{"id":"8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844","title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","argument":"Jannatul Shefa, Alejandro Salado, Paul Wach and Taylan G. Topcu present MARQ, a prototype that adds expert-derived systems engineering knowledge to large language models (LLMs) assessing the quality of requirements. The authors evaluated it against off-the-shelf baselines on an expert-labeled NASA requirement set, using three models and 100 executions per model. They report higher median F1 scores with MARQ for all three models, but say the gains differed: some models detected more defects while also producing more false alarms or remaining variable across executions. The reported gains were concentrated in Correctness and Conformance, and the authors say human-in-the-loop review remains necessary. The results concern this evaluation; the evidence does not establish how well MARQ would perform on other requirement sets or in operational systems.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The reported evaluation used one expert-labeled NASA requirement set; the supplied evidence does not establish performance on other requirement sets or in operational systems. The authors report model-dependent results, including more false alarms for GPT-5.6 Sol and high variability for Gemma 3 12B.","speaker":{"name":"Shefa, Jannatul, Salado, Alejandro, Wach, Paul, Topcu, Taylan G.","roleAtSource":"Article author"},"source":{"title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.36234","publishedAt":"2026-10-26","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors present MARQ as a modular prototype that supplements LLMs with expert-derived systems engineering knowledge for requirement quality assessment.","excerpt":"a modular, LLM-agnostic prototype that supplements LLMs with a criterion-specific knowledge stack","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The evaluation used an expert-labeled NASA requirement set, three named models, and 100 executions per model.","excerpt":"evaluated against off-the-shelf baselines on an expert-labeled NASA requirement set using Claude Opus 5, GPT-5.6 Sol, and the open-weight Gemma 3 12B, with 100 executions per model","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The authors report that MARQ achieved higher median F1 scores than off-the-shelf baselines for each evaluated model.","excerpt":"MARQ achieves median F1 scores of 0.80 for Claude Opus 5, 0.62 for GPT-5.6 Sol, and 0.53 for Gemma 3 12B, compared with off-the-shelf baselines of 0.55, 0.52, and 0.18, respectively.","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The authors report that the gains varied by model, including trade-offs involving defect detection, false alarms, and execution-level consistency.","excerpt":"the nature of these gains is model-dependent. Claude Opus 5 improves defect detection and false-alarm control; GPT-5.6 Sol gains detection and better execution-level consistency at the cost of more false alarms","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The authors say the evaluation's gains were concentrated in Correctness and Conformance and call for human-in-the-loop review.","excerpt":"Criterion-level analysis shows gains concentrated in Correctness and Conformance, trade-offs between misses and false alarms within individual criteria","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.36234","checkedAt":"2026-11-06T15:16:08.389Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Would MARQ show similar results on requirement sets beyond the expert-labeled NASA set used in this evaluation?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","summary":"Jannatul Shefa, Alejandro Salado, Paul Wach and Taylan G. Topcu present MARQ, a prototype that adds expert-derived systems engineering knowledge to large language models (LLMs) assessing the quality of requirements. The authors evaluated it against off-the-shelf baselines on an expert-labeled NASA requirement set, using three models and 100 executions per model. They report higher median F1 scores with MARQ for all three models, but say the gains differed: some models detected more defects while also producing more false alarms or remaining variable across executions. The reported gains were concentrated in Correctness and Conformance, and the authors say human-in-the-loop review remains necessary. The results concern this evaluation; the evidence does not establish how well MARQ would perform on other requirement sets or in operational systems.","question":"Would MARQ show similar results on requirement sets beyond the expert-labeled NASA set used in this evaluation?","claims":[{"text":"The authors present MARQ as a modular prototype that supplements LLMs with expert-derived systems engineering knowledge for requirement quality assessment.","excerpt":"a modular, LLM-agnostic prototype that supplements LLMs with a criterion-specific knowledge stack","support":"source-stated"},{"text":"The evaluation used an expert-labeled NASA requirement set, three named models, and 100 executions per model.","excerpt":"evaluated against off-the-shelf baselines on an expert-labeled NASA requirement set using Claude Opus 5, GPT-5.6 Sol, and the open-weight Gemma 3 12B, with 100 executions per model","support":"source-stated"},{"text":"The authors report that MARQ achieved higher median F1 scores than off-the-shelf baselines for each evaluated model.","excerpt":"MARQ achieves median F1 scores of 0.80 for Claude Opus 5, 0.62 for GPT-5.6 Sol, and 0.53 for Gemma 3 12B, compared with off-the-shelf baselines of 0.55, 0.52, and 0.18, respectively.","support":"source-stated"},{"text":"The authors report that the gains varied by model, including trade-offs involving defect detection, false alarms, and execution-level consistency.","excerpt":"the nature of these gains is model-dependent. Claude Opus 5 improves defect detection and false-alarm control; GPT-5.6 Sol gains detection and better execution-level consistency at the cost of more false alarms","support":"source-stated"},{"text":"The authors say the evaluation's gains were concentrated in Correctness and Conformance and call for human-in-the-loop review.","excerpt":"Criterion-level analysis shows gains concentrated in Correctness and Conformance, trade-offs between misses and false alarms within individual criteria","support":"source-stated"}],"limitations":["The reported evaluation used one expert-labeled NASA requirement set; the supplied evidence does not establish performance on other requirement sets or in operational systems.","The authors report model-dependent results, including more false alarms for GPT-5.6 Sol and high variability for Gemma 3 12B."],"mediaId":"035f71ec0325679b7fbcffecb529d66cbc298645a8b4c569e0839a18e3ee7c49","correctionReason":null,"id":"8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844","publicationId":"7d25ac92-0bc3-4db3-8db6-142825420485","version":1,"sourceUrl":"https://arxiv.org/abs/2610.36234","sourceTitle":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-26","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-28T07:52:39.778Z","createdAt":"2026-11-06T11:01:16.745Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"035f71ec0325679b7fbcffecb529d66cbc298645a8b4c569e0839a18e3ee7c49","kind":"article","url":"https://arxiv.org/abs/2610.36234","sourceUrl":"https://arxiv.org/abs/2610.36234","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-30T03:02:35.585Z","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/2610.36234","checkedAt":"2026-11-06T15:16:08.389Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"035f71ec0325679b7fbcffecb529d66cbc298645a8b4c569e0839a18e3ee7c49","confidence":0.81,"provider":"typesafe","model":"jev-1.13.0","inputHash":"2bd5460359ca46bcd2a108fc8e63a50c2cdf12d889b14b75f17a2ad419e69311","decidedAt":"2026-10-30T03:02:36.886Z"},"provenance":{"textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observationId":"8f370abba6a70900773f6aaf1e9e9fea546c06c2294375e86de439f6c1e6713c","retrievedAt":"2026-10-30T03:02:34.516Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"914b1183-fac7-4f1d-89f5-9552e499f531"},"reviewer":{"model":"gpt-6-luna","runId":"c36e8d95-3580-4b56-8f06-884ffbc56e1d"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-06T11:01:16.745Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Shefa, Jannatul","url":null,"kind":"unknown"},{"name":"Salado, Alejandro","url":null,"kind":"unknown"},{"name":"Wach, Paul","url":null,"kind":"unknown"},{"name":"Topcu, Taylan G.","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Shefa, Jannatul","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shefa, Jannatul","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}},{"actor":{"name":"Salado, Alejandro","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Salado, Alejandro","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}},{"actor":{"name":"Wach, Paul","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Wach, Paul","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}},{"actor":{"name":"Topcu, Taylan G.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Topcu, Taylan G.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}}],"participationCoverage":{"total":4,"returned":4,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.36234","snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z","basis":"publisher-page-metadata","sourceLinks":40,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844/source"}},"canonicalUrl":"https://escapemeridian.com/readings/marq-a-modular-llm-agnostic-requirement-quality-assessor-8185c111","formats":{"html":"/readings/marq-a-modular-llm-agnostic-requirement-quality-assessor-8185c111","markdown":"/readings/marq-a-modular-llm-agnostic-requirement-quality-assessor-8185c111.md","json":"/readings/marq-a-modular-llm-agnostic-requirement-quality-assessor-8185c111.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-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"},{"id":"nasa","name":"NASA","kind":"organization","context":"NASA is the US agency for civilian space exploration, aeronautics and related scientific research.","referenceUrls":["https://www.nasa.gov/","https://www.earthdata.nasa.gov/"],"path":"/entities/nasa","json":"/entities/nasa.json","markdown":"/entities/nasa.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844","revision":"7d25ac92-0bc3-4db3-8db6-142825420485","revisionKind":"publication-id","title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","url":"https://escapemeridian.com/readings/marq-a-modular-llm-agnostic-requirement-quality-assessor-8185c111"},"source":{"type":"source","id":"https://arxiv.org/abs/2610.36234","revision":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","revisionKind":"sha256-source-text","title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","url":"https://arxiv.org/abs/2610.36234"},"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":"8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844","title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","argument":"Jannatul Shefa, Alejandro Salado, Paul Wach and Taylan G. Topcu present MARQ, a prototype that adds expert-derived systems engineering knowledge to large language models (LLMs) assessing the quality of requirements. The authors evaluated it against off-the-shelf baselines on an expert-labeled NASA requirement set, using three models and 100 executions per model. They report higher median F1 scores with MARQ for all three models, but say the gains differed: some models detected more defects while also producing more false alarms or remaining variable across executions. The reported gains were concentrated in Correctness and Conformance, and the authors say human-in-the-loop review remains necessary. The results concern this evaluation; the evidence does not establish how well MARQ would perform on other requirement sets or in operational systems.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The reported evaluation used one expert-labeled NASA requirement set; the supplied evidence does not establish performance on other requirement sets or in operational systems. The authors report model-dependent results, including more false alarms for GPT-5.6 Sol and high variability for Gemma 3 12B.","speaker":{"name":"Shefa, Jannatul, Salado, Alejandro, Wach, Paul, Topcu, Taylan G.","roleAtSource":"Article author"},"source":{"title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.36234","publishedAt":"2026-10-26","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors present MARQ as a modular prototype that supplements LLMs with expert-derived systems engineering knowledge for requirement quality assessment.","excerpt":"a modular, LLM-agnostic prototype that supplements LLMs with a criterion-specific knowledge stack","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The evaluation used an expert-labeled NASA requirement set, three named models, and 100 executions per model.","excerpt":"evaluated against off-the-shelf baselines on an expert-labeled NASA requirement set using Claude Opus 5, GPT-5.6 Sol, and the open-weight Gemma 3 12B, with 100 executions per model","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The authors report that MARQ achieved higher median F1 scores than off-the-shelf baselines for each evaluated model.","excerpt":"MARQ achieves median F1 scores of 0.80 for Claude Opus 5, 0.62 for GPT-5.6 Sol, and 0.53 for Gemma 3 12B, compared with off-the-shelf baselines of 0.55, 0.52, and 0.18, respectively.","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The authors report that the gains varied by model, including trade-offs involving defect detection, false alarms, and execution-level consistency.","excerpt":"the nature of these gains is model-dependent. Claude Opus 5 improves defect detection and false-alarm control; GPT-5.6 Sol gains detection and better execution-level consistency at the cost of more false alarms","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"},{"text":"The authors say the evaluation's gains were concentrated in Correctness and Conformance and call for human-in-the-loop review.","excerpt":"Criterion-level analysis shows gains concentrated in Correctness and Conformance, trade-offs between misses and false alarms within individual criteria","url":"https://arxiv.org/abs/2610.36234","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.36234","checkedAt":"2026-11-06T15:16:08.389Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Would MARQ show similar results on requirement sets beyond the expert-labeled NASA set used in this evaluation?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","summary":"Jannatul Shefa, Alejandro Salado, Paul Wach and Taylan G. Topcu present MARQ, a prototype that adds expert-derived systems engineering knowledge to large language models (LLMs) assessing the quality of requirements. The authors evaluated it against off-the-shelf baselines on an expert-labeled NASA requirement set, using three models and 100 executions per model. They report higher median F1 scores with MARQ for all three models, but say the gains differed: some models detected more defects while also producing more false alarms or remaining variable across executions. The reported gains were concentrated in Correctness and Conformance, and the authors say human-in-the-loop review remains necessary. The results concern this evaluation; the evidence does not establish how well MARQ would perform on other requirement sets or in operational systems.","question":"Would MARQ show similar results on requirement sets beyond the expert-labeled NASA set used in this evaluation?","claims":[{"text":"The authors present MARQ as a modular prototype that supplements LLMs with expert-derived systems engineering knowledge for requirement quality assessment.","excerpt":"a modular, LLM-agnostic prototype that supplements LLMs with a criterion-specific knowledge stack","support":"source-stated"},{"text":"The evaluation used an expert-labeled NASA requirement set, three named models, and 100 executions per model.","excerpt":"evaluated against off-the-shelf baselines on an expert-labeled NASA requirement set using Claude Opus 5, GPT-5.6 Sol, and the open-weight Gemma 3 12B, with 100 executions per model","support":"source-stated"},{"text":"The authors report that MARQ achieved higher median F1 scores than off-the-shelf baselines for each evaluated model.","excerpt":"MARQ achieves median F1 scores of 0.80 for Claude Opus 5, 0.62 for GPT-5.6 Sol, and 0.53 for Gemma 3 12B, compared with off-the-shelf baselines of 0.55, 0.52, and 0.18, respectively.","support":"source-stated"},{"text":"The authors report that the gains varied by model, including trade-offs involving defect detection, false alarms, and execution-level consistency.","excerpt":"the nature of these gains is model-dependent. Claude Opus 5 improves defect detection and false-alarm control; GPT-5.6 Sol gains detection and better execution-level consistency at the cost of more false alarms","support":"source-stated"},{"text":"The authors say the evaluation's gains were concentrated in Correctness and Conformance and call for human-in-the-loop review.","excerpt":"Criterion-level analysis shows gains concentrated in Correctness and Conformance, trade-offs between misses and false alarms within individual criteria","support":"source-stated"}],"limitations":["The reported evaluation used one expert-labeled NASA requirement set; the supplied evidence does not establish performance on other requirement sets or in operational systems.","The authors report model-dependent results, including more false alarms for GPT-5.6 Sol and high variability for Gemma 3 12B."],"mediaId":"035f71ec0325679b7fbcffecb529d66cbc298645a8b4c569e0839a18e3ee7c49","correctionReason":null,"id":"8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844","publicationId":"7d25ac92-0bc3-4db3-8db6-142825420485","version":1,"sourceUrl":"https://arxiv.org/abs/2610.36234","sourceTitle":"MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-26","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-28T07:52:39.778Z","createdAt":"2026-11-06T11:01:16.745Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"035f71ec0325679b7fbcffecb529d66cbc298645a8b4c569e0839a18e3ee7c49","kind":"article","url":"https://arxiv.org/abs/2610.36234","sourceUrl":"https://arxiv.org/abs/2610.36234","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-30T03:02:35.585Z","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/2610.36234","checkedAt":"2026-11-06T15:16:08.389Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"035f71ec0325679b7fbcffecb529d66cbc298645a8b4c569e0839a18e3ee7c49","confidence":0.81,"provider":"typesafe","model":"jev-1.13.0","inputHash":"2bd5460359ca46bcd2a108fc8e63a50c2cdf12d889b14b75f17a2ad419e69311","decidedAt":"2026-10-30T03:02:36.886Z"},"provenance":{"textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observationId":"8f370abba6a70900773f6aaf1e9e9fea546c06c2294375e86de439f6c1e6713c","retrievedAt":"2026-10-30T03:02:34.516Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"914b1183-fac7-4f1d-89f5-9552e499f531"},"reviewer":{"model":"gpt-6-luna","runId":"c36e8d95-3580-4b56-8f06-884ffbc56e1d"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-06T11:01:16.745Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Shefa, Jannatul","url":null,"kind":"unknown"},{"name":"Salado, Alejandro","url":null,"kind":"unknown"},{"name":"Wach, Paul","url":null,"kind":"unknown"},{"name":"Topcu, Taylan G.","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Shefa, Jannatul","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shefa, Jannatul","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}},{"actor":{"name":"Salado, Alejandro","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Salado, Alejandro","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}},{"actor":{"name":"Wach, Paul","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Wach, Paul","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}},{"actor":{"name":"Topcu, Taylan G.","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Topcu, Taylan G.","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.36234","locator":"head/meta[name=author|citation_author]"},"assertionId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-06T11:02:07.293Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","sourceUrl":"https://arxiv.org/abs/2610.36234","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z"}}],"participationCoverage":{"total":4,"returned":4,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.36234","snapshotId":"500ba0f71807c765708b0268b55145687818fc611f43644efc028188fa346512","contentSha256":"9b31fa41af5009fe09806c4b07b3d5b334115b32508c700a69bc8f80a8585cd3","textHash":"362edce06bec4f03d0c165a43c72fe4676a4ca5ae76fc8cbe5340fefded59037","observedAt":"2026-11-06T11:02:07.293Z","basis":"publisher-page-metadata","sourceLinks":40,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/8185c11123378ec95202bf72929420b6ee067519a88a1a90e666210c3d5f9844/source"}}}],"audits":[{"publicationId":"7d25ac92-0bc3-4db3-8db6-142825420485","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly names the four authors listed on the arXiv page and keeps the results and recommendations attributed to them, not presented as Escape Meridian's view or as established fact.","excerpt":"Authors:Jannatul Shefa, Alejandro Salado, Paul Wach, Taylan G. Topcu"},{"dimension":"summary","status":"uncertain","explanation":"The summary is faithful, but a non-specialist would struggle to follow it. F1 score, the central metric, is never explained. Correctness and Conformance are INCOSE criteria and are unexplained, and INCOSE is not named. The draft also omits why requirement quality matters: the source says defects propagate downstream and harm development outcomes. Smallest repair: add tooltips for F1, requirements and the INCOSE criteria, and one sentence on downstream defect propagation. The draft also drops the source's point that Gemma 3 12B with MARQ reaches the level of off-the-shelf proprietary models.","excerpt":"requirement defects can propagate downstream and adversely affect development outcomes"},{"dimension":"date","status":"pass","explanation":"The originalPublished date of 2026-10-26 matches the v1 submission date on the arXiv page. It is not the later feed date.","excerpt":"[Submitted on 26 Oct 2026]"},{"dimension":"support","status":"pass","explanation":"The source abstract supports all claims and limitations, including the scores, model-dependent trade-offs, criterion concentration and the call for human-in-the-loop review. The question about other requirement sets is a fair, non-leading gap. A minor issue: claim 5's excerpt does not contain the human-in-the-loop phrase, though the source states it elsewhere. Do not read this as independent verification of the reported scores.","excerpt":"underscoring the need for criterion-specific interventions and human-in-the-loop review"},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper, which gives readers the easiest access to the source.","excerpt":null}],"createdAt":"2026-11-06T11:01:16.745Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}