{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WT5KYXBFXNFNS7LVACZJXQBDVZ","short_pith_number":"pith:WT5KYXBF","schema_version":"1.0","canonical_sha256":"b4faac5c25bb4ad97d7500b29bc023ae70821fb136403bfca767bb37cefd5dcb","source":{"kind":"arxiv","id":"2408.13464","version":2},"attestation_state":"computed","paper":{"title":"Uncovering Biases with Reflective Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Edward Y. Chang","submitted_at":"2024-08-24T04:48:32Z","abstract_excerpt":"Biases and errors in human-labeled data present significant challenges for machine learning, especially in supervised learning reliant on potentially flawed ground truth data. These flaws, including diagnostic errors and societal biases, risk being propagated and amplified through models trained using maximum likelihood estimation. We present the Reflective LLM Dialogue Framework RLDF, which leverages structured adversarial dialogues between multiple instances of a single LLM or different LLMs to uncover diverse perspectives and correct inconsistencies. By conditioning LLMs to adopt opposing s"},"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":"2408.13464","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-08-24T04:48:32Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"d649da36b5fa25e9653b2d48daf0417d53d8509dd50d5f71b05936dd7f6df016","abstract_canon_sha256":"4711b00e0c81391c54f762954265c6e1678c58c32817af0d8d7724df48934463"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:00.811941Z","signature_b64":"Jdfd/jbEXPdpFl5nyaPm75hj3CueQcWrhKpAhXLJixTOklebvtIRGi4UgMZnOdUEQG0XvPlBb+DsXxC4Ul+sBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4faac5c25bb4ad97d7500b29bc023ae70821fb136403bfca767bb37cefd5dcb","last_reissued_at":"2026-07-05T09:25:00.811322Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:00.811322Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Uncovering Biases with Reflective Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Edward Y. Chang","submitted_at":"2024-08-24T04:48:32Z","abstract_excerpt":"Biases and errors in human-labeled data present significant challenges for machine learning, especially in supervised learning reliant on potentially flawed ground truth data. These flaws, including diagnostic errors and societal biases, risk being propagated and amplified through models trained using maximum likelihood estimation. We present the Reflective LLM Dialogue Framework RLDF, which leverages structured adversarial dialogues between multiple instances of a single LLM or different LLMs to uncover diverse perspectives and correct inconsistencies. By conditioning LLMs to adopt opposing s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.13464","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/2408.13464/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":"2408.13464","created_at":"2026-07-05T09:25:00.811407+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.13464v2","created_at":"2026-07-05T09:25:00.811407+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.13464","created_at":"2026-07-05T09:25:00.811407+00:00"},{"alias_kind":"pith_short_12","alias_value":"WT5KYXBFXNFN","created_at":"2026-07-05T09:25:00.811407+00:00"},{"alias_kind":"pith_short_16","alias_value":"WT5KYXBFXNFNS7LV","created_at":"2026-07-05T09:25:00.811407+00:00"},{"alias_kind":"pith_short_8","alias_value":"WT5KYXBF","created_at":"2026-07-05T09:25:00.811407+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/WT5KYXBFXNFNS7LVACZJXQBDVZ","json":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ.json","graph_json":"https://pith.science/api/pith-number/WT5KYXBFXNFNS7LVACZJXQBDVZ/graph.json","events_json":"https://pith.science/api/pith-number/WT5KYXBFXNFNS7LVACZJXQBDVZ/events.json","paper":"https://pith.science/paper/WT5KYXBF"},"agent_actions":{"view_html":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ","download_json":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ.json","view_paper":"https://pith.science/paper/WT5KYXBF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.13464&json=true","fetch_graph":"https://pith.science/api/pith-number/WT5KYXBFXNFNS7LVACZJXQBDVZ/graph.json","fetch_events":"https://pith.science/api/pith-number/WT5KYXBFXNFNS7LVACZJXQBDVZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ/action/storage_attestation","attest_author":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ/action/author_attestation","sign_citation":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ/action/citation_signature","submit_replication":"https://pith.science/pith/WT5KYXBFXNFNS7LVACZJXQBDVZ/action/replication_record"}},"created_at":"2026-07-05T09:25:00.811407+00:00","updated_at":"2026-07-05T09:25:00.811407+00:00"}