{"id":"2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f","title":"arXiv cs.AI paper introduces LiveSearch, a live benchmark for financial search agents","argument":"In an arXiv paper, Sinuo Wang and coauthors introduce LiveSearch, a benchmark designed to test whether financial search agents can find current answers rather than repeat outdated ones. Its questions are stored as executable verifier code: instructions that call a professional financial data service and extract an answer. The question stays fixed, while the benchmark recalculates the answer from live data at evaluation time and reruns the checks daily. The authors say this makes memorized answers likely to go stale and requires models to use tools and analysis. They report that, under general web search, six contemporary tool-calling models reached 70.9% to 76.7% answer accuracy. They also report that 70% of served questions changed their gold value over three consecutive days, and that reused earlier answers failed on a majority of questions within one day, even with tolerance-aware matching. These results describe the benchmark and its reported evaluations; they do not establish how agents perform in real financial work. How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The supplied evidence is the paper's abstract; it does not provide details about benchmark coverage or how well its tasks represent financial work outside the benchmark. The reported accuracy results are benchmark evaluations and do not establish agents' reliability in real-world financial work.","speaker":{"name":"Wang, Sinuo, Zhao, Wentao, Lu, Jiacheng, Sun, Rui, Luan, Beidi, Shi, Qiming, Li, Jing, Jiang, Daxin, Bai, Zuo","roleAtSource":"Article author"},"source":{"title":"LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.37166","publishedAt":"2026-10-27","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors introduce LiveSearch, a benchmark that uses executable verifier code to determine answers from professional financial data at evaluation time.","excerpt":"We introduce LiveSearch, a live benchmark that stores each question as executable verifier code: a typed program of calls to a professional financial data service, followed by a declarative extractor.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"The authors say the benchmark reruns its answer checks daily without a human or language model in the loop.","excerpt":"a daily job re-executes it with no human or language model in the loop.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"Under general web search, six contemporary tool-calling models reached answer accuracy between 70.9% and 76.7%.","excerpt":"Under general web search, six contemporary tool-calling models reach 70.9% to 76.7% answer accuracy.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"The authors report that 70% of served questions changed their gold value over three consecutive days.","excerpt":"Over three consecutive days, 70% of served questions change their gold value","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"The authors report that reusing an earlier snapshot fails on a majority of questions within one day, even with tolerance-aware matching.","excerpt":"even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.37166","checkedAt":"2026-10-29T23:26:35.128Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv cs.AI paper introduces LiveSearch, a live benchmark for financial search agents","summary":"In an arXiv paper, Sinuo Wang and coauthors introduce LiveSearch, a benchmark designed to test whether financial search agents can find current answers rather than repeat outdated ones. Its questions are stored as executable verifier code: instructions that call a professional financial data service and extract an answer. The question stays fixed, while the benchmark recalculates the answer from live data at evaluation time and reruns the checks daily. The authors say this makes memorized answers likely to go stale and requires models to use tools and analysis. They report that, under general web search, six contemporary tool-calling models reached 70.9% to 76.7% answer accuracy. They also report that 70% of served questions changed their gold value over three consecutive days, and that reused earlier answers failed on a majority of questions within one day, even with tolerance-aware matching. These results describe the benchmark and its reported evaluations; they do not establish how agents perform in real financial work. How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?","question":"How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?","claims":[{"text":"The authors introduce LiveSearch, a benchmark that uses executable verifier code to determine answers from professional financial data at evaluation time.","excerpt":"We introduce LiveSearch, a live benchmark that stores each question as executable verifier code: a typed program of calls to a professional financial data service, followed by a declarative extractor.","support":"source-stated"},{"text":"The authors say the benchmark reruns its answer checks daily without a human or language model in the loop.","excerpt":"a daily job re-executes it with no human or language model in the loop.","support":"source-stated"},{"text":"Under general web search, six contemporary tool-calling models reached answer accuracy between 70.9% and 76.7%.","excerpt":"Under general web search, six contemporary tool-calling models reach 70.9% to 76.7% answer accuracy.","support":"source-stated"},{"text":"The authors report that 70% of served questions changed their gold value over three consecutive days.","excerpt":"Over three consecutive days, 70% of served questions change their gold value","support":"source-stated"},{"text":"The authors report that reusing an earlier snapshot fails on a majority of questions within one day, even with tolerance-aware matching.","excerpt":"even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day.","support":"source-stated"}],"limitations":["The supplied evidence is the paper's abstract; it does not provide details about benchmark coverage or how well its tasks represent financial work outside the benchmark.","The reported accuracy results are benchmark evaluations and do not establish agents' reliability in real-world financial work."],"mediaId":"a0a0598b6fec66a250a1df3bce57e0dbb4029255248eddd10461aa84fc456cab","correctionReason":null,"id":"2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f","publicationId":"34b0bb91-8586-4a73-87f1-e61aa459bb4c","version":1,"sourceUrl":"https://arxiv.org/abs/2610.37166","sourceTitle":"LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-27","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-28T07:50:32.254Z","createdAt":"2026-10-29T21:18:44.452Z","historical":false,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"a0a0598b6fec66a250a1df3bce57e0dbb4029255248eddd10461aa84fc456cab","kind":"article","url":"https://arxiv.org/abs/2610.37166","sourceUrl":"https://arxiv.org/abs/2610.37166","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-29T21:18:02.270Z","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/2610.37166","checkedAt":"2026-10-29T23:26:35.128Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"a0a0598b6fec66a250a1df3bce57e0dbb4029255248eddd10461aa84fc456cab","confidence":0.86,"provider":"typesafe","model":"jev-1.13.0","inputHash":"33ef9a9ced81a008db3d76d6ccd83eebb6ab1032ca31377cf9333352660da404","decidedAt":"2026-10-29T21:18:03.023Z"},"provenance":{"textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observationId":"3873202f5f77caf8a9f753d36d6cb40ea799dddb3251ea91d4167f75b5f5f8a2","retrievedAt":"2026-10-29T21:18:01.384Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"1425fd36-06af-46b2-837d-cae3bdf80f9a"},"reviewer":{"model":"gpt-6-luna","runId":"a7628ef6-540e-4fa9-8aa7-5f0562b69d8d"}},"audit":{"status":"passed","reviewedAt":"2026-10-29T21:18:44.452Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Wang, Sinuo","url":null,"kind":"unknown"},{"name":"Zhao, Wentao","url":null,"kind":"unknown"},{"name":"Lu, Jiacheng","url":null,"kind":"unknown"},{"name":"Sun, Rui","url":null,"kind":"unknown"},{"name":"Luan, Beidi","url":null,"kind":"unknown"},{"name":"Shi, Qiming","url":null,"kind":"unknown"},{"name":"Li, Jing","url":null,"kind":"unknown"},{"name":"Jiang, Daxin","url":null,"kind":"unknown"},{"name":"Bai, Zuo","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Wang, Sinuo","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Wang, Sinuo","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Zhao, Wentao","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Zhao, Wentao","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Lu, Jiacheng","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Lu, Jiacheng","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Sun, Rui","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Sun, Rui","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Luan, Beidi","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Luan, Beidi","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Shi, Qiming","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shi, Qiming","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Li, Jing","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Li, Jing","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:6","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Jiang, Daxin","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Jiang, Daxin","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:7","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Bai, Zuo","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Bai, Zuo","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:8","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}}],"participationCoverage":{"total":9,"returned":9,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.37166","snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z","basis":"publisher-page-metadata","sourceLinks":47,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f/source"}},"canonicalUrl":"https://escapemeridian.com/readings/arxiv-cs-ai-paper-introduces-livesearch-a-live-benchmark-for-2814be9f","formats":{"html":"/readings/arxiv-cs-ai-paper-introduces-livesearch-a-live-benchmark-for-2814be9f","markdown":"/readings/arxiv-cs-ai-paper-introduces-livesearch-a-live-benchmark-for-2814be9f.md","json":"/readings/arxiv-cs-ai-paper-introduces-livesearch-a-live-benchmark-for-2814be9f.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.","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-benchmark","name":"Benchmark","explanation":"A defined set of tests used to compare systems. Results depend on the tasks and measures selected.","referenceUrl":"https://arxiv.org/abs/2211.09110"},{"id":"term-tool-use","name":"Tool use","explanation":"A model's ability to operate other software, such as a search engine, a calculator or a code runner, while it works.","referenceUrl":"https://docs.anthropic.com/en/docs/build-with-claude/tool-use/overview"}],"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":"2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f","revision":"34b0bb91-8586-4a73-87f1-e61aa459bb4c","revisionKind":"publication-id","title":"arXiv cs.AI paper introduces LiveSearch, a live benchmark for financial search agents","url":"https://escapemeridian.com/readings/arxiv-cs-ai-paper-introduces-livesearch-a-live-benchmark-for-2814be9f"},"source":{"type":"source","id":"https://arxiv.org/abs/2610.37166","revision":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","revisionKind":"sha256-source-text","title":"LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents","url":"https://arxiv.org/abs/2610.37166"},"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":"2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f","title":"arXiv cs.AI paper introduces LiveSearch, a live benchmark for financial search agents","argument":"In an arXiv paper, Sinuo Wang and coauthors introduce LiveSearch, a benchmark designed to test whether financial search agents can find current answers rather than repeat outdated ones. Its questions are stored as executable verifier code: instructions that call a professional financial data service and extract an answer. The question stays fixed, while the benchmark recalculates the answer from live data at evaluation time and reruns the checks daily. The authors say this makes memorized answers likely to go stale and requires models to use tools and analysis. They report that, under general web search, six contemporary tool-calling models reached 70.9% to 76.7% answer accuracy. They also report that 70% of served questions changed their gold value over three consecutive days, and that reused earlier answers failed on a majority of questions within one day, even with tolerance-aware matching. These results describe the benchmark and its reported evaluations; they do not establish how agents perform in real financial work. How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The supplied evidence is the paper's abstract; it does not provide details about benchmark coverage or how well its tasks represent financial work outside the benchmark. The reported accuracy results are benchmark evaluations and do not establish agents' reliability in real-world financial work.","speaker":{"name":"Wang, Sinuo, Zhao, Wentao, Lu, Jiacheng, Sun, Rui, Luan, Beidi, Shi, Qiming, Li, Jing, Jiang, Daxin, Bai, Zuo","roleAtSource":"Article author"},"source":{"title":"LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2610.37166","publishedAt":"2026-10-27","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors introduce LiveSearch, a benchmark that uses executable verifier code to determine answers from professional financial data at evaluation time.","excerpt":"We introduce LiveSearch, a live benchmark that stores each question as executable verifier code: a typed program of calls to a professional financial data service, followed by a declarative extractor.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"The authors say the benchmark reruns its answer checks daily without a human or language model in the loop.","excerpt":"a daily job re-executes it with no human or language model in the loop.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"Under general web search, six contemporary tool-calling models reached answer accuracy between 70.9% and 76.7%.","excerpt":"Under general web search, six contemporary tool-calling models reach 70.9% to 76.7% answer accuracy.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"The authors report that 70% of served questions changed their gold value over three consecutive days.","excerpt":"Over three consecutive days, 70% of served questions change their gold value","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"},{"text":"The authors report that reusing an earlier snapshot fails on a majority of questions within one day, even with tolerance-aware matching.","excerpt":"even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day.","url":"https://arxiv.org/abs/2610.37166","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2610.37166","checkedAt":"2026-10-29T23:26:35.128Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"arXiv cs.AI paper introduces LiveSearch, a live benchmark for financial search agents","summary":"In an arXiv paper, Sinuo Wang and coauthors introduce LiveSearch, a benchmark designed to test whether financial search agents can find current answers rather than repeat outdated ones. Its questions are stored as executable verifier code: instructions that call a professional financial data service and extract an answer. The question stays fixed, while the benchmark recalculates the answer from live data at evaluation time and reruns the checks daily. The authors say this makes memorized answers likely to go stale and requires models to use tools and analysis. They report that, under general web search, six contemporary tool-calling models reached 70.9% to 76.7% answer accuracy. They also report that 70% of served questions changed their gold value over three consecutive days, and that reused earlier answers failed on a majority of questions within one day, even with tolerance-aware matching. These results describe the benchmark and its reported evaluations; they do not establish how agents perform in real financial work. How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?","question":"How well do LiveSearch’s questions and data-service calls represent the financial tasks agents would encounter outside the benchmark?","claims":[{"text":"The authors introduce LiveSearch, a benchmark that uses executable verifier code to determine answers from professional financial data at evaluation time.","excerpt":"We introduce LiveSearch, a live benchmark that stores each question as executable verifier code: a typed program of calls to a professional financial data service, followed by a declarative extractor.","support":"source-stated"},{"text":"The authors say the benchmark reruns its answer checks daily without a human or language model in the loop.","excerpt":"a daily job re-executes it with no human or language model in the loop.","support":"source-stated"},{"text":"Under general web search, six contemporary tool-calling models reached answer accuracy between 70.9% and 76.7%.","excerpt":"Under general web search, six contemporary tool-calling models reach 70.9% to 76.7% answer accuracy.","support":"source-stated"},{"text":"The authors report that 70% of served questions changed their gold value over three consecutive days.","excerpt":"Over three consecutive days, 70% of served questions change their gold value","support":"source-stated"},{"text":"The authors report that reusing an earlier snapshot fails on a majority of questions within one day, even with tolerance-aware matching.","excerpt":"even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day.","support":"source-stated"}],"limitations":["The supplied evidence is the paper's abstract; it does not provide details about benchmark coverage or how well its tasks represent financial work outside the benchmark.","The reported accuracy results are benchmark evaluations and do not establish agents' reliability in real-world financial work."],"mediaId":"a0a0598b6fec66a250a1df3bce57e0dbb4029255248eddd10461aa84fc456cab","correctionReason":null,"id":"2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f","publicationId":"34b0bb91-8586-4a73-87f1-e61aa459bb4c","version":1,"sourceUrl":"https://arxiv.org/abs/2610.37166","sourceTitle":"LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-10-27","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-10-28T07:50:32.254Z","createdAt":"2026-10-29T21:18:44.452Z","historical":false,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"a0a0598b6fec66a250a1df3bce57e0dbb4029255248eddd10461aa84fc456cab","kind":"article","url":"https://arxiv.org/abs/2610.37166","sourceUrl":"https://arxiv.org/abs/2610.37166","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-10-29T21:18:02.270Z","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/2610.37166","checkedAt":"2026-10-29T23:26:35.128Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"a0a0598b6fec66a250a1df3bce57e0dbb4029255248eddd10461aa84fc456cab","confidence":0.86,"provider":"typesafe","model":"jev-1.13.0","inputHash":"33ef9a9ced81a008db3d76d6ccd83eebb6ab1032ca31377cf9333352660da404","decidedAt":"2026-10-29T21:18:03.023Z"},"provenance":{"textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observationId":"3873202f5f77caf8a9f753d36d6cb40ea799dddb3251ea91d4167f75b5f5f8a2","retrievedAt":"2026-10-29T21:18:01.384Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"1425fd36-06af-46b2-837d-cae3bdf80f9a"},"reviewer":{"model":"gpt-6-luna","runId":"a7628ef6-540e-4fa9-8aa7-5f0562b69d8d"}},"audit":{"status":"passed","reviewedAt":"2026-10-29T21:18:44.452Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Wang, Sinuo","url":null,"kind":"unknown"},{"name":"Zhao, Wentao","url":null,"kind":"unknown"},{"name":"Lu, Jiacheng","url":null,"kind":"unknown"},{"name":"Sun, Rui","url":null,"kind":"unknown"},{"name":"Luan, Beidi","url":null,"kind":"unknown"},{"name":"Shi, Qiming","url":null,"kind":"unknown"},{"name":"Li, Jing","url":null,"kind":"unknown"},{"name":"Jiang, Daxin","url":null,"kind":"unknown"},{"name":"Bai, Zuo","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Wang, Sinuo","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Wang, Sinuo","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Zhao, Wentao","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Zhao, Wentao","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Lu, Jiacheng","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Lu, Jiacheng","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Sun, Rui","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Sun, Rui","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Luan, Beidi","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Luan, Beidi","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Shi, Qiming","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Shi, Qiming","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:5","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Li, Jing","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Li, Jing","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:6","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Jiang, Daxin","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Jiang, Daxin","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:7","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}},{"actor":{"name":"Bai, Zuo","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Bai, Zuo","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2610.37166","locator":"head/meta[name=author|citation_author]"},"assertionId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e:participation:8","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-10-29T23:26:35.128Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","sourceUrl":"https://arxiv.org/abs/2610.37166","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z"}}],"participationCoverage":{"total":9,"returned":9,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2610.37166","snapshotId":"a9d52b032633e045b24d2f26c9b40dca7484f0aeea15769931c1cbe2ed0dc97e","contentSha256":"ffdee731318fb5ad4b0ca2ba4f2d6388a0d56d7ddf77708ed02a9e362b857a7e","textHash":"8259318c6176a3cc0ea9fae797ffdeac1ebd5720862bc0cef0f609f5fb4a5929","observedAt":"2026-10-29T23:26:35.128Z","basis":"publisher-page-metadata","sourceLinks":47,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/2814be9f705d3518c4755f2a8c4299c2338c65255410ddb9a90159f995fdad2f/source"}}}],"audits":[{"publicationId":"34b0bb91-8586-4a73-87f1-e61aa459bb4c","verdict":"passed","findings":[{"dimension":"attribution","status":"pass","explanation":"The draft correctly attributes the paper to Sinuo Wang and coauthors on arXiv. Findings are framed as the authors' reports.","excerpt":"Authors:Sinuo Wang, Wentao Zhao, Jiacheng Lu, Rui Sun, Beidi Luan, Qiming Shi, Jing Li, Daxin Jiang, Zuo Bai"},{"dimension":"summary","status":"pass","explanation":"The summary explains the benchmark and executable verifier code in plain language. It also explains why live answers counter memorization and reports the key results with appropriate caveats. Terms like gold value and tolerance-aware matching are lightly explained by context but do not block understanding. The question is open and follows from the evidence.","excerpt":"The question is frozen, while its gold answer is determined by executing the program against professional financial data at evaluation time"},{"dimension":"date","status":"pass","explanation":"The original publication date of 2026-10-27 matches the arXiv v1 submission date.","excerpt":"[Submitted on 27 Oct 2026]"},{"dimension":"support","status":"pass","explanation":"Every claim matches the abstract, including the accuracy range, the 70% gold-value change, and the failure of a reused snapshot. The limitations correctly avoid turning benchmark scores into real-world reliability. No unsupported benefits or costs are implied.","excerpt":"Over three consecutive days, 70% of served questions change their gold value, and even under tolerance-aware matching, a literally reused earlier snapshot fails on a majority of questions within a single day."},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct link to the arXiv abstract page, which is the original source.","excerpt":null}],"createdAt":"2026-10-29T21:18:44.452Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}