{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:UWZGIIU2B55RE2LMGWY7WG3ENN","short_pith_number":"pith:UWZGIIU2","schema_version":"1.0","canonical_sha256":"a5b264229a0f7b12696c35b1fb1b646b7b96d982e5412592df420fca436ed31c","source":{"kind":"arxiv","id":"2211.11875","version":1},"attestation_state":"computed","paper":{"title":"Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ananth Agarwal, Chelsea Finn, Christopher D. Manning, Eric Mitchell, Joseph J. Noh, Patrick Liu, Siyan Li, William S. Armstrong","submitted_at":"2022-11-21T21:58:30Z","abstract_excerpt":"While large pre-trained language models are powerful, their predictions often lack logical consistency across test inputs. For example, a state-of-the-art Macaw question-answering (QA) model answers 'Yes' to 'Is a sparrow a bird?' and 'Does a bird have feet?' but answers 'No' to 'Does a sparrow have feet?'. To address this failure mode, we propose a framework, Consistency Correction through Relation Detection, or ConCoRD, for boosting the consistency and accuracy of pre-trained NLP models using pre-trained natural language inference (NLI) models without fine-tuning or re-training. Given a batc"},"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":"2211.11875","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-21T21:58:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4c083e4e1c164b3e3aea3e51e50b3666fd308bd548a97eed84b7c1df890f53ec","abstract_canon_sha256":"58d5ba61971212720ae5a2d152e963d167115ae21618acc68140c7457f11457d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:18:10.382300Z","signature_b64":"QM48GSqyq9ZHsr+4v9U8IQEWQrlHBUfH6o5P7Qh+M4D//6hX9L61moifQB1M8l4R7pw7yce8Z/OCq8tdJfGaDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5b264229a0f7b12696c35b1fb1b646b7b96d982e5412592df420fca436ed31c","last_reissued_at":"2026-07-05T05:18:10.381829Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:18:10.381829Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ananth Agarwal, Chelsea Finn, Christopher D. Manning, Eric Mitchell, Joseph J. Noh, Patrick Liu, Siyan Li, William S. Armstrong","submitted_at":"2022-11-21T21:58:30Z","abstract_excerpt":"While large pre-trained language models are powerful, their predictions often lack logical consistency across test inputs. For example, a state-of-the-art Macaw question-answering (QA) model answers 'Yes' to 'Is a sparrow a bird?' and 'Does a bird have feet?' but answers 'No' to 'Does a sparrow have feet?'. To address this failure mode, we propose a framework, Consistency Correction through Relation Detection, or ConCoRD, for boosting the consistency and accuracy of pre-trained NLP models using pre-trained natural language inference (NLI) models without fine-tuning or re-training. Given a batc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.11875","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/2211.11875/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":"2211.11875","created_at":"2026-07-05T05:18:10.381888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.11875v1","created_at":"2026-07-05T05:18:10.381888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.11875","created_at":"2026-07-05T05:18:10.381888+00:00"},{"alias_kind":"pith_short_12","alias_value":"UWZGIIU2B55R","created_at":"2026-07-05T05:18:10.381888+00:00"},{"alias_kind":"pith_short_16","alias_value":"UWZGIIU2B55RE2LM","created_at":"2026-07-05T05:18:10.381888+00:00"},{"alias_kind":"pith_short_8","alias_value":"UWZGIIU2","created_at":"2026-07-05T05:18:10.381888+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23189","citing_title":"Empirical Bayes Conformal Prediction for Vision and Language Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2506.22500","citing_title":"OR-VSKC: Resolving Visual-Semantic Knowledge Conflicts in Operating Rooms with Synthetic Data-Guided Alignment","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2511.22521","citing_title":"DocVAL: Validated Chain-of-Thought Distillation for Grounded Document VQA","ref_index":21,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN","json":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN.json","graph_json":"https://pith.science/api/pith-number/UWZGIIU2B55RE2LMGWY7WG3ENN/graph.json","events_json":"https://pith.science/api/pith-number/UWZGIIU2B55RE2LMGWY7WG3ENN/events.json","paper":"https://pith.science/paper/UWZGIIU2"},"agent_actions":{"view_html":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN","download_json":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN.json","view_paper":"https://pith.science/paper/UWZGIIU2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.11875&json=true","fetch_graph":"https://pith.science/api/pith-number/UWZGIIU2B55RE2LMGWY7WG3ENN/graph.json","fetch_events":"https://pith.science/api/pith-number/UWZGIIU2B55RE2LMGWY7WG3ENN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN/action/storage_attestation","attest_author":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN/action/author_attestation","sign_citation":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN/action/citation_signature","submit_replication":"https://pith.science/pith/UWZGIIU2B55RE2LMGWY7WG3ENN/action/replication_record"}},"created_at":"2026-07-05T05:18:10.381888+00:00","updated_at":"2026-07-05T05:18:10.381888+00:00"}