{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LKNC7YAGACFBDRBFLGPSWJS64V","short_pith_number":"pith:LKNC7YAG","schema_version":"1.0","canonical_sha256":"5a9a2fe006008a11c425599f2b265ee5763ba1b01fbd60b47913da6b68f009eb","source":{"kind":"arxiv","id":"2507.22915","version":1},"attestation_state":"computed","paper":{"title":"Theoretical Foundations and Mitigation of Hallucination in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Esmail Gumaan","submitted_at":"2025-07-20T15:22:34Z","abstract_excerpt":"Hallucination in Large Language Models (LLMs) refers to the generation of content that is not faithful to the input or the real-world facts. This paper provides a rigorous treatment of hallucination in LLMs, including formal definitions and theoretical analyses. We distinguish between intrinsic and extrinsic hallucinations, and define a \\textit{hallucination risk} for models. We derive bounds on this risk using learning-theoretic frameworks (PAC-Bayes and Rademacher complexity). We then survey detection strategies for hallucinations, such as token-level uncertainty estimation, confidence calib"},"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":"2507.22915","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-20T15:22:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"be80d61003362486bcbd5b5f648c84e04869fa24d07e83a48429cf8a0a64671e","abstract_canon_sha256":"eaca9df68b7a641877d1737a4c02df5560eb966e080001743054aefc82841a1d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:46:01.575977Z","signature_b64":"+Ou+qk5KK4AVVN2vU9bwmptMhgMTXKnWdPQW7hrZxJwv1B288Bq9g2jOKHhONRcIY6dLHjJQ/PSuMNizv6VPCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a9a2fe006008a11c425599f2b265ee5763ba1b01fbd60b47913da6b68f009eb","last_reissued_at":"2026-07-05T11:46:01.575384Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:46:01.575384Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Theoretical Foundations and Mitigation of Hallucination in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Esmail Gumaan","submitted_at":"2025-07-20T15:22:34Z","abstract_excerpt":"Hallucination in Large Language Models (LLMs) refers to the generation of content that is not faithful to the input or the real-world facts. This paper provides a rigorous treatment of hallucination in LLMs, including formal definitions and theoretical analyses. We distinguish between intrinsic and extrinsic hallucinations, and define a \\textit{hallucination risk} for models. We derive bounds on this risk using learning-theoretic frameworks (PAC-Bayes and Rademacher complexity). We then survey detection strategies for hallucinations, such as token-level uncertainty estimation, confidence calib"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.22915","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/2507.22915/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":"2507.22915","created_at":"2026-07-05T11:46:01.575462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.22915v1","created_at":"2026-07-05T11:46:01.575462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.22915","created_at":"2026-07-05T11:46:01.575462+00:00"},{"alias_kind":"pith_short_12","alias_value":"LKNC7YAGACFB","created_at":"2026-07-05T11:46:01.575462+00:00"},{"alias_kind":"pith_short_16","alias_value":"LKNC7YAGACFBDRBF","created_at":"2026-07-05T11:46:01.575462+00:00"},{"alias_kind":"pith_short_8","alias_value":"LKNC7YAG","created_at":"2026-07-05T11:46:01.575462+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.20591","citing_title":"Do No Harm? Hallucination and Actor-Level Abuse in Web-Deployed Medical Large Language Models","ref_index":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V","json":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V.json","graph_json":"https://pith.science/api/pith-number/LKNC7YAGACFBDRBFLGPSWJS64V/graph.json","events_json":"https://pith.science/api/pith-number/LKNC7YAGACFBDRBFLGPSWJS64V/events.json","paper":"https://pith.science/paper/LKNC7YAG"},"agent_actions":{"view_html":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V","download_json":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V.json","view_paper":"https://pith.science/paper/LKNC7YAG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.22915&json=true","fetch_graph":"https://pith.science/api/pith-number/LKNC7YAGACFBDRBFLGPSWJS64V/graph.json","fetch_events":"https://pith.science/api/pith-number/LKNC7YAGACFBDRBFLGPSWJS64V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V/action/storage_attestation","attest_author":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V/action/author_attestation","sign_citation":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V/action/citation_signature","submit_replication":"https://pith.science/pith/LKNC7YAGACFBDRBFLGPSWJS64V/action/replication_record"}},"created_at":"2026-07-05T11:46:01.575462+00:00","updated_at":"2026-07-05T11:46:01.575462+00:00"}