{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:4XV5LCE7PEXM34RWFTNTXVM6XI","short_pith_number":"pith:4XV5LCE7","schema_version":"1.0","canonical_sha256":"e5ebd5889f792ecdf2362cdb3bd59eba38504f25f8d594ae889ff2b88b49716a","source":{"kind":"arxiv","id":"1910.08684","version":3},"attestation_state":"computed","paper":{"title":"Sticking to the Facts: Confident Decoding for Faithful Data-to-Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ankur P. Parikh, Ran Tian, Shashi Narayan, Thibault Sellam","submitted_at":"2019-10-19T03:00:46Z","abstract_excerpt":"We address the issue of hallucination in data-to-text generation, i.e., reducing the generation of text that is unsupported by the source. We conjecture that hallucination can be caused by an encoder-decoder model generating content phrases without attending to the source; so we propose a confidence score to ensure that the model attends to the source whenever necessary, as well as a variational Bayes training framework that can learn the score from data. Experiments on the WikiBio (Lebretet al., 2016) dataset show that our approach is more faithful to the source than existing state-of-the-art"},"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":"1910.08684","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-10-19T03:00:46Z","cross_cats_sorted":[],"title_canon_sha256":"9de8095da2285c581b954fda4ab7dd9a899cfeaa11545a7477ec1b0ab58a46a2","abstract_canon_sha256":"0f31fc54faadc6b2e056e3255811853863aea66956c446a51b218205a7a5ff08"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:48:06.731870Z","signature_b64":"RCurb55obyGYNZ5byd1xRv6Rb0FNY43M06mQ+8OoPkKP2UIrW/tc6ELWmLrINcKvNTEh+asbgETVCoNSaT6tCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5ebd5889f792ecdf2362cdb3bd59eba38504f25f8d594ae889ff2b88b49716a","last_reissued_at":"2026-07-05T01:48:06.731432Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:48:06.731432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sticking to the Facts: Confident Decoding for Faithful Data-to-Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ankur P. Parikh, Ran Tian, Shashi Narayan, Thibault Sellam","submitted_at":"2019-10-19T03:00:46Z","abstract_excerpt":"We address the issue of hallucination in data-to-text generation, i.e., reducing the generation of text that is unsupported by the source. We conjecture that hallucination can be caused by an encoder-decoder model generating content phrases without attending to the source; so we propose a confidence score to ensure that the model attends to the source whenever necessary, as well as a variational Bayes training framework that can learn the score from data. Experiments on the WikiBio (Lebretet al., 2016) dataset show that our approach is more faithful to the source than existing state-of-the-art"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.08684","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/1910.08684/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":"1910.08684","created_at":"2026-07-05T01:48:06.731489+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.08684v3","created_at":"2026-07-05T01:48:06.731489+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.08684","created_at":"2026-07-05T01:48:06.731489+00:00"},{"alias_kind":"pith_short_12","alias_value":"4XV5LCE7PEXM","created_at":"2026-07-05T01:48:06.731489+00:00"},{"alias_kind":"pith_short_16","alias_value":"4XV5LCE7PEXM34RW","created_at":"2026-07-05T01:48:06.731489+00:00"},{"alias_kind":"pith_short_8","alias_value":"4XV5LCE7","created_at":"2026-07-05T01:48:06.731489+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20369","citing_title":"CATCH-ME if you RAG: a dataset of Contextually Annotated multi-Turn Counterspeech against Hate and Misinformation Exchanges","ref_index":256,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22435","citing_title":"Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation","ref_index":246,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI","json":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI.json","graph_json":"https://pith.science/api/pith-number/4XV5LCE7PEXM34RWFTNTXVM6XI/graph.json","events_json":"https://pith.science/api/pith-number/4XV5LCE7PEXM34RWFTNTXVM6XI/events.json","paper":"https://pith.science/paper/4XV5LCE7"},"agent_actions":{"view_html":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI","download_json":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI.json","view_paper":"https://pith.science/paper/4XV5LCE7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.08684&json=true","fetch_graph":"https://pith.science/api/pith-number/4XV5LCE7PEXM34RWFTNTXVM6XI/graph.json","fetch_events":"https://pith.science/api/pith-number/4XV5LCE7PEXM34RWFTNTXVM6XI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI/action/storage_attestation","attest_author":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI/action/author_attestation","sign_citation":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI/action/citation_signature","submit_replication":"https://pith.science/pith/4XV5LCE7PEXM34RWFTNTXVM6XI/action/replication_record"}},"created_at":"2026-07-05T01:48:06.731489+00:00","updated_at":"2026-07-05T01:48:06.731489+00:00"}