{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QAI3HPK3CTPMPX6RLPPSFAH2LP","short_pith_number":"pith:QAI3HPK3","schema_version":"1.0","canonical_sha256":"8011b3bd5b14dec7dfd15bdf2280fa5bd131e8a18045727779af263cd6e107eb","source":{"kind":"arxiv","id":"2502.05331","version":2},"attestation_state":"computed","paper":{"title":"Fine-Tuned LLMs are \"Time Capsules\" for Tracking Societal Bias Through Books","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali Emami, Nikta Gohari Sadr, Robert Morabito, Sangmitra Madhusudan, Skye Reid","submitted_at":"2025-02-07T21:13:27Z","abstract_excerpt":"Books, while often rich in cultural insights, can also mirror societal biases of their eras - biases that Large Language Models (LLMs) may learn and perpetuate during training. We introduce a novel method to trace and quantify these biases using fine-tuned LLMs. We develop BookPAGE, a corpus comprising 593 fictional books across seven decades (1950-2019), to track bias evolution. By fine-tuning LLMs on books from each decade and using targeted prompts, we examine shifts in biases related to gender, sexual orientation, race, and religion. Our findings indicate that LLMs trained on decade-specif"},"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":"2502.05331","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-07T21:13:27Z","cross_cats_sorted":[],"title_canon_sha256":"dae387592a39cd93dd45a7f12ed145269ae9f0b2297b52631f027e3c43cd5390","abstract_canon_sha256":"b6b241edb4ace1a77840562a0622383b9b4cb553d097195764343286feae0aa9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:47.765863Z","signature_b64":"C6nlE0ipPXcyALqqI3qO2Qn+FFxM72EzQWD5wsryMK0LQMDHs1Y3PYOk4xn04zfIF5TFJ7565SYLGQfU5yHABA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8011b3bd5b14dec7dfd15bdf2280fa5bd131e8a18045727779af263cd6e107eb","last_reissued_at":"2026-07-05T10:13:47.765348Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:47.765348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fine-Tuned LLMs are \"Time Capsules\" for Tracking Societal Bias Through Books","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali Emami, Nikta Gohari Sadr, Robert Morabito, Sangmitra Madhusudan, Skye Reid","submitted_at":"2025-02-07T21:13:27Z","abstract_excerpt":"Books, while often rich in cultural insights, can also mirror societal biases of their eras - biases that Large Language Models (LLMs) may learn and perpetuate during training. We introduce a novel method to trace and quantify these biases using fine-tuned LLMs. We develop BookPAGE, a corpus comprising 593 fictional books across seven decades (1950-2019), to track bias evolution. By fine-tuning LLMs on books from each decade and using targeted prompts, we examine shifts in biases related to gender, sexual orientation, race, and religion. Our findings indicate that LLMs trained on decade-specif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05331","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/2502.05331/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":"2502.05331","created_at":"2026-07-05T10:13:47.765425+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.05331v2","created_at":"2026-07-05T10:13:47.765425+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05331","created_at":"2026-07-05T10:13:47.765425+00:00"},{"alias_kind":"pith_short_12","alias_value":"QAI3HPK3CTPM","created_at":"2026-07-05T10:13:47.765425+00:00"},{"alias_kind":"pith_short_16","alias_value":"QAI3HPK3CTPMPX6R","created_at":"2026-07-05T10:13:47.765425+00:00"},{"alias_kind":"pith_short_8","alias_value":"QAI3HPK3","created_at":"2026-07-05T10:13:47.765425+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/QAI3HPK3CTPMPX6RLPPSFAH2LP","json":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP.json","graph_json":"https://pith.science/api/pith-number/QAI3HPK3CTPMPX6RLPPSFAH2LP/graph.json","events_json":"https://pith.science/api/pith-number/QAI3HPK3CTPMPX6RLPPSFAH2LP/events.json","paper":"https://pith.science/paper/QAI3HPK3"},"agent_actions":{"view_html":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP","download_json":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP.json","view_paper":"https://pith.science/paper/QAI3HPK3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.05331&json=true","fetch_graph":"https://pith.science/api/pith-number/QAI3HPK3CTPMPX6RLPPSFAH2LP/graph.json","fetch_events":"https://pith.science/api/pith-number/QAI3HPK3CTPMPX6RLPPSFAH2LP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP/action/storage_attestation","attest_author":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP/action/author_attestation","sign_citation":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP/action/citation_signature","submit_replication":"https://pith.science/pith/QAI3HPK3CTPMPX6RLPPSFAH2LP/action/replication_record"}},"created_at":"2026-07-05T10:13:47.765425+00:00","updated_at":"2026-07-05T10:13:47.765425+00:00"}