{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:VSEAS4RF5QIPSKBIBXSDE5ELOO","short_pith_number":"pith:VSEAS4RF","schema_version":"1.0","canonical_sha256":"ac88097225ec10f928280de432748b7382da28509a2b9f95f60954daf5fa66f2","source":{"kind":"arxiv","id":"2005.00813","version":1},"attestation_state":"computed","paper":{"title":"Social Biases in NLP Models as Barriers for Persons with Disabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ben Hutchinson, Emily Denton, Kellie Webster, Stephen Denuyl, Vinodkumar Prabhakaran, Yu Zhong","submitted_at":"2020-05-02T12:16:54Z","abstract_excerpt":"Building equitable and inclusive NLP technologies demands consideration of whether and how social attitudes are represented in ML models. In particular, representations encoded in models often inadvertently perpetuate undesirable social biases from the data on which they are trained. In this paper, we present evidence of such undesirable biases towards mentions of disability in two different English language models: toxicity prediction and sentiment analysis. Next, we demonstrate that the neural embeddings that are the critical first step in most NLP pipelines similarly contain undesirable bia"},"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":"2005.00813","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T12:16:54Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"31b6f2539a32a80ec42ce8431becf2208872d7f9db690fecb257779f874fa111","abstract_canon_sha256":"75192e028451065c3a916a7615f6c7a71392818b0484a1858a1faf742eb888ef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:00:01.022366Z","signature_b64":"XoHOfsmmaWo3KaSrucDjHK7ueznKCetSs1twUHDFyfqDAiLWsJPLab9mMTi0uHzLgSAlHQD2BXgrHeDiVworDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac88097225ec10f928280de432748b7382da28509a2b9f95f60954daf5fa66f2","last_reissued_at":"2026-07-05T01:00:01.021929Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:00:01.021929Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Social Biases in NLP Models as Barriers for Persons with Disabilities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ben Hutchinson, Emily Denton, Kellie Webster, Stephen Denuyl, Vinodkumar Prabhakaran, Yu Zhong","submitted_at":"2020-05-02T12:16:54Z","abstract_excerpt":"Building equitable and inclusive NLP technologies demands consideration of whether and how social attitudes are represented in ML models. In particular, representations encoded in models often inadvertently perpetuate undesirable social biases from the data on which they are trained. In this paper, we present evidence of such undesirable biases towards mentions of disability in two different English language models: toxicity prediction and sentiment analysis. Next, we demonstrate that the neural embeddings that are the critical first step in most NLP pipelines similarly contain undesirable bia"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.00813","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/2005.00813/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":"2005.00813","created_at":"2026-07-05T01:00:01.021985+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.00813v1","created_at":"2026-07-05T01:00:01.021985+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.00813","created_at":"2026-07-05T01:00:01.021985+00:00"},{"alias_kind":"pith_short_12","alias_value":"VSEAS4RF5QIP","created_at":"2026-07-05T01:00:01.021985+00:00"},{"alias_kind":"pith_short_16","alias_value":"VSEAS4RF5QIPSKBI","created_at":"2026-07-05T01:00:01.021985+00:00"},{"alias_kind":"pith_short_8","alias_value":"VSEAS4RF","created_at":"2026-07-05T01:00:01.021985+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2101.00027","citing_title":"The Pile: An 800GB Dataset of Diverse Text for Language Modeling","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO","json":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO.json","graph_json":"https://pith.science/api/pith-number/VSEAS4RF5QIPSKBIBXSDE5ELOO/graph.json","events_json":"https://pith.science/api/pith-number/VSEAS4RF5QIPSKBIBXSDE5ELOO/events.json","paper":"https://pith.science/paper/VSEAS4RF"},"agent_actions":{"view_html":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO","download_json":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO.json","view_paper":"https://pith.science/paper/VSEAS4RF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.00813&json=true","fetch_graph":"https://pith.science/api/pith-number/VSEAS4RF5QIPSKBIBXSDE5ELOO/graph.json","fetch_events":"https://pith.science/api/pith-number/VSEAS4RF5QIPSKBIBXSDE5ELOO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO/action/storage_attestation","attest_author":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO/action/author_attestation","sign_citation":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO/action/citation_signature","submit_replication":"https://pith.science/pith/VSEAS4RF5QIPSKBIBXSDE5ELOO/action/replication_record"}},"created_at":"2026-07-05T01:00:01.021985+00:00","updated_at":"2026-07-05T01:00:01.021985+00:00"}