous streams using any subset of its supported modalities. On evaluated benchmarks, the authors report that Hypnos outperforms existing foundation models on most tasks, matches strong supervised baselines in sleep stage classification while using 100× less labelled data, and surpasses a dedicated ECG foundation model at detecting atrial fibrillation. These results make the work relevant to researchers exploring whether one model can represent signals across different physiological tasks. The reported benchmark results do not by themselves establish performance in clinical use.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The abstract reports results on evaluated benchmarks and held-out test sets but does not detail their composition or establish performance in clinical use. The supplied evidence is the paper's abstract; it does not provide the methods or results needed to assess the reported comparisons in detail.","speaker":{"name":"Carter, Jonathan F., Tarassenko, Lionel","roleAtSource":"Article author"},"source":{"title":"Hypnos: Learning Generalisable Representations of Sleep Physiology via Next-Token Prediction","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2606.09605","publishedAt":"2026-06-08","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors trained Hypnos on eight sensing modalities from more than 20,000 overnight polysomnography recordings, using next-token prediction across modalities.","excerpt":"We develop Hypnos, a multi-modal sleep foundation model trained using eight different sensing modalities (e.g. EEG, ECG, respiratory signals) drawn from over 20{,}000 overnight polysomnography recordings.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"},{"text":"The authors report that Hypnos outperforms existing foundation models on most evaluated benchmarks.","excerpt":"Hypnos significantly outperforms existing foundation models on most benchmarks evaluated.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"},{"text":"In sleep stage classification, the authors report that Hypnos matches strong supervised baselines on held-out test sets while using 100 times less labelled data.","excerpt":"In sleep stage classification, we match the performance of strong supervised baselines on held-out test sets whilst using 100\\times less labelled data.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"},{"text":"The authors report that Hypnos surpasses a dedicated ECG foundation model at detecting atrial fibrillation.","excerpt":"Hypnos even generalises beyond sleep recordings, surpassing a dedicated ECG foundation model at detecting atrial fibrillation.","url":"https://arxiv.org/abs/2606.09605","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2606.09605","checkedAt":"2026-11-08T20:50:06.902Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How would Hypnos perform on independent data and in clinical settings beyond the evaluated benchmarks?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv cs.AI: Hypnos learns sleep physiology representations via next-token prediction","summary":"Authors Jonathan F. Carter and Lionel Tarassenko present Hypnos, a model trained to predict the next token across eight sensing modalities in more than 20,000 overnight polysomnography recordings. They argue that next-token prediction offers a scalable alternative to masked-reconstruction and contrastive training for physiological signals. The model generates embeddings from continuous streams using any subset of its supported modalities. On evaluated benchmarks, the authors report that Hypnos outperforms existing foundation models on most tasks, matches strong supervised baselines in sleep stage classification while using 100× less labelled data, and surpasses a dedicated ECG foundation model at detecting atrial fibrillation. These results make the work relevant to researchers exploring whether one model can represent signals across different physiological tasks. The reported benchmark results do not by themselves establish performance in clinical use.","question":"How would Hypnos perform on independent data and in clinical settings beyond the evaluated benchmarks?","claims":[{"text":"The authors trained Hypnos on eight sensing modalities from more than 20,000 overnight polysomnography recordings, using next-token prediction across modalities.","excerpt":"We develop Hypnos, a multi-modal sleep foundation model trained using eight different sensing modalities (e.g. EEG, ECG, respiratory signals) drawn from over 20{,}000 overnight polysomnography recordings.","support":"source-stated"},{"text":"The authors report that Hypnos outperforms existing foundation models on most evaluated benchmarks.","excerpt":"Hypnos significantly outperforms existing foundation models on most benchmarks evaluated.","support":"source-stated"},{"text":"In sleep stage classification, the authors report that Hypnos matches strong supervised baselines on held-out test sets while using 100 times less labelled data.","excerpt":"In sleep stage classification, we match the performance of strong supervised baselines on held-out test sets whilst using 100\\times less labelled data.","support":"source-stated"},{"text":"The authors report that Hypnos surpasses a dedicated ECG foundation model at detecting atrial fibrillation.","excerpt":"Hypnos even generalises beyond sleep recordings, surpassing a dedicated ECG foundation model at detecting atrial fibrillation.","support":"source-stated"}],"limitations":["The abstract reports results on evaluated benchmarks and held-out test sets but does not detail their composition or establish performance in clinical use.","The supplied evidence is the paper's abstract; it does not provide the methods or results needed to assess the reported comparisons in detail."],"mediaId":"da22c5a94ce86fcc3413e56e617e8baaf033a75e53658ba837f8ec4894b6b0a4","correctionReason":null,"id":"1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c","publicationId":"c0ba98e5-0467-4ceb-8c1e-f4ce6d5aec87","version":1,"sourceUrl":"https://arxiv.org/abs/2606.09605","sourceTitle":"Hypnos: Learning Generalisable Representations of Sleep Physiology via Next-Token Prediction","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-06-08","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-06T05:10:42.964Z","createdAt":"2026-11-08T19:46:31.659Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"da22c5a94ce86fcc3413e56e617e8baaf033a75e53658ba837f8ec4894b6b0a4","kind":"article","url":"https://arxiv.org/abs/2606.09605","sourceUrl":"https://arxiv.org/abs/2606.09605","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-08T19:39:20.249Z","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/2606.09605","checkedAt":"2026-11-08T20:50:06.902Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"da22c5a94ce86fcc3413e56e617e8baaf033a75e53658ba837f8ec4894b6b0a4","confidence":0.89,"provider":"typesafe","model":"jev-1.13.0","inputHash":"46586b3af805b14b5a70af50207eaa826d7254ca7d136472dbd95d11d142d38f","decidedAt":"2026-11-08T19:39:21.273Z"},"provenance":{"textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observationId":"b50cde0f19cf9255372843e7ffc8378df108785a898d940e47ef9521ae7c6149","retrievedAt":"2026-11-08T19:39:20.141Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"7ae0643b-86a6-4df5-8c06-0ff28170fbc8"},"reviewer":{"model":"gpt-6-luna","runId":"d2137151-4a4d-4ca8-848f-8b60a9491b86"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-08T19:46:31.659Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Carter, Jonathan F.","url":null,"kind":"unknown"},{"name":"Tarassenko, Lionel","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Carter, Jonathan F.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Carter, Jonathan F.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2606.09605","locator":"head/meta[name=author|citation_author]"},"assertionId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T20:50:06.902Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a","sourceUrl":"https://arxiv.org/abs/2606.09605","contentSha256":"abe5bec00ec02300209fa081e2dc7c1e393283a4e6a4fc72c1fc7102af8acec2","textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observedAt":"2026-11-08T20:50:06.902Z"}},{"actor":{"name":"Tarassenko, Lionel","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Tarassenko, Lionel","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2606.09605","locator":"head/meta[name=author|citation_author]"},"assertionId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-08T20:50:06.902Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a","sourceUrl":"https://arxiv.org/abs/2606.09605","contentSha256":"abe5bec00ec02300209fa081e2dc7c1e393283a4e6a4fc72c1fc7102af8acec2","textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observedAt":"2026-11-08T20:50:06.902Z"}}],"participationCoverage":{"total":2,"returned":2,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2606.09605","snapshotId":"285c86512906e24424f7c4f5ecbebb4d5017d949b35f757454e0ecffc1b6180a","contentSha256":"abe5bec00ec02300209fa081e2dc7c1e393283a4e6a4fc72c1fc7102af8acec2","textHash":"65ef3c8576a01c70bb5a92c9f30272d074b401bb2f5000b030fcdfa9748daa32","observedAt":"2026-11-08T20:50:06.902Z","basis":"publisher-page-metadata","sourceLinks":43,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/1df556cb6032c45897a4e7af993c4bc2e305c1c2c59dda9e9e6b76b83597579c/source"}}}],"audits":[{"publicationId":"c0ba98e5-0467-4ceb-8c1e-f4ce6d5aec87","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading attributes the work and claims to authors Jonathan F. Carter and Lionel Tarassenko and keeps results framed as author reports. This matches the arXiv page.","excerpt":"Authors:Jonathan F. Carter, Lionel Tarassenko"},{"dimension":"summary","status":"uncertain","explanation":"Facts are faithful, but a general reader may not follow the central terms. Polysomnography (an overnight clinical sleep study), next-token prediction, tokens and embeddings are used without plain-language explanation; tooltip coverage should be confirmed. The reading also drops the authors' stated reason for the method: masked reconstruction may suit stochastic signals poorly, and contrastive methods depend on poorly understood invariances. Smallest repair: add one clause on that rationale, gloss polysomnography, and keep 'significantly' and 'even generalises beyond sleep recordings' to preserve the source's force.","excerpt":"masked reconstruction may be poorly suited to the stochastic nature of these signals, while contrastive approaches rely on positive-pair definitions despite the semantic invariances of physiological signals being poorly understood"},{"dimension":"date","status":"pass","explanation":"originalPublished 2026-06-08 matches the v1 submission date. It does not use the 4 Nov 2026 revision or the November feed date.","excerpt":"Submitted on 8 Jun 2026 (v1)"},{"dimension":"support","status":"pass","explanation":"Each claim matches the abstract and is framed as an author report. The relevance statement is modest and tied to the cross-modality and atrial fibrillation results. The limitations correctly note that benchmark results do not establish clinical performance. The question follows from the evidence.","excerpt":"Hypnos significantly outperforms existing foundation models on most benchmarks evaluated."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper, which is the most relevant way to reach the source.","excerpt":null}],"createdAt":"2026-11-08T19:46:31.659Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}