{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YNWHKP66LPGT3WBGWQWHJPPDBM","short_pith_number":"pith:YNWHKP66","canonical_record":{"source":{"id":"2507.07924","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-07-10T17:06:24Z","cross_cats_sorted":[],"title_canon_sha256":"e8b6611571007920cb4190a28034540152075d0ded17cc1cc560497aa26656fd","abstract_canon_sha256":"ed08c7b4fe84f04193ed7d5c5e057be7b9e8328f9472674dac7c2ead574c32f0"},"schema_version":"1.0"},"canonical_sha256":"c36c753fde5bcd3dd826b42c74bde30b2949465324d2539d34efc7f2f4f92abb","source":{"kind":"arxiv","id":"2507.07924","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07924","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07924v1","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07924","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"pith_short_12","alias_value":"YNWHKP66LPGT","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"pith_short_16","alias_value":"YNWHKP66LPGT3WBG","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"pith_short_8","alias_value":"YNWHKP66","created_at":"2026-07-05T11:35:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YNWHKP66LPGT3WBGWQWHJPPDBM","target":"record","payload":{"canonical_record":{"source":{"id":"2507.07924","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-07-10T17:06:24Z","cross_cats_sorted":[],"title_canon_sha256":"e8b6611571007920cb4190a28034540152075d0ded17cc1cc560497aa26656fd","abstract_canon_sha256":"ed08c7b4fe84f04193ed7d5c5e057be7b9e8328f9472674dac7c2ead574c32f0"},"schema_version":"1.0"},"canonical_sha256":"c36c753fde5bcd3dd826b42c74bde30b2949465324d2539d34efc7f2f4f92abb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:07.128268Z","signature_b64":"80np7/yU/AOqOjYcFlNDBA5LZ/LHe10XCgePKRWDmcgr//BtsYG9uGkpe+ApkPndd5WonPQzLTEd7WccgLjFDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c36c753fde5bcd3dd826b42c74bde30b2949465324d2539d34efc7f2f4f92abb","last_reissued_at":"2026-07-05T11:35:07.127766Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:07.127766Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.07924","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:35:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HxsrIg4r/u/tRwtbNQSPyOSjHGFWpnTiyZb2mSIKY64ylWC6CO4dKDSRjZRXO2dNht8uQEmR9e91Y27YUfOwDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:21:33.906492Z"},"content_sha256":"5d59573d6e05c3ac1df832ff05cf2c19c68697e953659e1f074684ab20ec0727","schema_version":"1.0","event_id":"sha256:5d59573d6e05c3ac1df832ff05cf2c19c68697e953659e1f074684ab20ec0727"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YNWHKP66LPGT3WBGWQWHJPPDBM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measuring Hypothesis Testing Errors in the Evaluation of Retrieval Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Craig Macdonald, Graham McDonald, Jack McKechnie","submitted_at":"2025-07-10T17:06:24Z","abstract_excerpt":"The evaluation of Information Retrieval (IR) systems typically uses query-document pairs with corresponding human-labelled relevance assessments (qrels). These qrels are used to determine if one system is better than another based on average retrieval performance. Acquiring large volumes of human relevance assessments is expensive. Therefore, more efficient relevance assessment approaches have been proposed, necessitating comparisons between qrels to ascertain their efficacy. Discriminative power, i.e. the ability to correctly identify significant differences between systems, is important for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07924","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/2507.07924/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:35:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4Gyfsi8SkRkk8yuu/D6t6Xu1XzHXz6DR7EQX7QNrCO+AwmMzliCUOQ7kdYXDNeZXWMevyjser+JP1tedb52/AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:21:33.907045Z"},"content_sha256":"a20c0d5067fdc9e7eb7c7e35bd083874f16078e2d21fc4ffcefc9899a5193d08","schema_version":"1.0","event_id":"sha256:a20c0d5067fdc9e7eb7c7e35bd083874f16078e2d21fc4ffcefc9899a5193d08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YNWHKP66LPGT3WBGWQWHJPPDBM/bundle.json","state_url":"https://pith.science/pith/YNWHKP66LPGT3WBGWQWHJPPDBM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YNWHKP66LPGT3WBGWQWHJPPDBM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-09T00:21:33Z","links":{"resolver":"https://pith.science/pith/YNWHKP66LPGT3WBGWQWHJPPDBM","bundle":"https://pith.science/pith/YNWHKP66LPGT3WBGWQWHJPPDBM/bundle.json","state":"https://pith.science/pith/YNWHKP66LPGT3WBGWQWHJPPDBM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YNWHKP66LPGT3WBGWQWHJPPDBM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YNWHKP66LPGT3WBGWQWHJPPDBM","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":"ed08c7b4fe84f04193ed7d5c5e057be7b9e8328f9472674dac7c2ead574c32f0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-07-10T17:06:24Z","title_canon_sha256":"e8b6611571007920cb4190a28034540152075d0ded17cc1cc560497aa26656fd"},"schema_version":"1.0","source":{"id":"2507.07924","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.07924","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"arxiv_version","alias_value":"2507.07924v1","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07924","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"pith_short_12","alias_value":"YNWHKP66LPGT","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"pith_short_16","alias_value":"YNWHKP66LPGT3WBG","created_at":"2026-07-05T11:35:07Z"},{"alias_kind":"pith_short_8","alias_value":"YNWHKP66","created_at":"2026-07-05T11:35:07Z"}],"graph_snapshots":[{"event_id":"sha256:a20c0d5067fdc9e7eb7c7e35bd083874f16078e2d21fc4ffcefc9899a5193d08","target":"graph","created_at":"2026-07-05T11:35:07Z","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/2507.07924/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The evaluation of Information Retrieval (IR) systems typically uses query-document pairs with corresponding human-labelled relevance assessments (qrels). These qrels are used to determine if one system is better than another based on average retrieval performance. Acquiring large volumes of human relevance assessments is expensive. Therefore, more efficient relevance assessment approaches have been proposed, necessitating comparisons between qrels to ascertain their efficacy. Discriminative power, i.e. the ability to correctly identify significant differences between systems, is important for ","authors_text":"Craig Macdonald, Graham McDonald, Jack McKechnie","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-07-10T17:06:24Z","title":"Measuring Hypothesis Testing Errors in the Evaluation of Retrieval Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07924","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":"sha256:5d59573d6e05c3ac1df832ff05cf2c19c68697e953659e1f074684ab20ec0727","target":"record","created_at":"2026-07-05T11:35:07Z","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":"ed08c7b4fe84f04193ed7d5c5e057be7b9e8328f9472674dac7c2ead574c32f0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2025-07-10T17:06:24Z","title_canon_sha256":"e8b6611571007920cb4190a28034540152075d0ded17cc1cc560497aa26656fd"},"schema_version":"1.0","source":{"id":"2507.07924","kind":"arxiv","version":1}},"canonical_sha256":"c36c753fde5bcd3dd826b42c74bde30b2949465324d2539d34efc7f2f4f92abb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c36c753fde5bcd3dd826b42c74bde30b2949465324d2539d34efc7f2f4f92abb","first_computed_at":"2026-07-05T11:35:07.127766Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:35:07.127766Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"80np7/yU/AOqOjYcFlNDBA5LZ/LHe10XCgePKRWDmcgr//BtsYG9uGkpe+ApkPndd5WonPQzLTEd7WccgLjFDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:35:07.128268Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.07924","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5d59573d6e05c3ac1df832ff05cf2c19c68697e953659e1f074684ab20ec0727","sha256:a20c0d5067fdc9e7eb7c7e35bd083874f16078e2d21fc4ffcefc9899a5193d08"],"state_sha256":"a41c65c3cacecacea5cdc0e309ad93c806495c5a1a68ae8dcf17bc9c88406771"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xLac1FDzrx8bRsxzanNwolmL4NmFo2l98GiKnMt+FKkUXxbSWz7xTElVB1IIBP0KdScXANVyaE8eJ90dLGMDBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T00:21:33.911986Z","bundle_sha256":"b28f9ce69556c4d1a3f5bf54f7ebc22a2cc5cf5d3b4994fb2f2a5810bbdc2534"}}