{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5S3KLP446RBPY2RQCKJ4C6X6NH","short_pith_number":"pith:5S3KLP44","schema_version":"1.0","canonical_sha256":"ecb6a5bf9cf442fc6a301293c17afe69ef2306820c85257cafbf8df19651bd32","source":{"kind":"arxiv","id":"2507.13138","version":2},"attestation_state":"computed","paper":{"title":"Assessing the Reliability of LLMs Annotations in the Context of Demographic Bias and Model Explanation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anastasia Giachanou, Ayoub Bagheri, Hadi Mohammadi, Massimo Poesio, Pablo Mosteiro, Tina Shahedi","submitted_at":"2025-07-17T14:00:13Z","abstract_excerpt":"Understanding the sources of variability in annotations is crucial for developing fair NLP systems, especially for tasks like sexism detection where demographic bias is a concern. This study investigates the extent to which annotator demographic features influence labeling decisions compared to text content. Using a Generalized Linear Mixed Model, we quantify this inf luence, finding that while statistically present, demographic factors account for a minor fraction ( 8%) of the observed variance, with tweet content being the dominant factor. We then assess the reliability of Generative AI (Gen"},"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":"2507.13138","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T14:00:13Z","cross_cats_sorted":[],"title_canon_sha256":"e37334e3dca933fe310f1c2c2933142ebafa9b2a62b7f88437ef4f05a996b6e0","abstract_canon_sha256":"e7e989ee0f9444bb22121e23f3e40a1174eef1ab244c9f7a279a4398a5481c92"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:43:33.743972Z","signature_b64":"HM7UD4X4K8oN2iahhMU9IUptsK8n2k3hfPC9c4nk1OLCj7A02tOAK5gvf+e8jxVE8kS/jOKud+10O/TApogQAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ecb6a5bf9cf442fc6a301293c17afe69ef2306820c85257cafbf8df19651bd32","last_reissued_at":"2026-07-05T11:43:33.743388Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:43:33.743388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Assessing the Reliability of LLMs Annotations in the Context of Demographic Bias and Model Explanation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anastasia Giachanou, Ayoub Bagheri, Hadi Mohammadi, Massimo Poesio, Pablo Mosteiro, Tina Shahedi","submitted_at":"2025-07-17T14:00:13Z","abstract_excerpt":"Understanding the sources of variability in annotations is crucial for developing fair NLP systems, especially for tasks like sexism detection where demographic bias is a concern. This study investigates the extent to which annotator demographic features influence labeling decisions compared to text content. Using a Generalized Linear Mixed Model, we quantify this inf luence, finding that while statistically present, demographic factors account for a minor fraction ( 8%) of the observed variance, with tweet content being the dominant factor. We then assess the reliability of Generative AI (Gen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.13138","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/2507.13138/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":"2507.13138","created_at":"2026-07-05T11:43:33.743460+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.13138v2","created_at":"2026-07-05T11:43:33.743460+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.13138","created_at":"2026-07-05T11:43:33.743460+00:00"},{"alias_kind":"pith_short_12","alias_value":"5S3KLP446RBP","created_at":"2026-07-05T11:43:33.743460+00:00"},{"alias_kind":"pith_short_16","alias_value":"5S3KLP446RBPY2RQ","created_at":"2026-07-05T11:43:33.743460+00:00"},{"alias_kind":"pith_short_8","alias_value":"5S3KLP44","created_at":"2026-07-05T11:43:33.743460+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.21319","citing_title":"Do Large Language Models Understand Morality Across Cultures?","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH","json":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH.json","graph_json":"https://pith.science/api/pith-number/5S3KLP446RBPY2RQCKJ4C6X6NH/graph.json","events_json":"https://pith.science/api/pith-number/5S3KLP446RBPY2RQCKJ4C6X6NH/events.json","paper":"https://pith.science/paper/5S3KLP44"},"agent_actions":{"view_html":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH","download_json":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH.json","view_paper":"https://pith.science/paper/5S3KLP44","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.13138&json=true","fetch_graph":"https://pith.science/api/pith-number/5S3KLP446RBPY2RQCKJ4C6X6NH/graph.json","fetch_events":"https://pith.science/api/pith-number/5S3KLP446RBPY2RQCKJ4C6X6NH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH/action/storage_attestation","attest_author":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH/action/author_attestation","sign_citation":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH/action/citation_signature","submit_replication":"https://pith.science/pith/5S3KLP446RBPY2RQCKJ4C6X6NH/action/replication_record"}},"created_at":"2026-07-05T11:43:33.743460+00:00","updated_at":"2026-07-05T11:43:33.743460+00:00"}