{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4QR5K3MFXN34BQYFRPHCIBWME7","short_pith_number":"pith:4QR5K3MF","schema_version":"1.0","canonical_sha256":"e423d56d85bb77c0c3058bce2406cc27f157c6565eea79e0341a982f1fcd2a0a","source":{"kind":"arxiv","id":"2606.02184","version":1},"attestation_state":"computed","paper":{"title":"The Ghost Couple: Correlated LLM Name Priors and Their Haunting of the Web and Academic Publishing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DL","authors_text":"Micha{\\l} Brzozowski, Neo Christopher Chung","submitted_at":"2026-06-01T12:38:56Z","abstract_excerpt":"These names do not exist. Elena Vasquez and Marcus Chen have appeared as volcano experts, astronauts, thriller protagonists, podcast hosts, and academic co-authors across hundreds of independently produced AI-generated documents, never having lived. We show that large language models do not merely default to high-probability individual names when generating fictional experts: they produce correlated character ensembles, pairs and trios whose co-occurrence rates far exceed chance and are consistent across independent generations. These priors are model-family-specific (Claude: Elena Vasquez + M"},"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":"2606.02184","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DL","submitted_at":"2026-06-01T12:38:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a56bf5d75f968a444cb90985714414174e34cc04eb22058052e6857f88fc854e","abstract_canon_sha256":"4cd83c42e5cae66d434838ffd141521ee292737c59c79842bf9f4c0d357ad96a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T03:04:52.616167Z","signature_b64":"HsWG/xmuK6G07G1AhtgK+GFEHtS/LEAA6UvBTkPPRmp3suqe0mdlfo0aeVFBe67IEnT8kRKM24j+Ytnyw0nCDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e423d56d85bb77c0c3058bce2406cc27f157c6565eea79e0341a982f1fcd2a0a","last_reissued_at":"2026-06-02T03:04:52.615716Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T03:04:52.615716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Ghost Couple: Correlated LLM Name Priors and Their Haunting of the Web and Academic Publishing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DL","authors_text":"Micha{\\l} Brzozowski, Neo Christopher Chung","submitted_at":"2026-06-01T12:38:56Z","abstract_excerpt":"These names do not exist. Elena Vasquez and Marcus Chen have appeared as volcano experts, astronauts, thriller protagonists, podcast hosts, and academic co-authors across hundreds of independently produced AI-generated documents, never having lived. We show that large language models do not merely default to high-probability individual names when generating fictional experts: they produce correlated character ensembles, pairs and trios whose co-occurrence rates far exceed chance and are consistent across independent generations. These priors are model-family-specific (Claude: Elena Vasquez + M"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02184","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/2606.02184/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":"2606.02184","created_at":"2026-06-02T03:04:52.615773+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02184v1","created_at":"2026-06-02T03:04:52.615773+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02184","created_at":"2026-06-02T03:04:52.615773+00:00"},{"alias_kind":"pith_short_12","alias_value":"4QR5K3MFXN34","created_at":"2026-06-02T03:04:52.615773+00:00"},{"alias_kind":"pith_short_16","alias_value":"4QR5K3MFXN34BQYF","created_at":"2026-06-02T03:04:52.615773+00:00"},{"alias_kind":"pith_short_8","alias_value":"4QR5K3MF","created_at":"2026-06-02T03:04:52.615773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.25902","citing_title":"Reading the Finetuning Prior: Verbatim Content Recovery via Contrastive Decoding Diffing","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7","json":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7.json","graph_json":"https://pith.science/api/pith-number/4QR5K3MFXN34BQYFRPHCIBWME7/graph.json","events_json":"https://pith.science/api/pith-number/4QR5K3MFXN34BQYFRPHCIBWME7/events.json","paper":"https://pith.science/paper/4QR5K3MF"},"agent_actions":{"view_html":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7","download_json":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7.json","view_paper":"https://pith.science/paper/4QR5K3MF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02184&json=true","fetch_graph":"https://pith.science/api/pith-number/4QR5K3MFXN34BQYFRPHCIBWME7/graph.json","fetch_events":"https://pith.science/api/pith-number/4QR5K3MFXN34BQYFRPHCIBWME7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7/action/storage_attestation","attest_author":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7/action/author_attestation","sign_citation":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7/action/citation_signature","submit_replication":"https://pith.science/pith/4QR5K3MFXN34BQYFRPHCIBWME7/action/replication_record"}},"created_at":"2026-06-02T03:04:52.615773+00:00","updated_at":"2026-06-02T03:04:52.615773+00:00"}