{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MYUBK3FIDIEN5FF6EYOEKVW4CM","short_pith_number":"pith:MYUBK3FI","schema_version":"1.0","canonical_sha256":"6628156ca81a08de94be261c4556dc133587c38daf042b6f5a87d05c9661bece","source":{"kind":"arxiv","id":"2411.08003","version":3},"attestation_state":"computed","paper":{"title":"Can adversarial attacks by large language models be attributed?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.FL"],"primary_cat":"cs.AI","authors_text":"Andres Abeliuk, Jan Arne Telle, Manuel Cebrian","submitted_at":"2024-11-12T18:28:57Z","abstract_excerpt":"Attributing outputs from Large Language Models (LLMs) in adversarial settings-such as cyberattacks and disinformation campaigns-presents significant challenges that are likely to grow in importance. We approach this attribution problem from both a theoretical and an empirical perspective, drawing on formal language theory (identification in the limit) and data-driven analysis of the expanding LLM ecosystem. By modeling an LLM's set of possible outputs as a formal language, we analyze whether finite samples of text can uniquely pinpoint the originating model. Our results show that, under mild a"},"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":"2411.08003","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-12T18:28:57Z","cross_cats_sorted":["cs.CL","cs.CY","cs.FL"],"title_canon_sha256":"00ccfb8879e21fff0dd40b5cc91ac877e8eee57ad26a8c98e19d99cf45697aad","abstract_canon_sha256":"3b339665cf9136fe4d29d02f869d85846e2c7a0d1ff4d3063f71cd65f414dffd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:13.773512Z","signature_b64":"1gcLQX0Z4i6jNn9CUbDOdQ+PEMuSMRKupsIZXJkEg9PhtfzGO3tSa7FFh4dKn6VFMlPyAwl1QS2dz/PDwvBvAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6628156ca81a08de94be261c4556dc133587c38daf042b6f5a87d05c9661bece","last_reissued_at":"2026-07-05T11:45:13.773059Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:13.773059Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can adversarial attacks by large language models be attributed?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.FL"],"primary_cat":"cs.AI","authors_text":"Andres Abeliuk, Jan Arne Telle, Manuel Cebrian","submitted_at":"2024-11-12T18:28:57Z","abstract_excerpt":"Attributing outputs from Large Language Models (LLMs) in adversarial settings-such as cyberattacks and disinformation campaigns-presents significant challenges that are likely to grow in importance. We approach this attribution problem from both a theoretical and an empirical perspective, drawing on formal language theory (identification in the limit) and data-driven analysis of the expanding LLM ecosystem. By modeling an LLM's set of possible outputs as a formal language, we analyze whether finite samples of text can uniquely pinpoint the originating model. Our results show that, under mild a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.08003","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/2411.08003/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":"2411.08003","created_at":"2026-07-05T11:45:13.773119+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.08003v3","created_at":"2026-07-05T11:45:13.773119+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.08003","created_at":"2026-07-05T11:45:13.773119+00:00"},{"alias_kind":"pith_short_12","alias_value":"MYUBK3FIDIEN","created_at":"2026-07-05T11:45:13.773119+00:00"},{"alias_kind":"pith_short_16","alias_value":"MYUBK3FIDIEN5FF6","created_at":"2026-07-05T11:45:13.773119+00:00"},{"alias_kind":"pith_short_8","alias_value":"MYUBK3FI","created_at":"2026-07-05T11:45:13.773119+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/MYUBK3FIDIEN5FF6EYOEKVW4CM","json":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM.json","graph_json":"https://pith.science/api/pith-number/MYUBK3FIDIEN5FF6EYOEKVW4CM/graph.json","events_json":"https://pith.science/api/pith-number/MYUBK3FIDIEN5FF6EYOEKVW4CM/events.json","paper":"https://pith.science/paper/MYUBK3FI"},"agent_actions":{"view_html":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM","download_json":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM.json","view_paper":"https://pith.science/paper/MYUBK3FI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.08003&json=true","fetch_graph":"https://pith.science/api/pith-number/MYUBK3FIDIEN5FF6EYOEKVW4CM/graph.json","fetch_events":"https://pith.science/api/pith-number/MYUBK3FIDIEN5FF6EYOEKVW4CM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM/action/storage_attestation","attest_author":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM/action/author_attestation","sign_citation":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM/action/citation_signature","submit_replication":"https://pith.science/pith/MYUBK3FIDIEN5FF6EYOEKVW4CM/action/replication_record"}},"created_at":"2026-07-05T11:45:13.773119+00:00","updated_at":"2026-07-05T11:45:13.773119+00:00"}