{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QJYJUSFEW7YOEC4LUGH74NB3OW","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e2ba94c694dd1223b5b01ad4b5cb4a16df7963141d38d6dfaef7d8259e6eccb3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-03-13T20:44:15Z","title_canon_sha256":"b3c97df853bc8f9c71249939ca57781016dd5abfff3f757c7930c171969ae68d"},"schema_version":"1.0","source":{"id":"2603.15676","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.15676","created_at":"2026-05-22T01:04:00Z"},{"alias_kind":"arxiv_version","alias_value":"2603.15676v2","created_at":"2026-05-22T01:04:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.15676","created_at":"2026-05-22T01:04:00Z"},{"alias_kind":"pith_short_12","alias_value":"QJYJUSFEW7YO","created_at":"2026-05-22T01:04:00Z"},{"alias_kind":"pith_short_16","alias_value":"QJYJUSFEW7YOEC4L","created_at":"2026-05-22T01:04:00Z"},{"alias_kind":"pith_short_8","alias_value":"QJYJUSFE","created_at":"2026-05-22T01:04:00Z"}],"graph_snapshots":[{"event_id":"sha256:6ec39b47968c3e9b28ee27d96fe52fc6dea96ffe7c0620eeac3e2e932b8d2e77","target":"graph","created_at":"2026-05-22T01:04:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2603.15676/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM applications are AI systems whose nondeterministic outputs and evolving model behavior make traditional testing insufficient for release governance. We present an automated self-testing framework that introduces quality gates with evidence-based release decisions (PROMOTE/HOLD/ROLLBACK) across five empirically grounded dimensions: task success rate, research context preservation, P95 latency, safety pass rate, and evidence coverage. We evaluate the framework through a longitudinal case study of an internally deployed multi-agent conversational AI system with specific marketing capabilities","authors_text":"Alexandre Cristov\\~ao Maiorano","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-03-13T20:44:15Z","title":"Automated Self-Testing as a Quality Gate: Evidence-Driven Release Management for LLM Applications"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.15676","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7ac5dc8e93edb0df0ad6e23922fda6bcde63cfc0fefdb751cb40f68f88981021","target":"record","created_at":"2026-05-22T01:04:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e2ba94c694dd1223b5b01ad4b5cb4a16df7963141d38d6dfaef7d8259e6eccb3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-03-13T20:44:15Z","title_canon_sha256":"b3c97df853bc8f9c71249939ca57781016dd5abfff3f757c7930c171969ae68d"},"schema_version":"1.0","source":{"id":"2603.15676","kind":"arxiv","version":2}},"canonical_sha256":"82709a48a4b7f0e20b8ba18ffe343b7584c3e455409404cf3fe32335669e67cf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"82709a48a4b7f0e20b8ba18ffe343b7584c3e455409404cf3fe32335669e67cf","first_computed_at":"2026-05-22T01:04:00.270824Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:04:00.270824Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3+xSqDGCM2eK4yR+7mP0pPi5hI1gml6GVqImdnV+nFFm+He9C/m5DEgq0UksV0ZGUasPwwGgEarqqSu+NoVbCw==","signature_status":"signed_v1","signed_at":"2026-05-22T01:04:00.271462Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.15676","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7ac5dc8e93edb0df0ad6e23922fda6bcde63cfc0fefdb751cb40f68f88981021","sha256:6ec39b47968c3e9b28ee27d96fe52fc6dea96ffe7c0620eeac3e2e932b8d2e77"],"state_sha256":"9ebb9ba8fb3d3ac0ab5e3aa829b2aece33ef5c9067855f0388f69e96b88c6c3b"}