{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:HANDZI3AXYWTQV2H2BZT6NLGWL","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":"3d6fc033a2b4f0c19fdb18a4e433c1779eb7f4b5a2274324d9bed3b86a1eae59","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2022-04-07T17:09:07Z","title_canon_sha256":"bf517a568a9bf63b30e2164b6a8bdb7c08fe1a1bbaed43077fc3f5fe2c3b9213"},"schema_version":"1.0","source":{"id":"2204.04211","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2204.04211","created_at":"2026-07-05T04:12:46Z"},{"alias_kind":"arxiv_version","alias_value":"2204.04211v1","created_at":"2026-07-05T04:12:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.04211","created_at":"2026-07-05T04:12:46Z"},{"alias_kind":"pith_short_12","alias_value":"HANDZI3AXYWT","created_at":"2026-07-05T04:12:46Z"},{"alias_kind":"pith_short_16","alias_value":"HANDZI3AXYWTQV2H","created_at":"2026-07-05T04:12:46Z"},{"alias_kind":"pith_short_8","alias_value":"HANDZI3A","created_at":"2026-07-05T04:12:46Z"}],"graph_snapshots":[{"event_id":"sha256:acc9ba671174d617e7925b3ddf5329d21ec835229765de5eeecd88cf7fbb2a17","target":"graph","created_at":"2026-07-05T04:12:46Z","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/2204.04211/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current test and evaluation (T&E) methods for assessing machine learning (ML) system performance often rely on incomplete metrics. Testing is additionally often siloed from the other phases of the ML system lifecycle. Research investigating cross-domain approaches to ML T&E is needed to drive the state of the art forward and to build an Artificial Intelligence (AI) engineering discipline. This paper advocates for a robust, integrated approach to testing by outlining six key questions for guiding a holistic T&E strategy.","authors_text":"Anusha Sinha, Eric Heim, Jay Palat, Nathan VanHoudnos, Rachel Dzombak, Violet Turri","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2022-04-07T17:09:07Z","title":"Measuring AI Systems Beyond Accuracy"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.04211","kind":"arxiv","version":1},"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:c191605a670428ebba5e8393fd000eff4c5382e510777084becaea8e76da6e2c","target":"record","created_at":"2026-07-05T04:12:46Z","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":"3d6fc033a2b4f0c19fdb18a4e433c1779eb7f4b5a2274324d9bed3b86a1eae59","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2022-04-07T17:09:07Z","title_canon_sha256":"bf517a568a9bf63b30e2164b6a8bdb7c08fe1a1bbaed43077fc3f5fe2c3b9213"},"schema_version":"1.0","source":{"id":"2204.04211","kind":"arxiv","version":1}},"canonical_sha256":"381a3ca360be2d385747d0733f3566b2e7eeac1ca6475e80c93289ce666b6b4f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"381a3ca360be2d385747d0733f3566b2e7eeac1ca6475e80c93289ce666b6b4f","first_computed_at":"2026-07-05T04:12:46.283927Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:12:46.283927Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NhVJxMDBddHIUJ/fB6/97/L3A5kMNEJK4O9nRTtU4e08o7o6VT0pjdx51flRR5/pQ7DdTYxd+xjygTIu6zffBw==","signature_status":"signed_v1","signed_at":"2026-07-05T04:12:46.284394Z","signed_message":"canonical_sha256_bytes"},"source_id":"2204.04211","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c191605a670428ebba5e8393fd000eff4c5382e510777084becaea8e76da6e2c","sha256:acc9ba671174d617e7925b3ddf5329d21ec835229765de5eeecd88cf7fbb2a17"],"state_sha256":"8fa3fd8efed9eef172a86914b1e5ab766e35d11e3d4bf994e3374639644cc4e9"}