{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:IISO3BHK726W47K53BY5ZIYM2O","short_pith_number":"pith:IISO3BHK","canonical_record":{"source":{"id":"2109.05704","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T04:52:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f16613af20ebfe94aa89667df413268fa4fac72e95967d2448343df38aabd231","abstract_canon_sha256":"0bc8e9eeb5724591bad55523d07c63083dc1ff7d08cd29b8f99f3d0e62f9419e"},"schema_version":"1.0"},"canonical_sha256":"4224ed84eafebd6e7d5dd871dca30cd3aa82d2d999e80710e494d923168f9683","source":{"kind":"arxiv","id":"2109.05704","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.05704","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"arxiv_version","alias_value":"2109.05704v2","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05704","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"pith_short_12","alias_value":"IISO3BHK726W","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"pith_short_16","alias_value":"IISO3BHK726W47K5","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"pith_short_8","alias_value":"IISO3BHK","created_at":"2026-07-05T03:13:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:IISO3BHK726W47K53BY5ZIYM2O","target":"record","payload":{"canonical_record":{"source":{"id":"2109.05704","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T04:52:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f16613af20ebfe94aa89667df413268fa4fac72e95967d2448343df38aabd231","abstract_canon_sha256":"0bc8e9eeb5724591bad55523d07c63083dc1ff7d08cd29b8f99f3d0e62f9419e"},"schema_version":"1.0"},"canonical_sha256":"4224ed84eafebd6e7d5dd871dca30cd3aa82d2d999e80710e494d923168f9683","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:51.089130Z","signature_b64":"1DIJWriXJe1mKzsNlKj35cB0GFLNPuu00lFjZWnm0lp2dAK/ZVpAYYA3Xh4bgsrGL22klo66ZUvom3mjc2xkAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4224ed84eafebd6e7d5dd871dca30cd3aa82d2d999e80710e494d923168f9683","last_reissued_at":"2026-07-05T03:13:51.088714Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:51.088714Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.05704","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:13:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DLxcxc8+Y3oZUEjCDC5yw+VsIrST3vf+veTD2xVDmLUluIl0KRE0OdyYp0rVueRTVRIoPGdclN1aY+CkZ1IxCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T04:57:34.624890Z"},"content_sha256":"d4b9ca1a3141970788cf25f8e08a961c15ca740afb22f68a3b65fa4e46738c65","schema_version":"1.0","event_id":"sha256:d4b9ca1a3141970788cf25f8e08a961c15ca740afb22f68a3b65fa4e46738c65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:IISO3BHK726W47K53BY5ZIYM2O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mitigating Language-Dependent Ethnic Bias in BERT","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alice Oh, Jaimeen Ahn","submitted_at":"2021-09-13T04:52:41Z","abstract_excerpt":"BERT and other large-scale language models (LMs) contain gender and racial bias. They also exhibit other dimensions of social bias, most of which have not been studied in depth, and some of which vary depending on the language. In this paper, we study ethnic bias and how it varies across languages by analyzing and mitigating ethnic bias in monolingual BERT for English, German, Spanish, Korean, Turkish, and Chinese. To observe and quantify ethnic bias, we develop a novel metric called Categorical Bias score. Then we propose two methods for mitigation; first using a multilingual model, and secon"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05704","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/2109.05704/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:13:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yFnRB79URLvL51O/+E3WfpWhQv+ijYqdaaE7qXvwbpTaFo2/wPhc8FHT9tGIYlEz6ybHf8V0BOzxq35lSFkGAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T04:57:34.625388Z"},"content_sha256":"0305008c5dc7be2641f0ce25e4822111c5c877168f6e67a0f57a8cadad6f37e0","schema_version":"1.0","event_id":"sha256:0305008c5dc7be2641f0ce25e4822111c5c877168f6e67a0f57a8cadad6f37e0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IISO3BHK726W47K53BY5ZIYM2O/bundle.json","state_url":"https://pith.science/pith/IISO3BHK726W47K53BY5ZIYM2O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IISO3BHK726W47K53BY5ZIYM2O/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T04:57:34Z","links":{"resolver":"https://pith.science/pith/IISO3BHK726W47K53BY5ZIYM2O","bundle":"https://pith.science/pith/IISO3BHK726W47K53BY5ZIYM2O/bundle.json","state":"https://pith.science/pith/IISO3BHK726W47K53BY5ZIYM2O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IISO3BHK726W47K53BY5ZIYM2O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:IISO3BHK726W47K53BY5ZIYM2O","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0bc8e9eeb5724591bad55523d07c63083dc1ff7d08cd29b8f99f3d0e62f9419e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T04:52:41Z","title_canon_sha256":"f16613af20ebfe94aa89667df413268fa4fac72e95967d2448343df38aabd231"},"schema_version":"1.0","source":{"id":"2109.05704","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.05704","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"arxiv_version","alias_value":"2109.05704v2","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05704","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"pith_short_12","alias_value":"IISO3BHK726W","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"pith_short_16","alias_value":"IISO3BHK726W47K5","created_at":"2026-07-05T03:13:51Z"},{"alias_kind":"pith_short_8","alias_value":"IISO3BHK","created_at":"2026-07-05T03:13:51Z"}],"graph_snapshots":[{"event_id":"sha256:0305008c5dc7be2641f0ce25e4822111c5c877168f6e67a0f57a8cadad6f37e0","target":"graph","created_at":"2026-07-05T03:13:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2109.05704/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"BERT and other large-scale language models (LMs) contain gender and racial bias. They also exhibit other dimensions of social bias, most of which have not been studied in depth, and some of which vary depending on the language. In this paper, we study ethnic bias and how it varies across languages by analyzing and mitigating ethnic bias in monolingual BERT for English, German, Spanish, Korean, Turkish, and Chinese. To observe and quantify ethnic bias, we develop a novel metric called Categorical Bias score. Then we propose two methods for mitigation; first using a multilingual model, and secon","authors_text":"Alice Oh, Jaimeen Ahn","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T04:52:41Z","title":"Mitigating Language-Dependent Ethnic Bias in BERT"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05704","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d4b9ca1a3141970788cf25f8e08a961c15ca740afb22f68a3b65fa4e46738c65","target":"record","created_at":"2026-07-05T03:13:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0bc8e9eeb5724591bad55523d07c63083dc1ff7d08cd29b8f99f3d0e62f9419e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T04:52:41Z","title_canon_sha256":"f16613af20ebfe94aa89667df413268fa4fac72e95967d2448343df38aabd231"},"schema_version":"1.0","source":{"id":"2109.05704","kind":"arxiv","version":2}},"canonical_sha256":"4224ed84eafebd6e7d5dd871dca30cd3aa82d2d999e80710e494d923168f9683","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4224ed84eafebd6e7d5dd871dca30cd3aa82d2d999e80710e494d923168f9683","first_computed_at":"2026-07-05T03:13:51.088714Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:13:51.088714Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1DIJWriXJe1mKzsNlKj35cB0GFLNPuu00lFjZWnm0lp2dAK/ZVpAYYA3Xh4bgsrGL22klo66ZUvom3mjc2xkAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:13:51.089130Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.05704","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4b9ca1a3141970788cf25f8e08a961c15ca740afb22f68a3b65fa4e46738c65","sha256:0305008c5dc7be2641f0ce25e4822111c5c877168f6e67a0f57a8cadad6f37e0"],"state_sha256":"f3f3cd52cedf82da361ea234e5fc1641662f03f42cd406a676f8705006c818b1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0Z3h42miy/jusrWG+ez1rmxzArEQtQvKsCX2UmFax9zv42dahS4Yef1zHxrO+AEKPwDVDIHFXSaObShboskuBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T04:57:34.628701Z","bundle_sha256":"44905c2d8f269abdb03e883628861808cad6b9ff20f860aa834175f41ae5b4ba"}}