{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BRHKAOF67WNPC575N5EJ3LIVKN","short_pith_number":"pith:BRHKAOF6","schema_version":"1.0","canonical_sha256":"0c4ea038befd9af177fd6f489dad15536e49e086a2d5f834a124d71c8b933d15","source":{"kind":"arxiv","id":"2306.00374","version":1},"attestation_state":"computed","paper":{"title":"CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kahini Wadhawan, Rahul Madhavan, Rishabh Garg, Sameep Mehta","submitted_at":"2023-06-01T06:13:51Z","abstract_excerpt":"We propose a method to control the attributes of Language Models (LMs) for the text generation task using Causal Average Treatment Effect (ATE) scores and counterfactual augmentation. We explore this method, in the context of LM detoxification, and propose the Causally Fair Language (CFL) architecture for detoxifying pre-trained LMs in a plug-and-play manner. Our architecture is based on a Structural Causal Model (SCM) that is mathematically transparent and computationally efficient as compared with many existing detoxification techniques. We also propose several new metrics that aim to better"},"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":"2306.00374","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T06:13:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"03eca651abbe213eea5ff3ac6610da79070d0fab0455cf0e3f1b68aed97dd346","abstract_canon_sha256":"3d0bf2ef8a17bebf0982f8fa6697a1e1a0ba5bd9df23aeb639972fdad892591c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:56:34.096931Z","signature_b64":"6KCJMpKWVBOUbAiuyT66ogKo8GUrFiZWk6SWNhSfsJAeluW9htrrFiT1yJvCXd6PIkxC4VbkHlfJkTAiD+eZBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c4ea038befd9af177fd6f489dad15536e49e086a2d5f834a124d71c8b933d15","last_reissued_at":"2026-07-05T06:56:34.096459Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:56:34.096459Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kahini Wadhawan, Rahul Madhavan, Rishabh Garg, Sameep Mehta","submitted_at":"2023-06-01T06:13:51Z","abstract_excerpt":"We propose a method to control the attributes of Language Models (LMs) for the text generation task using Causal Average Treatment Effect (ATE) scores and counterfactual augmentation. We explore this method, in the context of LM detoxification, and propose the Causally Fair Language (CFL) architecture for detoxifying pre-trained LMs in a plug-and-play manner. Our architecture is based on a Structural Causal Model (SCM) that is mathematically transparent and computationally efficient as compared with many existing detoxification techniques. We also propose several new metrics that aim to better"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.00374","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/2306.00374/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":"2306.00374","created_at":"2026-07-05T06:56:34.096517+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.00374v1","created_at":"2026-07-05T06:56:34.096517+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.00374","created_at":"2026-07-05T06:56:34.096517+00:00"},{"alias_kind":"pith_short_12","alias_value":"BRHKAOF67WNP","created_at":"2026-07-05T06:56:34.096517+00:00"},{"alias_kind":"pith_short_16","alias_value":"BRHKAOF67WNPC575","created_at":"2026-07-05T06:56:34.096517+00:00"},{"alias_kind":"pith_short_8","alias_value":"BRHKAOF6","created_at":"2026-07-05T06:56:34.096517+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN","json":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN.json","graph_json":"https://pith.science/api/pith-number/BRHKAOF67WNPC575N5EJ3LIVKN/graph.json","events_json":"https://pith.science/api/pith-number/BRHKAOF67WNPC575N5EJ3LIVKN/events.json","paper":"https://pith.science/paper/BRHKAOF6"},"agent_actions":{"view_html":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN","download_json":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN.json","view_paper":"https://pith.science/paper/BRHKAOF6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.00374&json=true","fetch_graph":"https://pith.science/api/pith-number/BRHKAOF67WNPC575N5EJ3LIVKN/graph.json","fetch_events":"https://pith.science/api/pith-number/BRHKAOF67WNPC575N5EJ3LIVKN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN/action/storage_attestation","attest_author":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN/action/author_attestation","sign_citation":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN/action/citation_signature","submit_replication":"https://pith.science/pith/BRHKAOF67WNPC575N5EJ3LIVKN/action/replication_record"}},"created_at":"2026-07-05T06:56:34.096517+00:00","updated_at":"2026-07-05T06:56:34.096517+00:00"}