{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:UM4PNQ6SD7HWSTQVNWEBRCJTY7","short_pith_number":"pith:UM4PNQ6S","canonical_record":{"source":{"id":"1904.03035","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-05T12:40:28Z","cross_cats_sorted":[],"title_canon_sha256":"8461bcaad98765e2194d86cc71fff462eaabbb919ba1dbaec714a412bb750a0a","abstract_canon_sha256":"3ac2597c098fb4b6a3e88800039a32a3b872241b07503b7bc7bff318975f5d40"},"schema_version":"1.0"},"canonical_sha256":"a338f6c3d21fcf694e156d88188933c7c1b9842324bdeb572e09a344c704ef49","source":{"kind":"arxiv","id":"1904.03035","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1904.03035","created_at":"2026-05-17T23:49:18Z"},{"alias_kind":"arxiv_version","alias_value":"1904.03035v1","created_at":"2026-05-17T23:49:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.03035","created_at":"2026-05-17T23:49:18Z"},{"alias_kind":"pith_short_12","alias_value":"UM4PNQ6SD7HW","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"UM4PNQ6SD7HWSTQV","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"UM4PNQ6S","created_at":"2026-05-18T12:33:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:UM4PNQ6SD7HWSTQVNWEBRCJTY7","target":"record","payload":{"canonical_record":{"source":{"id":"1904.03035","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-05T12:40:28Z","cross_cats_sorted":[],"title_canon_sha256":"8461bcaad98765e2194d86cc71fff462eaabbb919ba1dbaec714a412bb750a0a","abstract_canon_sha256":"3ac2597c098fb4b6a3e88800039a32a3b872241b07503b7bc7bff318975f5d40"},"schema_version":"1.0"},"canonical_sha256":"a338f6c3d21fcf694e156d88188933c7c1b9842324bdeb572e09a344c704ef49","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:18.798662Z","signature_b64":"YxDWxR5z1fPwNECZbRPAyiJvHFWaxQQbM45/Y9f59echblCB5e+VYCaytofyhe0H5Yb5Rsq2GC0pX+/nukCqDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a338f6c3d21fcf694e156d88188933c7c1b9842324bdeb572e09a344c704ef49","last_reissued_at":"2026-05-17T23:49:18.798020Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:18.798020Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1904.03035","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:49:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"42DulYnYRt+k9/5nS+hyAMyaOdq1i2zJuOflViE2tVwPwX+4mYdqXMZlaWkUaPsl1AdyAAONu9gm0V3inTgsDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T07:49:23.452677Z"},"content_sha256":"257b2243a667957db4ef89a905ffe462bec5f959c8aea59e457b81db84d1b21d","schema_version":"1.0","event_id":"sha256:257b2243a667957db4ef89a905ffe462bec5f959c8aea59e457b81db84d1b21d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:UM4PNQ6SD7HWSTQVNWEBRCJTY7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Identifying and Reducing Gender Bias in Word-Level Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Samuel R. Bowman, Shikha Bordia","submitted_at":"2019-04-05T12:40:28Z","abstract_excerpt":"Many text corpora exhibit socially problematic biases, which can be propagated or amplified in the models trained on such data. For example, doctor cooccurs more frequently with male pronouns than female pronouns. In this study we (i) propose a metric to measure gender bias; (ii) measure bias in a text corpus and the text generated from a recurrent neural network language model trained on the text corpus; (iii) propose a regularization loss term for the language model that minimizes the projection of encoder-trained embeddings onto an embedding subspace that encodes gender; (iv) finally, evalu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.03035","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":""},"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":"2026-05-17T23:49:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l0RU3/RKp1+bBXiefNBEcqM64jFiXHw+hyLmDkYWw4b3u9+xKAVHOHPPFzHxidjW4S38iwHAfYU2U/kRpKpTCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T07:49:23.453023Z"},"content_sha256":"1653d2242afb58f5b2f0af9d11737dd2a6e2ad4170f13d9a7f5d8292ff133917","schema_version":"1.0","event_id":"sha256:1653d2242afb58f5b2f0af9d11737dd2a6e2ad4170f13d9a7f5d8292ff133917"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7/bundle.json","state_url":"https://pith.science/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7/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":"2026-06-03T07:49:23Z","links":{"resolver":"https://pith.science/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7","bundle":"https://pith.science/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7/bundle.json","state":"https://pith.science/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UM4PNQ6SD7HWSTQVNWEBRCJTY7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:UM4PNQ6SD7HWSTQVNWEBRCJTY7","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":"3ac2597c098fb4b6a3e88800039a32a3b872241b07503b7bc7bff318975f5d40","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-05T12:40:28Z","title_canon_sha256":"8461bcaad98765e2194d86cc71fff462eaabbb919ba1dbaec714a412bb750a0a"},"schema_version":"1.0","source":{"id":"1904.03035","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1904.03035","created_at":"2026-05-17T23:49:18Z"},{"alias_kind":"arxiv_version","alias_value":"1904.03035v1","created_at":"2026-05-17T23:49:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.03035","created_at":"2026-05-17T23:49:18Z"},{"alias_kind":"pith_short_12","alias_value":"UM4PNQ6SD7HW","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"UM4PNQ6SD7HWSTQV","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"UM4PNQ6S","created_at":"2026-05-18T12:33:30Z"}],"graph_snapshots":[{"event_id":"sha256:1653d2242afb58f5b2f0af9d11737dd2a6e2ad4170f13d9a7f5d8292ff133917","target":"graph","created_at":"2026-05-17T23:49:18Z","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"},"paper":{"abstract_excerpt":"Many text corpora exhibit socially problematic biases, which can be propagated or amplified in the models trained on such data. For example, doctor cooccurs more frequently with male pronouns than female pronouns. In this study we (i) propose a metric to measure gender bias; (ii) measure bias in a text corpus and the text generated from a recurrent neural network language model trained on the text corpus; (iii) propose a regularization loss term for the language model that minimizes the projection of encoder-trained embeddings onto an embedding subspace that encodes gender; (iv) finally, evalu","authors_text":"Samuel R. Bowman, Shikha Bordia","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-05T12:40:28Z","title":"Identifying and Reducing Gender Bias in Word-Level Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.03035","kind":"arxiv","version":1},"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":"sha256:257b2243a667957db4ef89a905ffe462bec5f959c8aea59e457b81db84d1b21d","target":"record","created_at":"2026-05-17T23:49:18Z","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":"3ac2597c098fb4b6a3e88800039a32a3b872241b07503b7bc7bff318975f5d40","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-05T12:40:28Z","title_canon_sha256":"8461bcaad98765e2194d86cc71fff462eaabbb919ba1dbaec714a412bb750a0a"},"schema_version":"1.0","source":{"id":"1904.03035","kind":"arxiv","version":1}},"canonical_sha256":"a338f6c3d21fcf694e156d88188933c7c1b9842324bdeb572e09a344c704ef49","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a338f6c3d21fcf694e156d88188933c7c1b9842324bdeb572e09a344c704ef49","first_computed_at":"2026-05-17T23:49:18.798020Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:49:18.798020Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YxDWxR5z1fPwNECZbRPAyiJvHFWaxQQbM45/Y9f59echblCB5e+VYCaytofyhe0H5Yb5Rsq2GC0pX+/nukCqDw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:49:18.798662Z","signed_message":"canonical_sha256_bytes"},"source_id":"1904.03035","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:257b2243a667957db4ef89a905ffe462bec5f959c8aea59e457b81db84d1b21d","sha256:1653d2242afb58f5b2f0af9d11737dd2a6e2ad4170f13d9a7f5d8292ff133917"],"state_sha256":"0994dd4e4e4c9723daadc719fcd01edf1d7c61078f9d3f1e5e7624a659e48166"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xW46MSgIDE4GCh1N62e3Ikx+2gXxlTWcy9wG2wwQR35hHNI4Iy8VdpPn5SCgDEwzGDDLM4HTko6s8WuyDiQkCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T07:49:23.454952Z","bundle_sha256":"ce77d5cacf95081ad0f0df6f5e5fcd2cda17d66a3c0ee1d15cd59e299e055e0c"}}