{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LRFUQRGCLBI4G2XPZRLPQI6MMW","short_pith_number":"pith:LRFUQRGC","canonical_record":{"source":{"id":"2507.16410","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T10:02:28Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"0c08d9698ef9cf7d0174f764aa8261aebb1886c193beda71c1570fd1243524eb","abstract_canon_sha256":"3a970916b0891ae1f76fe9b18ca0721054aa2bd850c090e8c774f59fa6fbaef2"},"schema_version":"1.0"},"canonical_sha256":"5c4b4844c25851c36aefcc56f823cc65928dad28170421f0e765a4e99da47ed0","source":{"kind":"arxiv","id":"2507.16410","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.16410","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"arxiv_version","alias_value":"2507.16410v1","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.16410","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"pith_short_12","alias_value":"LRFUQRGCLBI4","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"pith_short_16","alias_value":"LRFUQRGCLBI4G2XP","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"pith_short_8","alias_value":"LRFUQRGC","created_at":"2026-07-05T11:41:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LRFUQRGCLBI4G2XPZRLPQI6MMW","target":"record","payload":{"canonical_record":{"source":{"id":"2507.16410","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T10:02:28Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"0c08d9698ef9cf7d0174f764aa8261aebb1886c193beda71c1570fd1243524eb","abstract_canon_sha256":"3a970916b0891ae1f76fe9b18ca0721054aa2bd850c090e8c774f59fa6fbaef2"},"schema_version":"1.0"},"canonical_sha256":"5c4b4844c25851c36aefcc56f823cc65928dad28170421f0e765a4e99da47ed0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:41:08.972169Z","signature_b64":"PmKUPvGbnibwEWCktw1lGF386JZdJ3aacB4hQgDbo6wa24hMwRwCQhXem9zYDo3nNcPN3Tq1tp/1nBZxv1zmAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5c4b4844c25851c36aefcc56f823cc65928dad28170421f0e765a4e99da47ed0","last_reissued_at":"2026-07-05T11:41:08.971690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:41:08.971690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.16410","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":"2026-07-05T11:41:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/cUHTyOea5HCvoHmyInLgukliEOGggPMEnDndKv4urgavb786N9GLlRFD0HVHeidT+MC5Gwwmm29/3JBmJXaDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:48:01.798678Z"},"content_sha256":"4ff72c958c02324e9b55d6e87869d113383a5af36fcddc97bfdfaea3d47f2eea","schema_version":"1.0","event_id":"sha256:4ff72c958c02324e9b55d6e87869d113383a5af36fcddc97bfdfaea3d47f2eea"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LRFUQRGCLBI4G2XPZRLPQI6MMW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GG-BBQ: German Gender Bias Benchmark for Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.CL","authors_text":"H\\'ector Allende-Cid, Katharina Beckh, Katrin Klug, Sebastian Houben, Shalaka Satheesh, Teena Hassan","submitted_at":"2025-07-22T10:02:28Z","abstract_excerpt":"Within the context of Natural Language Processing (NLP), fairness evaluation is often associated with the assessment of bias and reduction of associated harm. In this regard, the evaluation is usually carried out by using a benchmark dataset, for a task such as Question Answering, created for the measurement of bias in the model's predictions along various dimensions, including gender identity. In our work, we evaluate gender bias in German Large Language Models (LLMs) using the Bias Benchmark for Question Answering by Parrish et al. (2022) as a reference. Specifically, the templates in the ge"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.16410","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/2507.16410/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":"2026-07-05T11:41:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k+GLvMXrqaJygGCF/NDoX5vZeUHIhuGFnQ6RAGzSvO3jol2ksjnlGeZNR02j8HecphI2zIVCgKlxxujr3dPTDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:48:01.799175Z"},"content_sha256":"db47ed7bb7b661e387d09b365190b90dcede0e203e74d77a91ce750f66f414e6","schema_version":"1.0","event_id":"sha256:db47ed7bb7b661e387d09b365190b90dcede0e203e74d77a91ce750f66f414e6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW/bundle.json","state_url":"https://pith.science/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW/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":"2026-08-07T19:48:01Z","links":{"resolver":"https://pith.science/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW","bundle":"https://pith.science/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW/bundle.json","state":"https://pith.science/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LRFUQRGCLBI4G2XPZRLPQI6MMW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LRFUQRGCLBI4G2XPZRLPQI6MMW","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":"3a970916b0891ae1f76fe9b18ca0721054aa2bd850c090e8c774f59fa6fbaef2","cross_cats_sorted":["cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T10:02:28Z","title_canon_sha256":"0c08d9698ef9cf7d0174f764aa8261aebb1886c193beda71c1570fd1243524eb"},"schema_version":"1.0","source":{"id":"2507.16410","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.16410","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"arxiv_version","alias_value":"2507.16410v1","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.16410","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"pith_short_12","alias_value":"LRFUQRGCLBI4","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"pith_short_16","alias_value":"LRFUQRGCLBI4G2XP","created_at":"2026-07-05T11:41:08Z"},{"alias_kind":"pith_short_8","alias_value":"LRFUQRGC","created_at":"2026-07-05T11:41:08Z"}],"graph_snapshots":[{"event_id":"sha256:db47ed7bb7b661e387d09b365190b90dcede0e203e74d77a91ce750f66f414e6","target":"graph","created_at":"2026-07-05T11:41:08Z","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/2507.16410/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Within the context of Natural Language Processing (NLP), fairness evaluation is often associated with the assessment of bias and reduction of associated harm. In this regard, the evaluation is usually carried out by using a benchmark dataset, for a task such as Question Answering, created for the measurement of bias in the model's predictions along various dimensions, including gender identity. In our work, we evaluate gender bias in German Large Language Models (LLMs) using the Bias Benchmark for Question Answering by Parrish et al. (2022) as a reference. Specifically, the templates in the ge","authors_text":"H\\'ector Allende-Cid, Katharina Beckh, Katrin Klug, Sebastian Houben, Shalaka Satheesh, Teena Hassan","cross_cats":["cs.CY","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T10:02:28Z","title":"GG-BBQ: German Gender Bias Benchmark for Question Answering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.16410","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":"sha256:4ff72c958c02324e9b55d6e87869d113383a5af36fcddc97bfdfaea3d47f2eea","target":"record","created_at":"2026-07-05T11:41:08Z","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":"3a970916b0891ae1f76fe9b18ca0721054aa2bd850c090e8c774f59fa6fbaef2","cross_cats_sorted":["cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T10:02:28Z","title_canon_sha256":"0c08d9698ef9cf7d0174f764aa8261aebb1886c193beda71c1570fd1243524eb"},"schema_version":"1.0","source":{"id":"2507.16410","kind":"arxiv","version":1}},"canonical_sha256":"5c4b4844c25851c36aefcc56f823cc65928dad28170421f0e765a4e99da47ed0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5c4b4844c25851c36aefcc56f823cc65928dad28170421f0e765a4e99da47ed0","first_computed_at":"2026-07-05T11:41:08.971690Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:41:08.971690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PmKUPvGbnibwEWCktw1lGF386JZdJ3aacB4hQgDbo6wa24hMwRwCQhXem9zYDo3nNcPN3Tq1tp/1nBZxv1zmAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:41:08.972169Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.16410","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ff72c958c02324e9b55d6e87869d113383a5af36fcddc97bfdfaea3d47f2eea","sha256:db47ed7bb7b661e387d09b365190b90dcede0e203e74d77a91ce750f66f414e6"],"state_sha256":"3e15eab86a57458c93b38e2e89ff2f0912fc509bd06e126a29ac54eea79670d5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Id5EP8FGtIlZ53xOmKn8vVthw0dR4DZE+q8oop7etISA/z4/pjOHw8BYtQpt6OWgzXIjGPjU4p6v6sTubJdCCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T19:48:01.805281Z","bundle_sha256":"7df46a90f0e8064fdda51223f38e5e4a73c91d55bc13c90b8dfe5d99b2c2e79f"}}