{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:RSZXT6MED6QDO7OOJLLHFW2Z2D","short_pith_number":"pith:RSZXT6ME","canonical_record":{"source":{"id":"2405.15760","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T17:56:03Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"4be6bbcd2f31bc5fa33c2a3fd8996f7b4f11ab47110a55c3d86a2d28ed9f509a","abstract_canon_sha256":"dff63203a22227158cfc007cb450f4e7ea85182fc5b6c48b4a0975685c425e11"},"schema_version":"1.0"},"canonical_sha256":"8cb379f9841fa0377dce4ad672db59d0f3296991a927d1356133e67d22b61932","source":{"kind":"arxiv","id":"2405.15760","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.15760","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"arxiv_version","alias_value":"2405.15760v1","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15760","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"pith_short_12","alias_value":"RSZXT6MED6QD","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"pith_short_16","alias_value":"RSZXT6MED6QDO7OO","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"pith_short_8","alias_value":"RSZXT6ME","created_at":"2026-07-05T08:22:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:RSZXT6MED6QDO7OOJLLHFW2Z2D","target":"record","payload":{"canonical_record":{"source":{"id":"2405.15760","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T17:56:03Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"4be6bbcd2f31bc5fa33c2a3fd8996f7b4f11ab47110a55c3d86a2d28ed9f509a","abstract_canon_sha256":"dff63203a22227158cfc007cb450f4e7ea85182fc5b6c48b4a0975685c425e11"},"schema_version":"1.0"},"canonical_sha256":"8cb379f9841fa0377dce4ad672db59d0f3296991a927d1356133e67d22b61932","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:53.704692Z","signature_b64":"D/+7kraRf5pd2DljpNWBNONbpJHoat0JgT7VWmPitbuT0wMm1wnVUBz9Dfgxm30oPp9wBDW9vhGcoz0pPimKDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8cb379f9841fa0377dce4ad672db59d0f3296991a927d1356133e67d22b61932","last_reissued_at":"2026-07-05T08:22:53.704131Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:53.704131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.15760","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":"2026-07-05T08:22:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tj6bEmmbinqdKan3rEwHatlibJc+rPPhDiN5F0Ba7pmQhM3p/2S7e7lO1n/1I20eFwWxacM2uglGmbs7aSO9Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:12:48.856304Z"},"content_sha256":"f12ebdcc9354832dbb0dee06de563b311e36778b94f5307f1451ef2a61085ec3","schema_version":"1.0","event_id":"sha256:f12ebdcc9354832dbb0dee06de563b311e36778b94f5307f1451ef2a61085ec3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:RSZXT6MED6QDO7OOJLLHFW2Z2D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GPT is Not an Annotator: The Necessity of Human Annotation in Fairness Benchmark Construction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Jennifer A. Thompson, Jonathan May, Virginia K. Felkner","submitted_at":"2024-05-24T17:56:03Z","abstract_excerpt":"Social biases in LLMs are usually measured via bias benchmark datasets. Current benchmarks have limitations in scope, grounding, quality, and human effort required. Previous work has shown success with a community-sourced, rather than crowd-sourced, approach to benchmark development. However, this work still required considerable effort from annotators with relevant lived experience. This paper explores whether an LLM (specifically, GPT-3.5-Turbo) can assist with the task of developing a bias benchmark dataset from responses to an open-ended community survey. We also extend the previous work t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15760","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/2405.15760/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":"2026-07-05T08:22:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uIFlzTuhJxjTIjeMvI1/3p3hl6FidQ21w1GSnCOrnrURiXiWNWU95+758QAu/sqJYqxXYec7APdMnM97sERNDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:12:48.857374Z"},"content_sha256":"b408ae0b268e38cbfb8c25e5975f18d663394b76f52f37edf79e2c84da236b42","schema_version":"1.0","event_id":"sha256:b408ae0b268e38cbfb8c25e5975f18d663394b76f52f37edf79e2c84da236b42"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D/bundle.json","state_url":"https://pith.science/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D/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":"2026-08-05T13:12:48Z","links":{"resolver":"https://pith.science/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D","bundle":"https://pith.science/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D/bundle.json","state":"https://pith.science/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RSZXT6MED6QDO7OOJLLHFW2Z2D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:RSZXT6MED6QDO7OOJLLHFW2Z2D","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":"dff63203a22227158cfc007cb450f4e7ea85182fc5b6c48b4a0975685c425e11","cross_cats_sorted":["cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T17:56:03Z","title_canon_sha256":"4be6bbcd2f31bc5fa33c2a3fd8996f7b4f11ab47110a55c3d86a2d28ed9f509a"},"schema_version":"1.0","source":{"id":"2405.15760","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.15760","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"arxiv_version","alias_value":"2405.15760v1","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15760","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"pith_short_12","alias_value":"RSZXT6MED6QD","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"pith_short_16","alias_value":"RSZXT6MED6QDO7OO","created_at":"2026-07-05T08:22:53Z"},{"alias_kind":"pith_short_8","alias_value":"RSZXT6ME","created_at":"2026-07-05T08:22:53Z"}],"graph_snapshots":[{"event_id":"sha256:b408ae0b268e38cbfb8c25e5975f18d663394b76f52f37edf79e2c84da236b42","target":"graph","created_at":"2026-07-05T08:22:53Z","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/2405.15760/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Social biases in LLMs are usually measured via bias benchmark datasets. Current benchmarks have limitations in scope, grounding, quality, and human effort required. Previous work has shown success with a community-sourced, rather than crowd-sourced, approach to benchmark development. However, this work still required considerable effort from annotators with relevant lived experience. This paper explores whether an LLM (specifically, GPT-3.5-Turbo) can assist with the task of developing a bias benchmark dataset from responses to an open-ended community survey. We also extend the previous work t","authors_text":"Jennifer A. Thompson, Jonathan May, Virginia K. Felkner","cross_cats":["cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T17:56:03Z","title":"GPT is Not an Annotator: The Necessity of Human Annotation in Fairness Benchmark Construction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15760","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:f12ebdcc9354832dbb0dee06de563b311e36778b94f5307f1451ef2a61085ec3","target":"record","created_at":"2026-07-05T08:22:53Z","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":"dff63203a22227158cfc007cb450f4e7ea85182fc5b6c48b4a0975685c425e11","cross_cats_sorted":["cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-24T17:56:03Z","title_canon_sha256":"4be6bbcd2f31bc5fa33c2a3fd8996f7b4f11ab47110a55c3d86a2d28ed9f509a"},"schema_version":"1.0","source":{"id":"2405.15760","kind":"arxiv","version":1}},"canonical_sha256":"8cb379f9841fa0377dce4ad672db59d0f3296991a927d1356133e67d22b61932","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8cb379f9841fa0377dce4ad672db59d0f3296991a927d1356133e67d22b61932","first_computed_at":"2026-07-05T08:22:53.704131Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:22:53.704131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"D/+7kraRf5pd2DljpNWBNONbpJHoat0JgT7VWmPitbuT0wMm1wnVUBz9Dfgxm30oPp9wBDW9vhGcoz0pPimKDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:22:53.704692Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.15760","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f12ebdcc9354832dbb0dee06de563b311e36778b94f5307f1451ef2a61085ec3","sha256:b408ae0b268e38cbfb8c25e5975f18d663394b76f52f37edf79e2c84da236b42"],"state_sha256":"2ecb9d69384200189f8386cbcfe4177a5f7a91062c2bd05e1e76096bc48faeac"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JTutdDEdzppxMy2hv8URI3ONB478RolO4/4EgAte5UX4HDPHqla6TQEyfhghVlCe/lSKcwYAmzXV7e3wDcUYCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T13:12:48.898565Z","bundle_sha256":"cd3467e97a7d35e9f836ffa2ddd4a4199ec9986a076e6a26e332a667c448c25b"}}