{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XBOFKHVE5BF5UJVUPYOCV3TU6U","short_pith_number":"pith:XBOFKHVE","canonical_record":{"source":{"id":"2501.14457","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T12:41:30Z","cross_cats_sorted":[],"title_canon_sha256":"4044a5be09024fc642969fd97d1740803975a72821f7c1112d952bc5ebe4b292","abstract_canon_sha256":"f9821b4e76cf4d692bb9fb9dde7580be8710e9afa641439f3ab5ec2383c5ac50"},"schema_version":"1.0"},"canonical_sha256":"b85c551ea4e84bda26b47e1c2aee74f5346f577878e5ca842ad204e5e79ff463","source":{"kind":"arxiv","id":"2501.14457","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.14457","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"arxiv_version","alias_value":"2501.14457v1","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.14457","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"pith_short_12","alias_value":"XBOFKHVE5BF5","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"pith_short_16","alias_value":"XBOFKHVE5BF5UJVU","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"pith_short_8","alias_value":"XBOFKHVE","created_at":"2026-07-05T10:05:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XBOFKHVE5BF5UJVUPYOCV3TU6U","target":"record","payload":{"canonical_record":{"source":{"id":"2501.14457","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T12:41:30Z","cross_cats_sorted":[],"title_canon_sha256":"4044a5be09024fc642969fd97d1740803975a72821f7c1112d952bc5ebe4b292","abstract_canon_sha256":"f9821b4e76cf4d692bb9fb9dde7580be8710e9afa641439f3ab5ec2383c5ac50"},"schema_version":"1.0"},"canonical_sha256":"b85c551ea4e84bda26b47e1c2aee74f5346f577878e5ca842ad204e5e79ff463","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:05:01.091045Z","signature_b64":"nIvFVjgZLNnshTbOCxRlJesxTGqitrzmjfx6rkecPA4TixwgLNcjl0ez6WrIyl0ixUcXc4r7w9+xorpm9OLVDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b85c551ea4e84bda26b47e1c2aee74f5346f577878e5ca842ad204e5e79ff463","last_reissued_at":"2026-07-05T10:05:01.090602Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:05:01.090602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.14457","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":"2026-07-05T10:05:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5C16Mr+xSGMRaSMHr8zOHETANkmA5YNqp6sEh/CkJ5aluXFxVSZfUcAdVkgqAcYg3iVuDafzeQPFeMTuJxEdBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T12:58:57.331257Z"},"content_sha256":"925776dc1010634324505c52260889ed3e30e8ff2d5fbe37e844f5ee47e73493","schema_version":"1.0","event_id":"sha256:925776dc1010634324505c52260889ed3e30e8ff2d5fbe37e844f5ee47e73493"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XBOFKHVE5BF5UJVUPYOCV3TU6U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Understanding and Mitigating Gender Bias in LLMs via Interpretable Neuron Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Sophia Ananiadou, Zeping Yu","submitted_at":"2025-01-24T12:41:30Z","abstract_excerpt":"Large language models (LLMs) often exhibit gender bias, posing challenges for their safe deployment. Existing methods to mitigate bias lack a comprehensive understanding of its mechanisms or compromise the model's core capabilities. To address these issues, we propose the CommonWords dataset, to systematically evaluate gender bias in LLMs. Our analysis reveals pervasive bias across models and identifies specific neuron circuits, including gender neurons and general neurons, responsible for this behavior. Notably, editing even a small number of general neurons can disrupt the model's overall ca"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.14457","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/2501.14457/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":"2026-07-05T10:05:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7qEAVJyijQOiwvOM/rk4SUxtj6rA1uCuWTrMiGMkJJntXT/Qc1zZo25Opi5XsvHnsr5XAeGBBj3TYClQYt1yCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T12:58:57.331556Z"},"content_sha256":"ee2ca385709899db196239cd9c2dbaef58217b4bbd8ff4cf3bef7b07b4468c03","schema_version":"1.0","event_id":"sha256:ee2ca385709899db196239cd9c2dbaef58217b4bbd8ff4cf3bef7b07b4468c03"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U/bundle.json","state_url":"https://pith.science/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U/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":"2026-08-05T12:58:57Z","links":{"resolver":"https://pith.science/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U","bundle":"https://pith.science/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U/bundle.json","state":"https://pith.science/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XBOFKHVE5BF5UJVUPYOCV3TU6U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XBOFKHVE5BF5UJVUPYOCV3TU6U","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":"f9821b4e76cf4d692bb9fb9dde7580be8710e9afa641439f3ab5ec2383c5ac50","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T12:41:30Z","title_canon_sha256":"4044a5be09024fc642969fd97d1740803975a72821f7c1112d952bc5ebe4b292"},"schema_version":"1.0","source":{"id":"2501.14457","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.14457","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"arxiv_version","alias_value":"2501.14457v1","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.14457","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"pith_short_12","alias_value":"XBOFKHVE5BF5","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"pith_short_16","alias_value":"XBOFKHVE5BF5UJVU","created_at":"2026-07-05T10:05:01Z"},{"alias_kind":"pith_short_8","alias_value":"XBOFKHVE","created_at":"2026-07-05T10:05:01Z"}],"graph_snapshots":[{"event_id":"sha256:ee2ca385709899db196239cd9c2dbaef58217b4bbd8ff4cf3bef7b07b4468c03","target":"graph","created_at":"2026-07-05T10:05:01Z","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/2501.14457/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) often exhibit gender bias, posing challenges for their safe deployment. Existing methods to mitigate bias lack a comprehensive understanding of its mechanisms or compromise the model's core capabilities. To address these issues, we propose the CommonWords dataset, to systematically evaluate gender bias in LLMs. Our analysis reveals pervasive bias across models and identifies specific neuron circuits, including gender neurons and general neurons, responsible for this behavior. Notably, editing even a small number of general neurons can disrupt the model's overall ca","authors_text":"Sophia Ananiadou, Zeping Yu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T12:41:30Z","title":"Understanding and Mitigating Gender Bias in LLMs via Interpretable Neuron Editing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.14457","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":"sha256:925776dc1010634324505c52260889ed3e30e8ff2d5fbe37e844f5ee47e73493","target":"record","created_at":"2026-07-05T10:05:01Z","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":"f9821b4e76cf4d692bb9fb9dde7580be8710e9afa641439f3ab5ec2383c5ac50","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-24T12:41:30Z","title_canon_sha256":"4044a5be09024fc642969fd97d1740803975a72821f7c1112d952bc5ebe4b292"},"schema_version":"1.0","source":{"id":"2501.14457","kind":"arxiv","version":1}},"canonical_sha256":"b85c551ea4e84bda26b47e1c2aee74f5346f577878e5ca842ad204e5e79ff463","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b85c551ea4e84bda26b47e1c2aee74f5346f577878e5ca842ad204e5e79ff463","first_computed_at":"2026-07-05T10:05:01.090602Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:05:01.090602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nIvFVjgZLNnshTbOCxRlJesxTGqitrzmjfx6rkecPA4TixwgLNcjl0ez6WrIyl0ixUcXc4r7w9+xorpm9OLVDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:05:01.091045Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.14457","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:925776dc1010634324505c52260889ed3e30e8ff2d5fbe37e844f5ee47e73493","sha256:ee2ca385709899db196239cd9c2dbaef58217b4bbd8ff4cf3bef7b07b4468c03"],"state_sha256":"2199634e8c00d9358e36a2a92e9ae5fd8d4e4e681df88c33d487dc74927319ec"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZxX/uhNnD84zncESEBhomCEnBtOa4IIw5VXMxNe8dvPGB9Zd5pfu5R9dLtzfSUdsVGzG5/S6ul+aJqO3fawbCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T12:58:57.334426Z","bundle_sha256":"c094d1aad700f04a4c630e5276beee809ec0227da655efffccbd9d180e07f7b7"}}