{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4S2PPD4BFBOIRYL6V7RBNKJSUW","short_pith_number":"pith:4S2PPD4B","canonical_record":{"source":{"id":"2506.18598","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-23T12:58:54Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"0651aa4f66939843c34b881a33b42b2b0fa1325e0bedfca4344855e7e5690838","abstract_canon_sha256":"91e4940c333279f1b8033ef8149f60855ad650ab526a07e79a8dc1364bd18fe8"},"schema_version":"1.0"},"canonical_sha256":"e4b4f78f81285c88e17eafe216a932a58a2c256887923297611c48c90dc5bdc3","source":{"kind":"arxiv","id":"2506.18598","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.18598","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"arxiv_version","alias_value":"2506.18598v1","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18598","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"pith_short_12","alias_value":"4S2PPD4BFBOI","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"pith_short_16","alias_value":"4S2PPD4BFBOIRYL6","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"pith_short_8","alias_value":"4S2PPD4B","created_at":"2026-07-05T11:25:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4S2PPD4BFBOIRYL6V7RBNKJSUW","target":"record","payload":{"canonical_record":{"source":{"id":"2506.18598","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-23T12:58:54Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"0651aa4f66939843c34b881a33b42b2b0fa1325e0bedfca4344855e7e5690838","abstract_canon_sha256":"91e4940c333279f1b8033ef8149f60855ad650ab526a07e79a8dc1364bd18fe8"},"schema_version":"1.0"},"canonical_sha256":"e4b4f78f81285c88e17eafe216a932a58a2c256887923297611c48c90dc5bdc3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:52.788105Z","signature_b64":"wVRBQUid6Uk3A5xl3A4bBSlcmwrUOOWdeLMbmnpGexzTl/4cmazlYTJInFEHEbp3DIO2hlsIpxRgLPhfu3DuAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4b4f78f81285c88e17eafe216a932a58a2c256887923297611c48c90dc5bdc3","last_reissued_at":"2026-07-05T11:25:52.787672Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:52.787672Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.18598","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":"2026-07-05T11:25:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xjHKEw3BjcV0uq72ZSxFhHX2Ve8/xu666ATOcHTpUfRmXHdCQh3XIgyo2APRyRWgNtuTQ8NJsUod1S3cRFYPCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:20:22.712444Z"},"content_sha256":"77d78fc6c1fc947987f27bc9b68a8b973974656a2d56f6a04d763460b730ad68","schema_version":"1.0","event_id":"sha256:77d78fc6c1fc947987f27bc9b68a8b973974656a2d56f6a04d763460b730ad68"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4S2PPD4BFBOIRYL6V7RBNKJSUW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"No Training Wheels: Steering Vectors for Bias Correction at Inference Time","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Ameesh Sethi, Armaan Sethi, Aviral Gupta","submitted_at":"2025-06-23T12:58:54Z","abstract_excerpt":"Neural network classifiers trained on datasets with uneven group representation often inherit class biases and learn spurious correlations. These models may perform well on average but consistently fail on atypical groups. For example, in hair color classification, datasets may over-represent females with blond hair, reinforcing stereotypes. Although various algorithmic and data-centric methods have been proposed to address such biases, they often require retraining or significant compute. In this work, we propose a cheap, training-free method inspired by steering vectors used to edit behavior"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18598","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/2506.18598/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":"2026-07-05T11:25:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iZ14Iw0W7WFjjkN2ui+8JspFWuQMy6Ea48ISr+gtgvUQjz3rr8ENNAK0aZQp8WzF9HYhRoAI3IkD5gkOs4VKDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T11:20:22.712967Z"},"content_sha256":"a7d2e41db2dd477bdd340fd6e7aba76488d8744c7a3e9146c41afd4d956cc9f3","schema_version":"1.0","event_id":"sha256:a7d2e41db2dd477bdd340fd6e7aba76488d8744c7a3e9146c41afd4d956cc9f3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW/bundle.json","state_url":"https://pith.science/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW/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":"2026-08-08T11:20:22Z","links":{"resolver":"https://pith.science/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW","bundle":"https://pith.science/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW/bundle.json","state":"https://pith.science/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4S2PPD4BFBOIRYL6V7RBNKJSUW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4S2PPD4BFBOIRYL6V7RBNKJSUW","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":"91e4940c333279f1b8033ef8149f60855ad650ab526a07e79a8dc1364bd18fe8","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-23T12:58:54Z","title_canon_sha256":"0651aa4f66939843c34b881a33b42b2b0fa1325e0bedfca4344855e7e5690838"},"schema_version":"1.0","source":{"id":"2506.18598","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.18598","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"arxiv_version","alias_value":"2506.18598v1","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18598","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"pith_short_12","alias_value":"4S2PPD4BFBOI","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"pith_short_16","alias_value":"4S2PPD4BFBOIRYL6","created_at":"2026-07-05T11:25:52Z"},{"alias_kind":"pith_short_8","alias_value":"4S2PPD4B","created_at":"2026-07-05T11:25:52Z"}],"graph_snapshots":[{"event_id":"sha256:a7d2e41db2dd477bdd340fd6e7aba76488d8744c7a3e9146c41afd4d956cc9f3","target":"graph","created_at":"2026-07-05T11:25:52Z","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/2506.18598/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural network classifiers trained on datasets with uneven group representation often inherit class biases and learn spurious correlations. These models may perform well on average but consistently fail on atypical groups. For example, in hair color classification, datasets may over-represent females with blond hair, reinforcing stereotypes. Although various algorithmic and data-centric methods have been proposed to address such biases, they often require retraining or significant compute. In this work, we propose a cheap, training-free method inspired by steering vectors used to edit behavior","authors_text":"Ameesh Sethi, Armaan Sethi, Aviral Gupta","cross_cats":["cs.CL","cs.CV"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-23T12:58:54Z","title":"No Training Wheels: Steering Vectors for Bias Correction at Inference Time"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18598","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":"sha256:77d78fc6c1fc947987f27bc9b68a8b973974656a2d56f6a04d763460b730ad68","target":"record","created_at":"2026-07-05T11:25:52Z","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":"91e4940c333279f1b8033ef8149f60855ad650ab526a07e79a8dc1364bd18fe8","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-23T12:58:54Z","title_canon_sha256":"0651aa4f66939843c34b881a33b42b2b0fa1325e0bedfca4344855e7e5690838"},"schema_version":"1.0","source":{"id":"2506.18598","kind":"arxiv","version":1}},"canonical_sha256":"e4b4f78f81285c88e17eafe216a932a58a2c256887923297611c48c90dc5bdc3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4b4f78f81285c88e17eafe216a932a58a2c256887923297611c48c90dc5bdc3","first_computed_at":"2026-07-05T11:25:52.787672Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:25:52.787672Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wVRBQUid6Uk3A5xl3A4bBSlcmwrUOOWdeLMbmnpGexzTl/4cmazlYTJInFEHEbp3DIO2hlsIpxRgLPhfu3DuAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:25:52.788105Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.18598","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:77d78fc6c1fc947987f27bc9b68a8b973974656a2d56f6a04d763460b730ad68","sha256:a7d2e41db2dd477bdd340fd6e7aba76488d8744c7a3e9146c41afd4d956cc9f3"],"state_sha256":"7abf49d6be8ffbc0a0d455520b707f29de69f5b2cbf14bbd755e8518dca09350"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"73UtOGyCA/7zmRp9GRlGOAbKFWgUaXcCWjas0GmQpebw6SL92h5bXZuPkXn6HpnlEHN00hyof5+42VlOkcKmAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T11:20:22.719429Z","bundle_sha256":"29fdcfc34ebe1d89a1ba69d797b9cca67717c7cf6d874179cd172343ed1a6a6a"}}