{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4QP3S4FCIY6762NJMIE7BQUBTC","short_pith_number":"pith:4QP3S4FC","schema_version":"1.0","canonical_sha256":"e41fb970a2463dff69a96209f0c281989e7acc597fd2422299d1b206bccd3cd3","source":{"kind":"arxiv","id":"2407.06946","version":2},"attestation_state":"computed","paper":{"title":"Self-Recognition in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caglar Gulcehre, Giuseppe Russo, Robert West, Tim R. Davidson, Veniamin Veselovsky, Viacheslav Surkov","submitted_at":"2024-07-09T15:23:28Z","abstract_excerpt":"A rapidly growing number of applications rely on a small set of closed-source language models (LMs). This dependency might introduce novel security risks if LMs develop self-recognition capabilities. Inspired by human identity verification methods, we propose a novel approach for assessing self-recognition in LMs using model-generated \"security questions\". Our test can be externally administered to monitor frontier models as it does not require access to internal model parameters or output probabilities. We use our test to examine self-recognition in ten of the most capable open- and closed-so"},"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":"2407.06946","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-07-09T15:23:28Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"ee9aa08cfdd0098972860c86e869db975e9d49c0e4a847ee2af27beeb9a19f48","abstract_canon_sha256":"08ab510156c63f044fe1bebac1bdeae072804d8f918bd7644a3496aee508b676"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:28.156842Z","signature_b64":"zyQlDjHIZiutTrDc/EyPybo/+8gK7hSgA5NDnnkJRPOD+/OH9fbbkNfXO40Af6bBgBSOL5lYCFKVUQrZyQt7DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e41fb970a2463dff69a96209f0c281989e7acc597fd2422299d1b206bccd3cd3","last_reissued_at":"2026-07-05T09:18:28.156360Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:28.156360Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Recognition in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caglar Gulcehre, Giuseppe Russo, Robert West, Tim R. Davidson, Veniamin Veselovsky, Viacheslav Surkov","submitted_at":"2024-07-09T15:23:28Z","abstract_excerpt":"A rapidly growing number of applications rely on a small set of closed-source language models (LMs). This dependency might introduce novel security risks if LMs develop self-recognition capabilities. Inspired by human identity verification methods, we propose a novel approach for assessing self-recognition in LMs using model-generated \"security questions\". Our test can be externally administered to monitor frontier models as it does not require access to internal model parameters or output probabilities. We use our test to examine self-recognition in ten of the most capable open- and closed-so"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.06946","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/2407.06946/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":"2407.06946","created_at":"2026-07-05T09:18:28.156416+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.06946v2","created_at":"2026-07-05T09:18:28.156416+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.06946","created_at":"2026-07-05T09:18:28.156416+00:00"},{"alias_kind":"pith_short_12","alias_value":"4QP3S4FCIY67","created_at":"2026-07-05T09:18:28.156416+00:00"},{"alias_kind":"pith_short_16","alias_value":"4QP3S4FCIY6762NJ","created_at":"2026-07-05T09:18:28.156416+00:00"},{"alias_kind":"pith_short_8","alias_value":"4QP3S4FC","created_at":"2026-07-05T09:18:28.156416+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00545","citing_title":"The Assistant as a Privileged Persona: A canonical reference in cross-persona self-recognition","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC","json":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC.json","graph_json":"https://pith.science/api/pith-number/4QP3S4FCIY6762NJMIE7BQUBTC/graph.json","events_json":"https://pith.science/api/pith-number/4QP3S4FCIY6762NJMIE7BQUBTC/events.json","paper":"https://pith.science/paper/4QP3S4FC"},"agent_actions":{"view_html":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC","download_json":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC.json","view_paper":"https://pith.science/paper/4QP3S4FC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.06946&json=true","fetch_graph":"https://pith.science/api/pith-number/4QP3S4FCIY6762NJMIE7BQUBTC/graph.json","fetch_events":"https://pith.science/api/pith-number/4QP3S4FCIY6762NJMIE7BQUBTC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC/action/storage_attestation","attest_author":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC/action/author_attestation","sign_citation":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC/action/citation_signature","submit_replication":"https://pith.science/pith/4QP3S4FCIY6762NJMIE7BQUBTC/action/replication_record"}},"created_at":"2026-07-05T09:18:28.156416+00:00","updated_at":"2026-07-05T09:18:28.156416+00:00"}