{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7SXAISIQ5TXSFBUO272BAHHC76","short_pith_number":"pith:7SXAISIQ","schema_version":"1.0","canonical_sha256":"fcae044910ecef22868ed7f4101ce2ff85347796dedebaed0beea5215b72233c","source":{"kind":"arxiv","id":"2412.11414","version":1},"attestation_state":"computed","paper":{"title":"Biased or Flawed? Mitigating Stereotypes in Generative Language Models by Addressing Task-Specific Flaws","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Akshita Jha, Chandan K. Reddy, Sanchit Kabra","submitted_at":"2024-12-16T03:29:08Z","abstract_excerpt":"Recent studies have shown that generative language models often reflect and amplify societal biases in their outputs. However, these studies frequently conflate observed biases with other task-specific shortcomings, such as comprehension failure. For example, when a model misinterprets a text and produces a response that reinforces a stereotype, it becomes difficult to determine whether the issue arises from inherent bias or from a misunderstanding of the given content. In this paper, we conduct a multi-faceted evaluation that distinctly disentangles bias from flaws within the reading comprehe"},"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":"2412.11414","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-16T03:29:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3ab609ebca6e6d689a28151f75e412a472f39347097092c4f73b5cd751a3a202","abstract_canon_sha256":"b2922c7bc475e4290784c98ae264b8f01d36480e319405a4486c51bd0ba690d2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:49:32.607143Z","signature_b64":"7FoAKoCyj5znjLZGpTEWC2JYmO76B5DZWXItcs0jtaJqWS7I1XYVZJyG9Df3N9cg91p6DBIyzJVjXxT2CSvfCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcae044910ecef22868ed7f4101ce2ff85347796dedebaed0beea5215b72233c","last_reissued_at":"2026-07-05T09:49:32.606680Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:49:32.606680Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Biased or Flawed? Mitigating Stereotypes in Generative Language Models by Addressing Task-Specific Flaws","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Akshita Jha, Chandan K. Reddy, Sanchit Kabra","submitted_at":"2024-12-16T03:29:08Z","abstract_excerpt":"Recent studies have shown that generative language models often reflect and amplify societal biases in their outputs. However, these studies frequently conflate observed biases with other task-specific shortcomings, such as comprehension failure. For example, when a model misinterprets a text and produces a response that reinforces a stereotype, it becomes difficult to determine whether the issue arises from inherent bias or from a misunderstanding of the given content. In this paper, we conduct a multi-faceted evaluation that distinctly disentangles bias from flaws within the reading comprehe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11414","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/2412.11414/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":"2412.11414","created_at":"2026-07-05T09:49:32.606732+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.11414v1","created_at":"2026-07-05T09:49:32.606732+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11414","created_at":"2026-07-05T09:49:32.606732+00:00"},{"alias_kind":"pith_short_12","alias_value":"7SXAISIQ5TXS","created_at":"2026-07-05T09:49:32.606732+00:00"},{"alias_kind":"pith_short_16","alias_value":"7SXAISIQ5TXSFBUO","created_at":"2026-07-05T09:49:32.606732+00:00"},{"alias_kind":"pith_short_8","alias_value":"7SXAISIQ","created_at":"2026-07-05T09:49:32.606732+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.00963","citing_title":"Who Gets Left Behind? Auditing Disability Inclusivity in Large Language Models","ref_index":2024,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76","json":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76.json","graph_json":"https://pith.science/api/pith-number/7SXAISIQ5TXSFBUO272BAHHC76/graph.json","events_json":"https://pith.science/api/pith-number/7SXAISIQ5TXSFBUO272BAHHC76/events.json","paper":"https://pith.science/paper/7SXAISIQ"},"agent_actions":{"view_html":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76","download_json":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76.json","view_paper":"https://pith.science/paper/7SXAISIQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.11414&json=true","fetch_graph":"https://pith.science/api/pith-number/7SXAISIQ5TXSFBUO272BAHHC76/graph.json","fetch_events":"https://pith.science/api/pith-number/7SXAISIQ5TXSFBUO272BAHHC76/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76/action/storage_attestation","attest_author":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76/action/author_attestation","sign_citation":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76/action/citation_signature","submit_replication":"https://pith.science/pith/7SXAISIQ5TXSFBUO272BAHHC76/action/replication_record"}},"created_at":"2026-07-05T09:49:32.606732+00:00","updated_at":"2026-07-05T09:49:32.606732+00:00"}