{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7AA6NQCQAD5SEX76LHDXT3XUDR","short_pith_number":"pith:7AA6NQCQ","schema_version":"1.0","canonical_sha256":"f801e6c05000fb225ffe59c779eef41c7a281a5a07101d7bb77e60694323c9f4","source":{"kind":"arxiv","id":"2405.10431","version":1},"attestation_state":"computed","paper":{"title":"Thinking Fair and Slow: On the Efficacy of Structured Prompts for Debiasing Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhinav Java, Kokil Jaidka, Pragyan Banerjee, Shaz Furniturewala, Simra Shahid, Sumit Bhatia, Surgan Jandial","submitted_at":"2024-05-16T20:27:58Z","abstract_excerpt":"Existing debiasing techniques are typically training-based or require access to the model's internals and output distributions, so they are inaccessible to end-users looking to adapt LLM outputs for their particular needs. In this study, we examine whether structured prompting techniques can offer opportunities for fair text generation. We evaluate a comprehensive end-user-focused iterative framework of debiasing that applies System 2 thinking processes for prompts to induce logical, reflective, and critical text generation, with single, multi-step, instruction, and role-based variants. By sys"},"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":"2405.10431","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-16T20:27:58Z","cross_cats_sorted":[],"title_canon_sha256":"71da548851f77fe05491f40c97d6d5eaef94efd2284d5b50400364d62323ea0f","abstract_canon_sha256":"127e8ff9d1c172086f0e80432f00b172a33098a9064337e6c6e7f494f2308dba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:19:59.980567Z","signature_b64":"Um8C1XB0klwnFTsID9k4pZD4hkUNQgBa3EJ3Vur70wrJl3+BF+uTJB25skwJZND2c/MF6nnuFQ+sss2ttIWUCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f801e6c05000fb225ffe59c779eef41c7a281a5a07101d7bb77e60694323c9f4","last_reissued_at":"2026-07-05T08:19:59.980121Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:19:59.980121Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Thinking Fair and Slow: On the Efficacy of Structured Prompts for Debiasing Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhinav Java, Kokil Jaidka, Pragyan Banerjee, Shaz Furniturewala, Simra Shahid, Sumit Bhatia, Surgan Jandial","submitted_at":"2024-05-16T20:27:58Z","abstract_excerpt":"Existing debiasing techniques are typically training-based or require access to the model's internals and output distributions, so they are inaccessible to end-users looking to adapt LLM outputs for their particular needs. In this study, we examine whether structured prompting techniques can offer opportunities for fair text generation. We evaluate a comprehensive end-user-focused iterative framework of debiasing that applies System 2 thinking processes for prompts to induce logical, reflective, and critical text generation, with single, multi-step, instruction, and role-based variants. By sys"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.10431","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/2405.10431/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":"2405.10431","created_at":"2026-07-05T08:19:59.980177+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.10431v1","created_at":"2026-07-05T08:19:59.980177+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.10431","created_at":"2026-07-05T08:19:59.980177+00:00"},{"alias_kind":"pith_short_12","alias_value":"7AA6NQCQAD5S","created_at":"2026-07-05T08:19:59.980177+00:00"},{"alias_kind":"pith_short_16","alias_value":"7AA6NQCQAD5SEX76","created_at":"2026-07-05T08:19:59.980177+00:00"},{"alias_kind":"pith_short_8","alias_value":"7AA6NQCQ","created_at":"2026-07-05T08:19:59.980177+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.16756","citing_title":"Mitigating Prompt-Induced Cognitive Biases in General-Purpose AI for Software Engineering","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR","json":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR.json","graph_json":"https://pith.science/api/pith-number/7AA6NQCQAD5SEX76LHDXT3XUDR/graph.json","events_json":"https://pith.science/api/pith-number/7AA6NQCQAD5SEX76LHDXT3XUDR/events.json","paper":"https://pith.science/paper/7AA6NQCQ"},"agent_actions":{"view_html":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR","download_json":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR.json","view_paper":"https://pith.science/paper/7AA6NQCQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.10431&json=true","fetch_graph":"https://pith.science/api/pith-number/7AA6NQCQAD5SEX76LHDXT3XUDR/graph.json","fetch_events":"https://pith.science/api/pith-number/7AA6NQCQAD5SEX76LHDXT3XUDR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR/action/storage_attestation","attest_author":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR/action/author_attestation","sign_citation":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR/action/citation_signature","submit_replication":"https://pith.science/pith/7AA6NQCQAD5SEX76LHDXT3XUDR/action/replication_record"}},"created_at":"2026-07-05T08:19:59.980177+00:00","updated_at":"2026-07-05T08:19:59.980177+00:00"}