{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:IGAPJOB4SQ2WXLZXZP7TJBVSTJ","short_pith_number":"pith:IGAPJOB4","schema_version":"1.0","canonical_sha256":"4180f4b83c94356baf37cbff3486b29a6c4db4033dfe8b7851bec888f5a48c87","source":{"kind":"arxiv","id":"2210.04873","version":2},"attestation_state":"computed","paper":{"title":"CORE: A Retrieve-then-Edit Framework for Counterfactual Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bhargavi Paranjape, Hannaneh Hajishirzi, Luke Zettlemoyer, Tanay Dixit","submitted_at":"2022-10-10T17:45:38Z","abstract_excerpt":"Counterfactual data augmentation (CDA) -- i.e., adding minimally perturbed inputs during training -- helps reduce model reliance on spurious correlations and improves generalization to out-of-distribution (OOD) data. Prior work on generating counterfactuals only considered restricted classes of perturbations, limiting their effectiveness. We present COunterfactual Generation via Retrieval and Editing (CORE), a retrieval-augmented generation framework for creating diverse counterfactual perturbations for CDA. For each training example, CORE first performs a dense retrieval over a task-related u"},"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":"2210.04873","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-10T17:45:38Z","cross_cats_sorted":[],"title_canon_sha256":"d1bb2b8dbd19d383406430b5d903058f326c6b4768e5913f416d4c902e6b38e5","abstract_canon_sha256":"a5a5b3c1aebae7f708d3f5ade5d5b662fe43817c5991842339432e2101721e1e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:16.501211Z","signature_b64":"3ywwZHWATDGEIbRbG2dRsFkyLAevFYP1ph1k5CwO9cR0yo4S4Rbq040d9dd7TlfKZIdxQpBsebOjbVLF0vXrAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4180f4b83c94356baf37cbff3486b29a6c4db4033dfe8b7851bec888f5a48c87","last_reissued_at":"2026-07-05T05:12:16.500814Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:16.500814Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CORE: A Retrieve-then-Edit Framework for Counterfactual Data Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bhargavi Paranjape, Hannaneh Hajishirzi, Luke Zettlemoyer, Tanay Dixit","submitted_at":"2022-10-10T17:45:38Z","abstract_excerpt":"Counterfactual data augmentation (CDA) -- i.e., adding minimally perturbed inputs during training -- helps reduce model reliance on spurious correlations and improves generalization to out-of-distribution (OOD) data. Prior work on generating counterfactuals only considered restricted classes of perturbations, limiting their effectiveness. We present COunterfactual Generation via Retrieval and Editing (CORE), a retrieval-augmented generation framework for creating diverse counterfactual perturbations for CDA. For each training example, CORE first performs a dense retrieval over a task-related u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.04873","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/2210.04873/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":"2210.04873","created_at":"2026-07-05T05:12:16.500866+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.04873v2","created_at":"2026-07-05T05:12:16.500866+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.04873","created_at":"2026-07-05T05:12:16.500866+00:00"},{"alias_kind":"pith_short_12","alias_value":"IGAPJOB4SQ2W","created_at":"2026-07-05T05:12:16.500866+00:00"},{"alias_kind":"pith_short_16","alias_value":"IGAPJOB4SQ2WXLZX","created_at":"2026-07-05T05:12:16.500866+00:00"},{"alias_kind":"pith_short_8","alias_value":"IGAPJOB4","created_at":"2026-07-05T05:12:16.500866+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27249","citing_title":"Gumbel Machine: Counterfactual Student Writing Generation via Gumbel Noise Steering","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ","json":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ.json","graph_json":"https://pith.science/api/pith-number/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/graph.json","events_json":"https://pith.science/api/pith-number/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/events.json","paper":"https://pith.science/paper/IGAPJOB4"},"agent_actions":{"view_html":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ","download_json":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ.json","view_paper":"https://pith.science/paper/IGAPJOB4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.04873&json=true","fetch_graph":"https://pith.science/api/pith-number/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/graph.json","fetch_events":"https://pith.science/api/pith-number/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/action/storage_attestation","attest_author":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/action/author_attestation","sign_citation":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/action/citation_signature","submit_replication":"https://pith.science/pith/IGAPJOB4SQ2WXLZXZP7TJBVSTJ/action/replication_record"}},"created_at":"2026-07-05T05:12:16.500866+00:00","updated_at":"2026-07-05T05:12:16.500866+00:00"}