{"id":"1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af","title":"arXiv: OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","argument":"The authors of OmitEHR argue that omissions—clinically important information missing from a summary—are the most common failure mode in patient-record summaries generated by large language models (LLMs). Their method learns which source-note sentences clinicians included in historical summaries, then flags high-salience sentences that an LLM-generated summary does not cover. The authors evaluate OmitEHR on brief hospital course (BHC) summarization using 5,000 admissions from one large academic medical center. On held-out admissions, it outperformed the evaluated baselines at predicting salient source sentences, including a prompted LLM judge. The authors also report that its full omission detection pipeline was faster and cheaper than scoring every source sentence with an LLM judge. In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline; candidates counted if at least one clinician confirmed them. The paper presents OmitEHR as a way for health systems to flag omissions before summaries reach clinicians. These results describe the authors’ evaluations; the evidence provided here is from one medical center and a 25-admission blinded review.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The reported evaluation uses admissions from a large academic medical center; the supplied evidence does not report results from other medical centers. The blinded review covered 25 admissions, and candidates counted as confirmed if at least one of the two clinicians confirmed them.","speaker":{"name":"Lin, Bridget, Welch, Ryan, Seddighzadeh, Bobak, Riley, Fran, Stoehr, Kaitlyn, Pipinos, Margarita I., Speron, Dimitri, Sage, Emma V., Jain, Sneha, Shah, Nigam H., Alsentzer, Emily","roleAtSource":"Article author"},"source":{"title":"OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.19343","publishedAt":"2026-11-14","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors say OmitEHR learns which source-note sentences clinicians include in historical summaries and flags high-salience sentences missing from LLM-generated summaries.","excerpt":"OmitEHR learns to predict which source-note sentences clinicians include in their summaries, then surfaces high-salience sentences not covered by an LLM-generated summary as candidate omissions.","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"},{"text":"The authors evaluated OmitEHR on BHC summarization using 5,000 admissions from a large academic medical center.","excerpt":"We apply OmitEHR to brief hospital course (BHC) summarization using 5,000 admissions from a large academic medical center.","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"},{"text":"On held-out admissions, the authors report that OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","excerpt":"On held-out admissions, OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"},{"text":"In a blinded review of 25 admissions, the authors report that clinicians confirmed more clinically meaningful omissions surfaced by OmitEHR than by a full LLM pipeline.","excerpt":"In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.19343","checkedAt":"2026-11-17T13:51:01.725Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How would OmitEHR perform on admissions from medical centers beyond the one used in this evaluation?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv: OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","summary":"The authors of OmitEHR argue that omissions—clinically important information missing from a summary—are the most common failure mode in patient-record summaries generated by large language models (LLMs). Their method learns which source-note sentences clinicians included in historical summaries, then flags high-salience sentences that an LLM-generated summary does not cover. The authors evaluate OmitEHR on brief hospital course (BHC) summarization using 5,000 admissions from one large academic medical center. On held-out admissions, it outperformed the evaluated baselines at predicting salient source sentences, including a prompted LLM judge. The authors also report that its full omission detection pipeline was faster and cheaper than scoring every source sentence with an LLM judge. In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline; candidates counted if at least one clinician confirmed them. The paper presents OmitEHR as a way for health systems to flag omissions before summaries reach clinicians. These results describe the authors’ evaluations; the evidence provided here is from one medical center and a 25-admission blinded review.","question":"How would OmitEHR perform on admissions from medical centers beyond the one used in this evaluation?","claims":[{"text":"The authors say OmitEHR learns which source-note sentences clinicians include in historical summaries and flags high-salience sentences missing from LLM-generated summaries.","excerpt":"OmitEHR learns to predict which source-note sentences clinicians include in their summaries, then surfaces high-salience sentences not covered by an LLM-generated summary as candidate omissions.","support":"source-stated"},{"text":"The authors evaluated OmitEHR on BHC summarization using 5,000 admissions from a large academic medical center.","excerpt":"We apply OmitEHR to brief hospital course (BHC) summarization using 5,000 admissions from a large academic medical center.","support":"source-stated"},{"text":"On held-out admissions, the authors report that OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","excerpt":"On held-out admissions, OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","support":"source-stated"},{"text":"In a blinded review of 25 admissions, the authors report that clinicians confirmed more clinically meaningful omissions surfaced by OmitEHR than by a full LLM pipeline.","excerpt":"In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline","support":"source-stated"}],"limitations":["The reported evaluation uses admissions from a large academic medical center; the supplied evidence does not report results from other medical centers.","The blinded review covered 25 admissions, and candidates counted as confirmed if at least one of the two clinicians confirmed them."],"mediaId":"bb9a7a50a7963d5250e97240569146bd429414e8325b2ec4f4b6355c743a0baa","correctionReason":null,"id":"1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af","publicationId":"bcfe6b92-0b87-4d35-85f6-cce230a76776","version":1,"sourceUrl":"https://arxiv.org/abs/2611.19343","sourceTitle":"OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-14","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-17T08:29:09.856Z","createdAt":"2026-11-17T13:18:47.337Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"bb9a7a50a7963d5250e97240569146bd429414e8325b2ec4f4b6355c743a0baa","kind":"article","url":"https://arxiv.org/abs/2611.19343","sourceUrl":"https://arxiv.org/abs/2611.19343","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-17T13:17:25.139Z","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/2611.19343","checkedAt":"2026-11-17T13:51:01.725Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"bb9a7a50a7963d5250e97240569146bd429414e8325b2ec4f4b6355c743a0baa","confidence":0.65,"provider":"typesafe","model":"jev-1.13.0","inputHash":"c0683316b3a5f171bbc748b76dcfc82b8bbb1e44cbb84f342c83cdd1adfbcede","decidedAt":"2026-11-17T13:17:27.380Z"},"provenance":{"textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observationId":"5fe1da8350a0f1c1276bc7adbf28e89ac4c77e24e9ee0c473b9cc0f2ffb2e329","retrievedAt":"2026-11-17T13:17:24.833Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"f9c0330d-3852-48bf-838e-d60e6da0b02f"},"reviewer":{"model":"gpt-6-luna","runId":"45e2fe52-38c6-411b-819d-579d8e5d24b6"}},"audit":{"status":"passed","reviewedAt":"2026-11-17T13:18:47.337Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Lin, Bridget","url":null,"kind":"unknown"},{"name":"Welch, Ryan","url":null,"kind":"unknown"},{"name":"Seddighzadeh, Bobak","url":null,"kind":"unknown"},{"name":"Riley, Fran","url":null,"kind":"unknown"},{"name":"Stoehr, Kaitlyn","url":null,"kind":"unknown"},{"name":"Pipinos, Margarita I.","url":null,"kind":"unknown"},{"name":"Speron, Dimitri","url":null,"kind":"unknown"},{"name":"Sage, Emma V.","url":null,"kind":"unknown"},{"name":"Jain, Sneha","url":null,"kind":"unknown"},{"name":"Shah, Nigam H.","url":null,"kind":"unknown"},{"name":"Alsentzer, Emily","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Lin, Bridget","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Lin, Bridget","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Welch, Ryan","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Welch, Ryan","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Seddighzadeh, Bobak","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Seddighzadeh, Bobak","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Riley, Fran","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Riley, Fran","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Stoehr, Kaitlyn","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Stoehr, Kaitlyn","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Pipinos, Margarita I.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Pipinos, Margarita I.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Speron, Dimitri","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Speron, Dimitri","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:6","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Sage, Emma V.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Sage, Emma V.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:7","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Jain, Sneha","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Jain, Sneha","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:8","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Shah, Nigam H.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shah, Nigam H.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:9","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Alsentzer, Emily","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Alsentzer, Emily","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:10","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}}],"participationCoverage":{"total":11,"returned":11,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.19343","snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z","basis":"publisher-page-metadata","sourceLinks":50,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af/source"}},"canonicalUrl":"https://escapemeridian.com/readings/arxiv-omitehr-scalable-omission-detection-in-llm-generated-1ca83ad8","formats":{"html":"/readings/arxiv-omitehr-scalable-omission-detection-in-llm-generated-1ca83ad8","markdown":"/readings/arxiv-omitehr-scalable-omission-detection-in-llm-generated-1ca83ad8.md","json":"/readings/arxiv-omitehr-scalable-omission-detection-in-llm-generated-1ca83ad8.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":[],"conditions":[],"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-llm","name":"Large language model","explanation":"A language model with many learned numerical settings, used to process and generate language.","referenceUrl":"https://developers.google.com/machine-learning/glossary#large-language-model"}],"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":"1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af","revision":"bcfe6b92-0b87-4d35-85f6-cce230a76776","revisionKind":"publication-id","title":"arXiv: OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","url":"https://escapemeridian.com/readings/arxiv-omitehr-scalable-omission-detection-in-llm-generated-1ca83ad8"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.19343","revision":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","revisionKind":"sha256-source-text","title":"OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","url":"https://arxiv.org/abs/2611.19343"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings to this reading. Shared topics are not citations. Other record types need an explicit reference adapter."},"revisions":[{"id":"1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af","title":"arXiv: OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","argument":"The authors of OmitEHR argue that omissions—clinically important information missing from a summary—are the most common failure mode in patient-record summaries generated by large language models (LLMs). Their method learns which source-note sentences clinicians included in historical summaries, then flags high-salience sentences that an LLM-generated summary does not cover. The authors evaluate OmitEHR on brief hospital course (BHC) summarization using 5,000 admissions from one large academic medical center. On held-out admissions, it outperformed the evaluated baselines at predicting salient source sentences, including a prompted LLM judge. The authors also report that its full omission detection pipeline was faster and cheaper than scoring every source sentence with an LLM judge. In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline; candidates counted if at least one clinician confirmed them. The paper presents OmitEHR as a way for health systems to flag omissions before summaries reach clinicians. These results describe the authors’ evaluations; the evidence provided here is from one medical center and a 25-admission blinded review.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The reported evaluation uses admissions from a large academic medical center; the supplied evidence does not report results from other medical centers. The blinded review covered 25 admissions, and candidates counted as confirmed if at least one of the two clinicians confirmed them.","speaker":{"name":"Lin, Bridget, Welch, Ryan, Seddighzadeh, Bobak, Riley, Fran, Stoehr, Kaitlyn, Pipinos, Margarita I., Speron, Dimitri, Sage, Emma V., Jain, Sneha, Shah, Nigam H., Alsentzer, Emily","roleAtSource":"Article author"},"source":{"title":"OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.19343","publishedAt":"2026-11-14","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors say OmitEHR learns which source-note sentences clinicians include in historical summaries and flags high-salience sentences missing from LLM-generated summaries.","excerpt":"OmitEHR learns to predict which source-note sentences clinicians include in their summaries, then surfaces high-salience sentences not covered by an LLM-generated summary as candidate omissions.","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"},{"text":"The authors evaluated OmitEHR on BHC summarization using 5,000 admissions from a large academic medical center.","excerpt":"We apply OmitEHR to brief hospital course (BHC) summarization using 5,000 admissions from a large academic medical center.","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"},{"text":"On held-out admissions, the authors report that OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","excerpt":"On held-out admissions, OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"},{"text":"In a blinded review of 25 admissions, the authors report that clinicians confirmed more clinically meaningful omissions surfaced by OmitEHR than by a full LLM pipeline.","excerpt":"In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline","url":"https://arxiv.org/abs/2611.19343","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.19343","checkedAt":"2026-11-17T13:51:01.725Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How would OmitEHR perform on admissions from medical centers beyond the one used in this evaluation?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv: OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","summary":"The authors of OmitEHR argue that omissions—clinically important information missing from a summary—are the most common failure mode in patient-record summaries generated by large language models (LLMs). Their method learns which source-note sentences clinicians included in historical summaries, then flags high-salience sentences that an LLM-generated summary does not cover. The authors evaluate OmitEHR on brief hospital course (BHC) summarization using 5,000 admissions from one large academic medical center. On held-out admissions, it outperformed the evaluated baselines at predicting salient source sentences, including a prompted LLM judge. The authors also report that its full omission detection pipeline was faster and cheaper than scoring every source sentence with an LLM judge. In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline; candidates counted if at least one clinician confirmed them. The paper presents OmitEHR as a way for health systems to flag omissions before summaries reach clinicians. These results describe the authors’ evaluations; the evidence provided here is from one medical center and a 25-admission blinded review.","question":"How would OmitEHR perform on admissions from medical centers beyond the one used in this evaluation?","claims":[{"text":"The authors say OmitEHR learns which source-note sentences clinicians include in historical summaries and flags high-salience sentences missing from LLM-generated summaries.","excerpt":"OmitEHR learns to predict which source-note sentences clinicians include in their summaries, then surfaces high-salience sentences not covered by an LLM-generated summary as candidate omissions.","support":"source-stated"},{"text":"The authors evaluated OmitEHR on BHC summarization using 5,000 admissions from a large academic medical center.","excerpt":"We apply OmitEHR to brief hospital course (BHC) summarization using 5,000 admissions from a large academic medical center.","support":"source-stated"},{"text":"On held-out admissions, the authors report that OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","excerpt":"On held-out admissions, OmitEHR outperformed all evaluated baselines at predicting salient source sentences, including a prompted LLM judge.","support":"source-stated"},{"text":"In a blinded review of 25 admissions, the authors report that clinicians confirmed more clinically meaningful omissions surfaced by OmitEHR than by a full LLM pipeline.","excerpt":"In a blinded review of 25 admissions, two board-certified clinicians confirmed a mean of 5.6 clinically meaningful omissions per summary surfaced by OmitEHR, compared with 3.3 for a full LLM pipeline","support":"source-stated"}],"limitations":["The reported evaluation uses admissions from a large academic medical center; the supplied evidence does not report results from other medical centers.","The blinded review covered 25 admissions, and candidates counted as confirmed if at least one of the two clinicians confirmed them."],"mediaId":"bb9a7a50a7963d5250e97240569146bd429414e8325b2ec4f4b6355c743a0baa","correctionReason":null,"id":"1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af","publicationId":"bcfe6b92-0b87-4d35-85f6-cce230a76776","version":1,"sourceUrl":"https://arxiv.org/abs/2611.19343","sourceTitle":"OmitEHR: Scalable Omission Detection in LLM-Generated Clinical Summaries","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-14","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-17T08:29:09.856Z","createdAt":"2026-11-17T13:18:47.337Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"bb9a7a50a7963d5250e97240569146bd429414e8325b2ec4f4b6355c743a0baa","kind":"article","url":"https://arxiv.org/abs/2611.19343","sourceUrl":"https://arxiv.org/abs/2611.19343","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-17T13:17:25.139Z","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/2611.19343","checkedAt":"2026-11-17T13:51:01.725Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"bb9a7a50a7963d5250e97240569146bd429414e8325b2ec4f4b6355c743a0baa","confidence":0.65,"provider":"typesafe","model":"jev-1.13.0","inputHash":"c0683316b3a5f171bbc748b76dcfc82b8bbb1e44cbb84f342c83cdd1adfbcede","decidedAt":"2026-11-17T13:17:27.380Z"},"provenance":{"textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observationId":"5fe1da8350a0f1c1276bc7adbf28e89ac4c77e24e9ee0c473b9cc0f2ffb2e329","retrievedAt":"2026-11-17T13:17:24.833Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"f9c0330d-3852-48bf-838e-d60e6da0b02f"},"reviewer":{"model":"gpt-6-luna","runId":"45e2fe52-38c6-411b-819d-579d8e5d24b6"}},"audit":{"status":"passed","reviewedAt":"2026-11-17T13:18:47.337Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Lin, Bridget","url":null,"kind":"unknown"},{"name":"Welch, Ryan","url":null,"kind":"unknown"},{"name":"Seddighzadeh, Bobak","url":null,"kind":"unknown"},{"name":"Riley, Fran","url":null,"kind":"unknown"},{"name":"Stoehr, Kaitlyn","url":null,"kind":"unknown"},{"name":"Pipinos, Margarita I.","url":null,"kind":"unknown"},{"name":"Speron, Dimitri","url":null,"kind":"unknown"},{"name":"Sage, Emma V.","url":null,"kind":"unknown"},{"name":"Jain, Sneha","url":null,"kind":"unknown"},{"name":"Shah, Nigam H.","url":null,"kind":"unknown"},{"name":"Alsentzer, Emily","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Lin, Bridget","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Lin, Bridget","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Welch, Ryan","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Welch, Ryan","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Seddighzadeh, Bobak","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Seddighzadeh, Bobak","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Riley, Fran","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Riley, Fran","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Stoehr, Kaitlyn","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Stoehr, Kaitlyn","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Pipinos, Margarita I.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Pipinos, Margarita I.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Speron, Dimitri","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Speron, Dimitri","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:6","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Sage, Emma V.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Sage, Emma V.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:7","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Jain, Sneha","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Jain, Sneha","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:8","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Shah, Nigam H.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shah, Nigam H.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:9","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}},{"actor":{"name":"Alsentzer, Emily","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Alsentzer, Emily","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.19343","locator":"head/meta[name=author|citation_author]"},"assertionId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6:participation:10","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-17T13:51:01.725Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","sourceUrl":"https://arxiv.org/abs/2611.19343","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z"}}],"participationCoverage":{"total":11,"returned":11,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.19343","snapshotId":"2061a33fc381f30e176edc55f7504fa578fc6dd912a4474b57b04a43c03ce3a6","contentSha256":"56b7ac40633e5d233b152f25309c995ab60120b5d9d6bcc52389e4bace706a31","textHash":"8386d7ba6f62f629d9a12e34b92d9f0fbb259392643c226044ca34009ded470a","observedAt":"2026-11-17T13:51:01.725Z","basis":"publisher-page-metadata","sourceLinks":50,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/1ca83ad8c4511551c28571d27b100377066d64a5f3962505d7052c35219989af/source"}}}],"audits":[{"publicationId":"bcfe6b92-0b87-4d35-85f6-cce230a76776","verdict":"passed","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading consistently attributes the method, results, and practical framing to the paper's authors on arXiv. It does not present the findings as independently verified or as Escape Meridian's view.","excerpt":"Authors:Bridget Lin, Ryan Welch, Bobak Seddighzadeh, Fran Riley, Kaitlyn Stoehr, Margarita I. Pipinos, Dimitri Speron, Emma V. Sage, Sneha Jain, Nigam H. Shah, Emily Alsentzer"},{"dimension":"summary","status":"pass","explanation":"The reading leads with the problem (omissions in LLM patient-record summaries), explains omissions and LLMs in plain terms, and describes the method and results accurately. It keeps the comparison condition and the authors' practical claim, and states the setting's limits. Terms such as salience and BHC could use tooltips, but the meaning still comes through. The question follows from the single-site evidence.","excerpt":"evaluations now find omissions, the absence of clinically important information, to be the most common failure mode"},{"dimension":"date","status":"pass","explanation":"originalPublished 2026-11-14 matches the v1 submission date. The feed date of 2026-11-17 is later and correctly not used.","excerpt":"[Submitted on 14 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Every claim and excerpt matches the abstract. The faster-and-cheaper statement and the at-least-one-clinician counting rule are preserved. The practical-tool framing is attributed to the paper. The limitations are drawn from the stated single center and the 25-admission review, and the reading adds no new benefits or costs.","excerpt":"The full omission detection pipeline took 17.7 to 19.5 seconds and cost \\$0.03 to \\$0.13 per summary depending on the base model, 1.6 to 1.7 times faster and 5.6 to 6.7 times cheaper than scoring every source sentence with an LLM judge."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified candidate linking directly to the arXiv abstract page. This is the original source and the easiest way for readers to access it.","excerpt":null}],"createdAt":"2026-11-17T13:18:47.337Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}