{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MDOSSZOKDCCBWTT75OVSGK3SJE","short_pith_number":"pith:MDOSSZOK","schema_version":"1.0","canonical_sha256":"60dd2965ca18841b4e7febab232b72490bc38c6c0188f145f9fb3e17386b4346","source":{"kind":"arxiv","id":"2607.26159","version":1},"attestation_state":"computed","paper":{"title":"When benchmark inferences do not compose: Projectibility in AI evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.AI","authors_text":"Brett Reynolds","submitted_at":"2026-07-28T18:07:04Z","abstract_excerpt":"An AI benchmark result rarely reaches a consequential claim in one step. Evaluators generalize it to further cases, interpret it as evidence of capability, extrapolate it to new tasks, transport it to another system or site, and combine it with assumptions about human review and downstream consequences. Validity-centred approaches require evidence for each claim. This paper identifies a further epistemic problem: warranted links don't automatically make a warranted chain. The target of one study may not be the source of the next; system, population, outcome, or conditions may change at the int"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.26159","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T18:07:04Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"4572fbe5c956a304af3e5790f04e875d35b9c7632a9ad692e99279493260dbc9","abstract_canon_sha256":"c891d347055f661c64eb44fa261694be3c94588d8878fb97d59089d0fff39b34"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"60dd2965ca18841b4e7febab232b72490bc38c6c0188f145f9fb3e17386b4346","last_reissued_at":"2026-07-30T00:08:35.639554Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T00:08:35.639554Z"},"graph_snapshot":{"paper":{"title":"When benchmark inferences do not compose: Projectibility in AI evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.AI","authors_text":"Brett Reynolds","submitted_at":"2026-07-28T18:07:04Z","abstract_excerpt":"An AI benchmark result rarely reaches a consequential claim in one step. Evaluators generalize it to further cases, interpret it as evidence of capability, extrapolate it to new tasks, transport it to another system or site, and combine it with assumptions about human review and downstream consequences. Validity-centred approaches require evidence for each claim. This paper identifies a further epistemic problem: warranted links don't automatically make a warranted chain. The target of one study may not be the source of the next; system, population, outcome, or conditions may change at the int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26159","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.26159/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.26159","created_at":"2026-07-30T00:08:35.643296+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.26159v1","created_at":"2026-07-30T00:08:35.643296+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26159","created_at":"2026-07-30T00:08:35.643296+00:00"},{"alias_kind":"pith_short_12","alias_value":"MDOSSZOKDCCB","created_at":"2026-07-30T00:08:35.643296+00:00"},{"alias_kind":"pith_short_16","alias_value":"MDOSSZOKDCCBWTT7","created_at":"2026-07-30T00:08:35.643296+00:00"},{"alias_kind":"pith_short_8","alias_value":"MDOSSZOK","created_at":"2026-07-30T00:08:35.643296+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE","json":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE.json","graph_json":"https://pith.science/api/pith-number/MDOSSZOKDCCBWTT75OVSGK3SJE/graph.json","events_json":"https://pith.science/api/pith-number/MDOSSZOKDCCBWTT75OVSGK3SJE/events.json","paper":"https://pith.science/paper/MDOSSZOK"},"agent_actions":{"view_html":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE","download_json":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE.json","view_paper":"https://pith.science/paper/MDOSSZOK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.26159&json=true","fetch_graph":"https://pith.science/api/pith-number/MDOSSZOKDCCBWTT75OVSGK3SJE/graph.json","fetch_events":"https://pith.science/api/pith-number/MDOSSZOKDCCBWTT75OVSGK3SJE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE/action/storage_attestation","attest_author":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE/action/author_attestation","sign_citation":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE/action/citation_signature","submit_replication":"https://pith.science/pith/MDOSSZOKDCCBWTT75OVSGK3SJE/action/replication_record"}},"created_at":"2026-07-30T00:08:35.643296+00:00","updated_at":"2026-07-30T00:08:35.643296+00:00"}