{"id":"fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4","title":"How LLM Capability in Software Vulnerability Research Is Changing","argument":"Authors Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini and Dinesh Manocha argue that claims about LLM-based agents finding and fixing software vulnerabilities are difficult to compare because they are scattered across papers, vendor reports and disclosures. They introduce a framework that rates capability on an autonomy scale and assesses each claim’s correctness, availability and strength of evidence. Applying it, they report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models appeared in 2026; open-weight models later reached comparable autonomy for discovery and validation. The authors also examine public vulnerability and exploitation records. They report that disclosed vulnerabilities more than doubled in 2026, with a growing share rated critical or high, and that severe bugs long present in widely used code were reported several times more often than before. Yet the intervals from report to fix and from disclosure to attack had not noticeably changed. The contrast matters: more reported vulnerabilities did not, in the authors’ account, coincide with a noticeable shift in those intervals. The abstract describes the framework’s results and says the paper grades proposed mitigations, but does not detail its evidence or rubric.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The supplied evidence is the arXiv abstract; it does not include the paper’s detailed methods, underlying records, evidence assessments or mitigation rubric. The abstract reports associations and timing patterns but does not explain why the number of disclosures increased or why the intervals had not noticeably shifted.","speaker":{"name":"Lbath, Amine, Suri, Manan, Delaitre, Aurelien, Okun, Vadim, Amini, Massih-Reza, Manocha, Dinesh","roleAtSource":"Article author"},"source":{"title":"SoK: Assessing the Evolution of LLM Capability in Software Vulnerability Research","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.26601","publishedAt":"2026-11-19","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors say claims about LLM-based vulnerability research are scattered and rarely comparable, and introduce a framework to assess capability and evidence.","excerpt":"These claims, however, are scattered across papers, vendor reports and disclosures and are rarely comparable.","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"},{"text":"The authors report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models of 2026; open-weight models later reached comparable autonomy for discovery and validation.","excerpt":"agents discovered, validated, and patched vulnerabilities in real-world software without human intervention before the restricted frontier models of 2026","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"},{"text":"The authors report a sharp rise in disclosed vulnerabilities in 2026, including more critical or high-rated vulnerabilities and more reports of severe bugs that had persisted in widely used code.","excerpt":"In 2026 the number of disclosed vulnerabilities more than doubled, with an increasing share rated critical or high","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"},{"text":"The authors say the time between vulnerability report and fix, and between disclosure and attack, had not noticeably shifted so far.","excerpt":"the time between the stages of a vulnerability's life, from report to fix and from disclosure to attack, has not shifted noticeably so far","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.26601","checkedAt":"2026-11-21T02:06:09.763Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"What explains the authors’ finding that the intervals between vulnerability report, fix, disclosure and attack had not noticeably changed?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"How LLM Capability in Software Vulnerability Research Is Changing","summary":"Authors Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini and Dinesh Manocha argue that claims about LLM-based agents finding and fixing software vulnerabilities are difficult to compare because they are scattered across papers, vendor reports and disclosures. They introduce a framework that rates capability on an autonomy scale and assesses each claim’s correctness, availability and strength of evidence. Applying it, they report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models appeared in 2026; open-weight models later reached comparable autonomy for discovery and validation. The authors also examine public vulnerability and exploitation records. They report that disclosed vulnerabilities more than doubled in 2026, with a growing share rated critical or high, and that severe bugs long present in widely used code were reported several times more often than before. Yet the intervals from report to fix and from disclosure to attack had not noticeably changed. The contrast matters: more reported vulnerabilities did not, in the authors’ account, coincide with a noticeable shift in those intervals. The abstract describes the framework’s results and says the paper grades proposed mitigations, but does not detail its evidence or rubric.","question":"What explains the authors’ finding that the intervals between vulnerability report, fix, disclosure and attack had not noticeably changed?","claims":[{"text":"The authors say claims about LLM-based vulnerability research are scattered and rarely comparable, and introduce a framework to assess capability and evidence.","excerpt":"These claims, however, are scattered across papers, vendor reports and disclosures and are rarely comparable.","support":"source-stated"},{"text":"The authors report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models of 2026; open-weight models later reached comparable autonomy for discovery and validation.","excerpt":"agents discovered, validated, and patched vulnerabilities in real-world software without human intervention before the restricted frontier models of 2026","support":"source-stated"},{"text":"The authors report a sharp rise in disclosed vulnerabilities in 2026, including more critical or high-rated vulnerabilities and more reports of severe bugs that had persisted in widely used code.","excerpt":"In 2026 the number of disclosed vulnerabilities more than doubled, with an increasing share rated critical or high","support":"source-stated"},{"text":"The authors say the time between vulnerability report and fix, and between disclosure and attack, had not noticeably shifted so far.","excerpt":"the time between the stages of a vulnerability's life, from report to fix and from disclosure to attack, has not shifted noticeably so far","support":"source-stated"}],"limitations":["The supplied evidence is the arXiv abstract; it does not include the paper’s detailed methods, underlying records, evidence assessments or mitigation rubric.","The abstract reports associations and timing patterns but does not explain why the number of disclosures increased or why the intervals had not noticeably shifted."],"mediaId":"8051b8fe8ceefcca2d27793a79dd5d856aaa5f707ede08812fd1f8b8dbebae3f","correctionReason":null,"id":"fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4","publicationId":"8069f6bc-1d3e-4e37-815b-5da165e595c8","version":1,"sourceUrl":"https://arxiv.org/abs/2611.26601","sourceTitle":"SoK: Assessing the Evolution of LLM Capability in Software Vulnerability Research","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-19","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-20T05:10:23.421Z","createdAt":"2026-11-20T09:03:29.938Z","historical":false,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"8051b8fe8ceefcca2d27793a79dd5d856aaa5f707ede08812fd1f8b8dbebae3f","kind":"article","url":"https://arxiv.org/abs/2611.26601","sourceUrl":"https://arxiv.org/abs/2611.26601","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-20T09:02:41.052Z","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/2611.26601","checkedAt":"2026-11-21T02:06:09.763Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"8051b8fe8ceefcca2d27793a79dd5d856aaa5f707ede08812fd1f8b8dbebae3f","confidence":0.8,"provider":"typesafe","model":"jev-1.13.0","inputHash":"0189d44a593ac1dc07b15a6cefdcd06ad4241a06f1547e26de6054ca797f2fa0","decidedAt":"2026-11-20T09:02:41.761Z"},"provenance":{"textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observationId":"f835ab29c8adee2d98d155ff460cdb8ee037edaa8eff1bd9c32cc25a47e46c3a","retrievedAt":"2026-11-20T09:02:40.959Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"dfc24835-4bde-48bb-88f7-a50a96737bfd"},"reviewer":{"model":"gpt-6-luna","runId":"9036e79f-c390-4d59-8a54-66411c7c4b60"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-20T09:03:29.938Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Lbath, Amine","url":null,"kind":"unknown"},{"name":"Suri, Manan","url":null,"kind":"unknown"},{"name":"Delaitre, Aurelien","url":null,"kind":"unknown"},{"name":"Okun, Vadim","url":null,"kind":"unknown"},{"name":"Amini, Massih-Reza","url":null,"kind":"unknown"},{"name":"Manocha, Dinesh","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Lbath, Amine","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Lbath, Amine","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Suri, Manan","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Suri, Manan","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Delaitre, Aurelien","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Delaitre, Aurelien","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Okun, Vadim","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Okun, Vadim","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Amini, Massih-Reza","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Amini, Massih-Reza","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Manocha, Dinesh","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Manocha, Dinesh","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}}],"participationCoverage":{"total":6,"returned":6,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.26601","snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z","basis":"publisher-page-metadata","sourceLinks":46,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4/source"}},"canonicalUrl":"https://escapemeridian.com/readings/how-llm-capability-in-software-vulnerability-research-is-fba7135c","formats":{"html":"/readings/how-llm-capability-in-software-vulnerability-research-is-fba7135c","markdown":"/readings/how-llm-capability-in-software-vulnerability-research-is-fba7135c.md","json":"/readings/how-llm-capability-in-software-vulnerability-research-is-fba7135c.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"},{"id":"term-frontier-model","name":"Frontier model","explanation":"One of the most capable AI models available at a given time; the set shifts as models improve.","referenceUrl":"https://www.gov.uk/government/publications/frontier-ai-capabilities-and-risks-discussion-paper"},{"id":"term-open-weights","name":"Open weights","explanation":"A model whose trained numbers are published so others can download, run and inspect it; the license and training data may still be restricted.","referenceUrl":"https://opensource.org/ai/open-weights"}],"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":"fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4","revision":"8069f6bc-1d3e-4e37-815b-5da165e595c8","revisionKind":"publication-id","title":"How LLM Capability in Software Vulnerability Research Is Changing","url":"https://escapemeridian.com/readings/how-llm-capability-in-software-vulnerability-research-is-fba7135c"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.26601","revision":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","revisionKind":"sha256-source-text","title":"SoK: Assessing the Evolution of LLM Capability in Software Vulnerability Research","url":"https://arxiv.org/abs/2611.26601"},"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":"fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4","title":"How LLM Capability in Software Vulnerability Research Is Changing","argument":"Authors Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini and Dinesh Manocha argue that claims about LLM-based agents finding and fixing software vulnerabilities are difficult to compare because they are scattered across papers, vendor reports and disclosures. They introduce a framework that rates capability on an autonomy scale and assesses each claim’s correctness, availability and strength of evidence. Applying it, they report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models appeared in 2026; open-weight models later reached comparable autonomy for discovery and validation. The authors also examine public vulnerability and exploitation records. They report that disclosed vulnerabilities more than doubled in 2026, with a growing share rated critical or high, and that severe bugs long present in widely used code were reported several times more often than before. Yet the intervals from report to fix and from disclosure to attack had not noticeably changed. The contrast matters: more reported vulnerabilities did not, in the authors’ account, coincide with a noticeable shift in those intervals. The abstract describes the framework’s results and says the paper grades proposed mitigations, but does not detail its evidence or rubric.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The supplied evidence is the arXiv abstract; it does not include the paper’s detailed methods, underlying records, evidence assessments or mitigation rubric. The abstract reports associations and timing patterns but does not explain why the number of disclosures increased or why the intervals had not noticeably shifted.","speaker":{"name":"Lbath, Amine, Suri, Manan, Delaitre, Aurelien, Okun, Vadim, Amini, Massih-Reza, Manocha, Dinesh","roleAtSource":"Article author"},"source":{"title":"SoK: Assessing the Evolution of LLM Capability in Software Vulnerability Research","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.26601","publishedAt":"2026-11-19","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors say claims about LLM-based vulnerability research are scattered and rarely comparable, and introduce a framework to assess capability and evidence.","excerpt":"These claims, however, are scattered across papers, vendor reports and disclosures and are rarely comparable.","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"},{"text":"The authors report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models of 2026; open-weight models later reached comparable autonomy for discovery and validation.","excerpt":"agents discovered, validated, and patched vulnerabilities in real-world software without human intervention before the restricted frontier models of 2026","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"},{"text":"The authors report a sharp rise in disclosed vulnerabilities in 2026, including more critical or high-rated vulnerabilities and more reports of severe bugs that had persisted in widely used code.","excerpt":"In 2026 the number of disclosed vulnerabilities more than doubled, with an increasing share rated critical or high","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"},{"text":"The authors say the time between vulnerability report and fix, and between disclosure and attack, had not noticeably shifted so far.","excerpt":"the time between the stages of a vulnerability's life, from report to fix and from disclosure to attack, has not shifted noticeably so far","url":"https://arxiv.org/abs/2611.26601","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.26601","checkedAt":"2026-11-21T02:06:09.763Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"What explains the authors’ finding that the intervals between vulnerability report, fix, disclosure and attack had not noticeably changed?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"How LLM Capability in Software Vulnerability Research Is Changing","summary":"Authors Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini and Dinesh Manocha argue that claims about LLM-based agents finding and fixing software vulnerabilities are difficult to compare because they are scattered across papers, vendor reports and disclosures. They introduce a framework that rates capability on an autonomy scale and assesses each claim’s correctness, availability and strength of evidence. Applying it, they report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models appeared in 2026; open-weight models later reached comparable autonomy for discovery and validation. The authors also examine public vulnerability and exploitation records. They report that disclosed vulnerabilities more than doubled in 2026, with a growing share rated critical or high, and that severe bugs long present in widely used code were reported several times more often than before. Yet the intervals from report to fix and from disclosure to attack had not noticeably changed. The contrast matters: more reported vulnerabilities did not, in the authors’ account, coincide with a noticeable shift in those intervals. The abstract describes the framework’s results and says the paper grades proposed mitigations, but does not detail its evidence or rubric.","question":"What explains the authors’ finding that the intervals between vulnerability report, fix, disclosure and attack had not noticeably changed?","claims":[{"text":"The authors say claims about LLM-based vulnerability research are scattered and rarely comparable, and introduce a framework to assess capability and evidence.","excerpt":"These claims, however, are scattered across papers, vendor reports and disclosures and are rarely comparable.","support":"source-stated"},{"text":"The authors report that agents found, validated and patched vulnerabilities in real-world software without human intervention before restricted frontier models of 2026; open-weight models later reached comparable autonomy for discovery and validation.","excerpt":"agents discovered, validated, and patched vulnerabilities in real-world software without human intervention before the restricted frontier models of 2026","support":"source-stated"},{"text":"The authors report a sharp rise in disclosed vulnerabilities in 2026, including more critical or high-rated vulnerabilities and more reports of severe bugs that had persisted in widely used code.","excerpt":"In 2026 the number of disclosed vulnerabilities more than doubled, with an increasing share rated critical or high","support":"source-stated"},{"text":"The authors say the time between vulnerability report and fix, and between disclosure and attack, had not noticeably shifted so far.","excerpt":"the time between the stages of a vulnerability's life, from report to fix and from disclosure to attack, has not shifted noticeably so far","support":"source-stated"}],"limitations":["The supplied evidence is the arXiv abstract; it does not include the paper’s detailed methods, underlying records, evidence assessments or mitigation rubric.","The abstract reports associations and timing patterns but does not explain why the number of disclosures increased or why the intervals had not noticeably shifted."],"mediaId":"8051b8fe8ceefcca2d27793a79dd5d856aaa5f707ede08812fd1f8b8dbebae3f","correctionReason":null,"id":"fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4","publicationId":"8069f6bc-1d3e-4e37-815b-5da165e595c8","version":1,"sourceUrl":"https://arxiv.org/abs/2611.26601","sourceTitle":"SoK: Assessing the Evolution of LLM Capability in Software Vulnerability Research","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-19","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-20T05:10:23.421Z","createdAt":"2026-11-20T09:03:29.938Z","historical":false,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"8051b8fe8ceefcca2d27793a79dd5d856aaa5f707ede08812fd1f8b8dbebae3f","kind":"article","url":"https://arxiv.org/abs/2611.26601","sourceUrl":"https://arxiv.org/abs/2611.26601","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-20T09:02:41.052Z","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/2611.26601","checkedAt":"2026-11-21T02:06:09.763Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"8051b8fe8ceefcca2d27793a79dd5d856aaa5f707ede08812fd1f8b8dbebae3f","confidence":0.8,"provider":"typesafe","model":"jev-1.13.0","inputHash":"0189d44a593ac1dc07b15a6cefdcd06ad4241a06f1547e26de6054ca797f2fa0","decidedAt":"2026-11-20T09:02:41.761Z"},"provenance":{"textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observationId":"f835ab29c8adee2d98d155ff460cdb8ee037edaa8eff1bd9c32cc25a47e46c3a","retrievedAt":"2026-11-20T09:02:40.959Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"dfc24835-4bde-48bb-88f7-a50a96737bfd"},"reviewer":{"model":"gpt-6-luna","runId":"9036e79f-c390-4d59-8a54-66411c7c4b60"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-20T09:03:29.938Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Lbath, Amine","url":null,"kind":"unknown"},{"name":"Suri, Manan","url":null,"kind":"unknown"},{"name":"Delaitre, Aurelien","url":null,"kind":"unknown"},{"name":"Okun, Vadim","url":null,"kind":"unknown"},{"name":"Amini, Massih-Reza","url":null,"kind":"unknown"},{"name":"Manocha, Dinesh","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Lbath, Amine","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Lbath, Amine","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Suri, Manan","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Suri, Manan","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Delaitre, Aurelien","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Delaitre, Aurelien","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Okun, Vadim","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Okun, Vadim","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Amini, Massih-Reza","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Amini, Massih-Reza","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}},{"actor":{"name":"Manocha, Dinesh","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Manocha, Dinesh","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.26601","locator":"head/meta[name=author|citation_author]"},"assertionId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-21T02:06:09.763Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","sourceUrl":"https://arxiv.org/abs/2611.26601","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z"}}],"participationCoverage":{"total":6,"returned":6,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.26601","snapshotId":"4c5e0c3d6cfbd0e95812224df4103a52382affab0258f021febb8a59a533777e","contentSha256":"576e56504a40cd70b83d1a1debf537e3ed00a6724242f72beb9f300d67e6ed94","textHash":"51c68bc798d251fa952f781a4997a8309a424637588ad1110fc45a03930d598e","observedAt":"2026-11-21T02:06:09.763Z","basis":"publisher-page-metadata","sourceLinks":46,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/fba7135c60cecdea6417e2c56e570478910f6d634f70a1607f0cc27381f2beb4/source"}}}],"audits":[{"publicationId":"8069f6bc-1d3e-4e37-815b-5da165e595c8","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading attributes the framework, findings and measurements to the six listed arXiv authors, and the names match the source. Findings are framed as the authors' reports rather than established fact.","excerpt":"Authors:Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini, Dinesh Manocha"},{"dimension":"summary","status":"uncertain","explanation":"The summary follows the abstract faithfully, but a general reader may struggle with central terms. LLM, LLM-based agents, restricted frontier models and open-weight models are never explained, and the open-weight finding depends on them. The autonomy scale loses the source's plain-language anchor: it is modeled on the levels of driving automation. 'Long present' also softens the source's 'a decade or more'. The 'contrast matters' sentence repeats the finding without saying why it matters. Smallest repair: briefly gloss these terms, restore the driving-automation analogy and the decade-or-more scope, and keep 'so far' in the summary text.","excerpt":"rates these capabilities on an autonomy scale modeled on the levels of driving automation"},{"dimension":"date","status":"pass","explanation":"The originalPublished date of 2026-11-19 matches the v1 submission date shown on the arXiv page. The feed date of 2026-11-20 is correctly not used.","excerpt":"[Submitted on 19 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Each claim's excerpt is verbatim and supports it. The limitations correctly note that the evidence is only the abstract and that it gives no causal explanation. The question asks about an open issue the abstract leaves unexplained rather than asserting an answer. No benefits, costs or forecasts are invented.","excerpt":"Yet the time between the stages of a vulnerability's life, from report to fix and from disclosure to attack, has not shifted noticeably so far."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper, which is the most direct access to the source.","excerpt":null}],"createdAt":"2026-11-20T09:03:29.938Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}