{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KMYOIPFBHJKHYNV44E5QRMOLZZ","short_pith_number":"pith:KMYOIPFB","canonical_record":{"source":{"id":"2506.10491","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-12T08:47:40Z","cross_cats_sorted":[],"title_canon_sha256":"31ea80291f2a657bc1af61005dd20dfaab5224457b0022ea2a136a27bee7bad4","abstract_canon_sha256":"7e7cfb887916f7865355cdc5dc038ac0f4e059d98d58c65f6717290da96d7ee8"},"schema_version":"1.0"},"canonical_sha256":"5330e43ca13a547c36bce13b08b1cbce73badc52c5f16e882a038f51f83327a1","source":{"kind":"arxiv","id":"2506.10491","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10491","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10491v2","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10491","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"pith_short_12","alias_value":"KMYOIPFBHJKH","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"pith_short_16","alias_value":"KMYOIPFBHJKHYNV4","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"pith_short_8","alias_value":"KMYOIPFB","created_at":"2026-07-05T12:02:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KMYOIPFBHJKHYNV44E5QRMOLZZ","target":"record","payload":{"canonical_record":{"source":{"id":"2506.10491","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-12T08:47:40Z","cross_cats_sorted":[],"title_canon_sha256":"31ea80291f2a657bc1af61005dd20dfaab5224457b0022ea2a136a27bee7bad4","abstract_canon_sha256":"7e7cfb887916f7865355cdc5dc038ac0f4e059d98d58c65f6717290da96d7ee8"},"schema_version":"1.0"},"canonical_sha256":"5330e43ca13a547c36bce13b08b1cbce73badc52c5f16e882a038f51f83327a1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:47.143169Z","signature_b64":"SQW1c1+0zhdh3J0oe1UT9zRO3RyqHQQkzLCIbKtVKW1funCiHYYhCoMb+YbUNyaT7Vl4YsrSO2FhCIEo/c2ICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5330e43ca13a547c36bce13b08b1cbce73badc52c5f16e882a038f51f83327a1","last_reissued_at":"2026-07-05T12:02:47.142594Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:47.142594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.10491","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":"2026-07-05T12:02:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"G1OOUdCxFaNYG6TvZ6+ljV8kxlv5Gu6q4xC4bAsvOmWsuqR+2Zom/UTNuydUVsL3ZnpdrO3kKdqore8IvO6yBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T06:27:38.293499Z"},"content_sha256":"8edd07a5cf2bd572112e8c09d2f3cef919af7c437e7c28aaa8a747adb461e345","schema_version":"1.0","event_id":"sha256:8edd07a5cf2bd572112e8c09d2f3cef919af7c437e7c28aaa8a747adb461e345"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KMYOIPFBHJKHYNV44E5QRMOLZZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Surface Fairness, Deep Bias: A Comparative Study of Bias in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aleksandra Sorokovikova, Iuliia Eremenko, Ivan P. Yamshchikov, Pavel Chizhov","submitted_at":"2025-06-12T08:47:40Z","abstract_excerpt":"Modern language models are trained on large amounts of data. These data inevitably include controversial and stereotypical content, which contains all sorts of biases related to gender, origin, age, etc. As a result, the models express biased points of view or produce different results based on the assigned personality or the personality of the user. In this paper, we investigate various proxy measures of bias in large language models (LLMs). We find that evaluating models with pre-prompted personae on a multi-subject benchmark (MMLU) leads to negligible and mostly random differences in scores"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10491","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/2506.10491/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":"2026-07-05T12:02:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j6piUOXSh5tOEEyQdvpHgu9va8LVDCdnL/55xUSUPtg2T6m3RVbWBrkVjz5rRb8TUUD4cUtzUUOELDPvQZzVCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T06:27:38.294502Z"},"content_sha256":"758b6814c0ac99fa526edb58339613b9bc269668bf90280c10e1edb73d003be5","schema_version":"1.0","event_id":"sha256:758b6814c0ac99fa526edb58339613b9bc269668bf90280c10e1edb73d003be5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ/bundle.json","state_url":"https://pith.science/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ/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":"2026-08-10T06:27:38Z","links":{"resolver":"https://pith.science/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ","bundle":"https://pith.science/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ/bundle.json","state":"https://pith.science/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KMYOIPFBHJKHYNV44E5QRMOLZZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KMYOIPFBHJKHYNV44E5QRMOLZZ","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":"7e7cfb887916f7865355cdc5dc038ac0f4e059d98d58c65f6717290da96d7ee8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-12T08:47:40Z","title_canon_sha256":"31ea80291f2a657bc1af61005dd20dfaab5224457b0022ea2a136a27bee7bad4"},"schema_version":"1.0","source":{"id":"2506.10491","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10491","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10491v2","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10491","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"pith_short_12","alias_value":"KMYOIPFBHJKH","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"pith_short_16","alias_value":"KMYOIPFBHJKHYNV4","created_at":"2026-07-05T12:02:47Z"},{"alias_kind":"pith_short_8","alias_value":"KMYOIPFB","created_at":"2026-07-05T12:02:47Z"}],"graph_snapshots":[{"event_id":"sha256:758b6814c0ac99fa526edb58339613b9bc269668bf90280c10e1edb73d003be5","target":"graph","created_at":"2026-07-05T12:02:47Z","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/2506.10491/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern language models are trained on large amounts of data. These data inevitably include controversial and stereotypical content, which contains all sorts of biases related to gender, origin, age, etc. As a result, the models express biased points of view or produce different results based on the assigned personality or the personality of the user. In this paper, we investigate various proxy measures of bias in large language models (LLMs). We find that evaluating models with pre-prompted personae on a multi-subject benchmark (MMLU) leads to negligible and mostly random differences in scores","authors_text":"Aleksandra Sorokovikova, Iuliia Eremenko, Ivan P. Yamshchikov, Pavel Chizhov","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-12T08:47:40Z","title":"Surface Fairness, Deep Bias: A Comparative Study of Bias in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10491","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":"sha256:8edd07a5cf2bd572112e8c09d2f3cef919af7c437e7c28aaa8a747adb461e345","target":"record","created_at":"2026-07-05T12:02:47Z","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":"7e7cfb887916f7865355cdc5dc038ac0f4e059d98d58c65f6717290da96d7ee8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-12T08:47:40Z","title_canon_sha256":"31ea80291f2a657bc1af61005dd20dfaab5224457b0022ea2a136a27bee7bad4"},"schema_version":"1.0","source":{"id":"2506.10491","kind":"arxiv","version":2}},"canonical_sha256":"5330e43ca13a547c36bce13b08b1cbce73badc52c5f16e882a038f51f83327a1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5330e43ca13a547c36bce13b08b1cbce73badc52c5f16e882a038f51f83327a1","first_computed_at":"2026-07-05T12:02:47.142594Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:47.142594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SQW1c1+0zhdh3J0oe1UT9zRO3RyqHQQkzLCIbKtVKW1funCiHYYhCoMb+YbUNyaT7Vl4YsrSO2FhCIEo/c2ICg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:47.143169Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.10491","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8edd07a5cf2bd572112e8c09d2f3cef919af7c437e7c28aaa8a747adb461e345","sha256:758b6814c0ac99fa526edb58339613b9bc269668bf90280c10e1edb73d003be5"],"state_sha256":"408aa32b85eaaff9c811c6c7174f2298471a3c929295ad506aa9dc6863962390"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ymp3gTzuYbim37vL20ELjReDtLnIkhbNYjGldGJnELan1Ma5Ibhvl6VVyz2qyqKLVF51FUylZfQ2ig7xVz8RBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T06:27:38.305568Z","bundle_sha256":"ec7950f608d319f2ce363cb2f45aed7a7ace3bfbc13923c76abc43e4555165a7"}}