{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NMIHJIL3I6FWNALDFTKPKQWWTP","short_pith_number":"pith:NMIHJIL3","schema_version":"1.0","canonical_sha256":"6b1074a17b478b6681632cd4f542d69be7f75f1fdf5dd77504de9029c396f9d2","source":{"kind":"arxiv","id":"2506.22271","version":3},"attestation_state":"computed","paper":{"title":"On the Theoretical Limitations of Embedding-based Link Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Emile van Krieken, Luciano Serafini, Samy Badreddine","submitted_at":"2025-06-27T14:41:22Z","abstract_excerpt":"Neural networks often map low-dimensional embeddings to high-dimensional output spaces. Usually, the output layer is linear, which can create a \"rank bottleneck\" that limits the functions a model can represent. Such bottlenecks are ubiquitous in link prediction models, such as knowledge graph embeddings (KGEs), as the output space of entities can be orders of magnitude larger than the embedding dimension. We investigate how rank bottlenecks limit model expressivity for fitting the training data. While previous work focused on sufficient bounds on the embedding dimension required for specific K"},"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":"2506.22271","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-27T14:41:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c7252be8fd8adcb44a335c38cccf2744aa0b50d0709ee018b286e5042d1776ef","abstract_canon_sha256":"8aba13e1452a8c27bf29eef05cf9c655f12234154acd40a6e39ede276b4c30ad"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:15.001846Z","signature_b64":"JpfQO8GCDtr2uOxB5iMcVdBPZgdTFT3+abdEeYATximMwP9UkbcHcMS9iqpY0G4yE68WsS4iPKTtAwG30qFEAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6b1074a17b478b6681632cd4f542d69be7f75f1fdf5dd77504de9029c396f9d2","last_reissued_at":"2026-06-02T01:03:15.000405Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:15.000405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Theoretical Limitations of Embedding-based Link Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Emile van Krieken, Luciano Serafini, Samy Badreddine","submitted_at":"2025-06-27T14:41:22Z","abstract_excerpt":"Neural networks often map low-dimensional embeddings to high-dimensional output spaces. Usually, the output layer is linear, which can create a \"rank bottleneck\" that limits the functions a model can represent. Such bottlenecks are ubiquitous in link prediction models, such as knowledge graph embeddings (KGEs), as the output space of entities can be orders of magnitude larger than the embedding dimension. We investigate how rank bottlenecks limit model expressivity for fitting the training data. While previous work focused on sufficient bounds on the embedding dimension required for specific K"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.22271","kind":"arxiv","version":3},"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/2506.22271/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":"2506.22271","created_at":"2026-06-02T01:03:15.000465+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.22271v3","created_at":"2026-06-02T01:03:15.000465+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.22271","created_at":"2026-06-02T01:03:15.000465+00:00"},{"alias_kind":"pith_short_12","alias_value":"NMIHJIL3I6FW","created_at":"2026-06-02T01:03:15.000465+00:00"},{"alias_kind":"pith_short_16","alias_value":"NMIHJIL3I6FWNALD","created_at":"2026-06-02T01:03:15.000465+00:00"},{"alias_kind":"pith_short_8","alias_value":"NMIHJIL3","created_at":"2026-06-02T01:03:15.000465+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/NMIHJIL3I6FWNALDFTKPKQWWTP","json":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP.json","graph_json":"https://pith.science/api/pith-number/NMIHJIL3I6FWNALDFTKPKQWWTP/graph.json","events_json":"https://pith.science/api/pith-number/NMIHJIL3I6FWNALDFTKPKQWWTP/events.json","paper":"https://pith.science/paper/NMIHJIL3"},"agent_actions":{"view_html":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP","download_json":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP.json","view_paper":"https://pith.science/paper/NMIHJIL3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.22271&json=true","fetch_graph":"https://pith.science/api/pith-number/NMIHJIL3I6FWNALDFTKPKQWWTP/graph.json","fetch_events":"https://pith.science/api/pith-number/NMIHJIL3I6FWNALDFTKPKQWWTP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP/action/storage_attestation","attest_author":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP/action/author_attestation","sign_citation":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP/action/citation_signature","submit_replication":"https://pith.science/pith/NMIHJIL3I6FWNALDFTKPKQWWTP/action/replication_record"}},"created_at":"2026-06-02T01:03:15.000465+00:00","updated_at":"2026-06-02T01:03:15.000465+00:00"}