{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ISPILVZLXFH3RZQC6X7RGLSVAZ","short_pith_number":"pith:ISPILVZL","schema_version":"1.0","canonical_sha256":"449e85d72bb94fb8e602f5ff132e55067171aa77a74f87dfe2a8caa14ee92718","source":{"kind":"arxiv","id":"2504.08970","version":2},"attestation_state":"computed","paper":{"title":"On Large-scale Evaluation of Embedding Models for Knowledge Graph Completion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chengkai Li, Farahnaz Akrami, Nasim Shirvani-Mahdavi","submitted_at":"2025-04-11T20:49:02Z","abstract_excerpt":"Knowledge graph embedding (KGE) models are extensively studied for knowledge graph completion, yet their evaluation remains constrained by unrealistic benchmarks. Standard evaluation metrics rely on the closed-world assumption, which penalizes models for correctly predicting missing triples, contradicting the fundamental goals of link prediction. These metrics often compress accuracy assessment into a single value, obscuring models' specific strengths and weaknesses. The prevailing evaluation protocol, link prediction, operates under the unrealistic assumption that an entity's properties, for "},"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":"2504.08970","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-11T20:49:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"db0c50d2b528d889b152b4f778c1e3d1f73c59bae26713a6baa7cd4540cd760e","abstract_canon_sha256":"7bc6ea535fdb7fc0bea5dc19b6f06f6216967bfc2c91266cecd80ba22122a30c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:05.892027Z","signature_b64":"+XK2R1OJ2eLUeI1cySqzVxqESUTsTrw8X6B33bK3bOAIg4ONPA1GLOPuDNh+EWTJlpP4DShpQH0nfBRu1p//DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"449e85d72bb94fb8e602f5ff132e55067171aa77a74f87dfe2a8caa14ee92718","last_reissued_at":"2026-07-05T11:19:05.891546Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:05.891546Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Large-scale Evaluation of Embedding Models for Knowledge Graph Completion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chengkai Li, Farahnaz Akrami, Nasim Shirvani-Mahdavi","submitted_at":"2025-04-11T20:49:02Z","abstract_excerpt":"Knowledge graph embedding (KGE) models are extensively studied for knowledge graph completion, yet their evaluation remains constrained by unrealistic benchmarks. Standard evaluation metrics rely on the closed-world assumption, which penalizes models for correctly predicting missing triples, contradicting the fundamental goals of link prediction. These metrics often compress accuracy assessment into a single value, obscuring models' specific strengths and weaknesses. The prevailing evaluation protocol, link prediction, operates under the unrealistic assumption that an entity's properties, for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.08970","kind":"arxiv","version":2},"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/2504.08970/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":"2504.08970","created_at":"2026-07-05T11:19:05.891600+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.08970v2","created_at":"2026-07-05T11:19:05.891600+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.08970","created_at":"2026-07-05T11:19:05.891600+00:00"},{"alias_kind":"pith_short_12","alias_value":"ISPILVZLXFH3","created_at":"2026-07-05T11:19:05.891600+00:00"},{"alias_kind":"pith_short_16","alias_value":"ISPILVZLXFH3RZQC","created_at":"2026-07-05T11:19:05.891600+00:00"},{"alias_kind":"pith_short_8","alias_value":"ISPILVZL","created_at":"2026-07-05T11:19:05.891600+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.10971","citing_title":"Rule2Text: A Framework for Generating and Evaluating Natural Language Explanations of Knowledge Graph Rules","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ","json":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ.json","graph_json":"https://pith.science/api/pith-number/ISPILVZLXFH3RZQC6X7RGLSVAZ/graph.json","events_json":"https://pith.science/api/pith-number/ISPILVZLXFH3RZQC6X7RGLSVAZ/events.json","paper":"https://pith.science/paper/ISPILVZL"},"agent_actions":{"view_html":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ","download_json":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ.json","view_paper":"https://pith.science/paper/ISPILVZL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.08970&json=true","fetch_graph":"https://pith.science/api/pith-number/ISPILVZLXFH3RZQC6X7RGLSVAZ/graph.json","fetch_events":"https://pith.science/api/pith-number/ISPILVZLXFH3RZQC6X7RGLSVAZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ/action/storage_attestation","attest_author":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ/action/author_attestation","sign_citation":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ/action/citation_signature","submit_replication":"https://pith.science/pith/ISPILVZLXFH3RZQC6X7RGLSVAZ/action/replication_record"}},"created_at":"2026-07-05T11:19:05.891600+00:00","updated_at":"2026-07-05T11:19:05.891600+00:00"}