{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:YGUKBHHUOAYCY4J3CPV3UWJCJI","short_pith_number":"pith:YGUKBHHU","schema_version":"1.0","canonical_sha256":"c1a8a09cf470302c713b13ebba59224a2222a2002d7d0a8678e1d26017a5ff7f","source":{"kind":"arxiv","id":"2104.08704","version":2},"attestation_state":"computed","paper":{"title":"A Token-level Reference-free Hallucination Detection Benchmark for Free-form Text Generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bill Dolan, Chris Brockett, Tianyu Liu, Weizhu Chen, Yi Mao, Yizhe Zhang, Zhifang Sui","submitted_at":"2021-04-18T04:09:48Z","abstract_excerpt":"Large pretrained generative models like GPT-3 often suffer from hallucinating non-existent or incorrect content, which undermines their potential merits in real applications. Existing work usually attempts to detect these hallucinations based on a corresponding oracle reference at a sentence or document level. However ground-truth references may not be readily available for many free-form text generation applications, and sentence- or document-level detection may fail to provide the fine-grained signals that would prevent fallacious content in real time. As a first step to addressing these iss"},"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":"2104.08704","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-18T04:09:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9b2558eb484c1bc8869b6b6eeffada01f5e3ad1cdaa73c10d3991742c6821e86","abstract_canon_sha256":"2ff8d1ac463d6e284a89b054968c2764e847dde2a1ade95acb89417da642adf4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:10:50.247985Z","signature_b64":"ifNHQovk/a6Q1JfizwIDeT57pirWPMNG4/mH2/UvL5XqpQT4YROSstv1LRAygqvOvTCYHGev45dhwazRr5FbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1a8a09cf470302c713b13ebba59224a2222a2002d7d0a8678e1d26017a5ff7f","last_reissued_at":"2026-07-05T04:10:50.247482Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:10:50.247482Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Token-level Reference-free Hallucination Detection Benchmark for Free-form Text Generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bill Dolan, Chris Brockett, Tianyu Liu, Weizhu Chen, Yi Mao, Yizhe Zhang, Zhifang Sui","submitted_at":"2021-04-18T04:09:48Z","abstract_excerpt":"Large pretrained generative models like GPT-3 often suffer from hallucinating non-existent or incorrect content, which undermines their potential merits in real applications. Existing work usually attempts to detect these hallucinations based on a corresponding oracle reference at a sentence or document level. However ground-truth references may not be readily available for many free-form text generation applications, and sentence- or document-level detection may fail to provide the fine-grained signals that would prevent fallacious content in real time. As a first step to addressing these iss"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.08704","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/2104.08704/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":"2104.08704","created_at":"2026-07-05T04:10:50.247553+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.08704v2","created_at":"2026-07-05T04:10:50.247553+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.08704","created_at":"2026-07-05T04:10:50.247553+00:00"},{"alias_kind":"pith_short_12","alias_value":"YGUKBHHUOAYC","created_at":"2026-07-05T04:10:50.247553+00:00"},{"alias_kind":"pith_short_16","alias_value":"YGUKBHHUOAYCY4J3","created_at":"2026-07-05T04:10:50.247553+00:00"},{"alias_kind":"pith_short_8","alias_value":"YGUKBHHU","created_at":"2026-07-05T04:10:50.247553+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.18426","citing_title":"Retrieval Augmented Generation-based Large Language Models for Bridging Transportation Cybersecurity Legal Knowledge Gaps","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI","json":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI.json","graph_json":"https://pith.science/api/pith-number/YGUKBHHUOAYCY4J3CPV3UWJCJI/graph.json","events_json":"https://pith.science/api/pith-number/YGUKBHHUOAYCY4J3CPV3UWJCJI/events.json","paper":"https://pith.science/paper/YGUKBHHU"},"agent_actions":{"view_html":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI","download_json":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI.json","view_paper":"https://pith.science/paper/YGUKBHHU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.08704&json=true","fetch_graph":"https://pith.science/api/pith-number/YGUKBHHUOAYCY4J3CPV3UWJCJI/graph.json","fetch_events":"https://pith.science/api/pith-number/YGUKBHHUOAYCY4J3CPV3UWJCJI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI/action/storage_attestation","attest_author":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI/action/author_attestation","sign_citation":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI/action/citation_signature","submit_replication":"https://pith.science/pith/YGUKBHHUOAYCY4J3CPV3UWJCJI/action/replication_record"}},"created_at":"2026-07-05T04:10:50.247553+00:00","updated_at":"2026-07-05T04:10:50.247553+00:00"}