{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:VJSTI6M6PFXM3WNIDQ7NP6ZNNQ","short_pith_number":"pith:VJSTI6M6","canonical_record":{"source":{"id":"2303.00586","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-03-01T15:28:26Z","cross_cats_sorted":["cs.AI","cs.CV","cs.CY","cs.LG"],"title_canon_sha256":"1155b3045b123ed01ac923fbaa63df48772a106c45b5d657037285db68929aea","abstract_canon_sha256":"e4b03443747f62268bdb8935a8ae1eef9f765d98bdec542027853ed491f6ef72"},"schema_version":"1.0"},"canonical_sha256":"aa6534799e796ecdd9a81c3ed7fb2d6c1e69d372f49a5ee522e787eb46e02a59","source":{"kind":"arxiv","id":"2303.00586","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.00586","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"arxiv_version","alias_value":"2303.00586v2","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.00586","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"pith_short_12","alias_value":"VJSTI6M6PFXM","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"pith_short_16","alias_value":"VJSTI6M6PFXM3WNI","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"pith_short_8","alias_value":"VJSTI6M6","created_at":"2026-07-05T07:26:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:VJSTI6M6PFXM3WNIDQ7NP6ZNNQ","target":"record","payload":{"canonical_record":{"source":{"id":"2303.00586","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-03-01T15:28:26Z","cross_cats_sorted":["cs.AI","cs.CV","cs.CY","cs.LG"],"title_canon_sha256":"1155b3045b123ed01ac923fbaa63df48772a106c45b5d657037285db68929aea","abstract_canon_sha256":"e4b03443747f62268bdb8935a8ae1eef9f765d98bdec542027853ed491f6ef72"},"schema_version":"1.0"},"canonical_sha256":"aa6534799e796ecdd9a81c3ed7fb2d6c1e69d372f49a5ee522e787eb46e02a59","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:26:36.404612Z","signature_b64":"uk8TCezFvMNrSpNHOdqjyYALkyvspfBxxxpF/I9jvse26HkIcivnk65mlpjtd09HYSz5AfjPyOWIN+utV0N4AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa6534799e796ecdd9a81c3ed7fb2d6c1e69d372f49a5ee522e787eb46e02a59","last_reissued_at":"2026-07-05T07:26:36.404106Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:26:36.404106Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.00586","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":"2026-07-05T07:26:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A1KtVrIuDrQraETZxOKQ2m7RsCXB+7ERjQtjr7sInaWJ3jusUQknXr7ZCmzcPTO3wxZGbBVPJmTJRJXRHh1oCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T19:38:37.883029Z"},"content_sha256":"023c542a6146d4737542fc3a388f7bb19fdd3df62cb3342a31d073afdd06844b","schema_version":"1.0","event_id":"sha256:023c542a6146d4737542fc3a388f7bb19fdd3df62cb3342a31d073afdd06844b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:VJSTI6M6PFXM3WNIDQ7NP6ZNNQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensembling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.CY","cs.LG"],"primary_cat":"stat.ML","authors_text":"Daniel D'souza, Karina Nguyen, Randall Balestriero, Sara Hooker, Wei-Yin Ko","submitted_at":"2023-03-01T15:28:26Z","abstract_excerpt":"Ensembling multiple Deep Neural Networks (DNNs) is a simple and effective way to improve top-line metrics and to outperform a larger single model. In this work, we go beyond top-line metrics and instead explore the impact of ensembling on subgroup performances. Surprisingly, we observe that even with a simple homogeneous ensemble -- all the individual DNNs share the same training set, architecture, and design choices -- the minority group performance disproportionately improves with the number of models compared to the majority group, i.e. fairness naturally emerges from ensembling. Even more "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.00586","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/2303.00586/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":"2026-07-05T07:26:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cFjHDB6+fVq8wS3CjGbPm/oV2CioDIb07MuFk0d6VqAXbCc9cgAc4nK5Yh7gbXSHrIgWe3Q45FFwLjDXmRHmCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T19:38:37.883436Z"},"content_sha256":"fa8f1d767a901448a4dcce6f1e3927b1c2f55eca077e919eb1c6e1d3df226479","schema_version":"1.0","event_id":"sha256:fa8f1d767a901448a4dcce6f1e3927b1c2f55eca077e919eb1c6e1d3df226479"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ/bundle.json","state_url":"https://pith.science/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ/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":"2026-07-06T19:38:37Z","links":{"resolver":"https://pith.science/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ","bundle":"https://pith.science/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ/bundle.json","state":"https://pith.science/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VJSTI6M6PFXM3WNIDQ7NP6ZNNQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:VJSTI6M6PFXM3WNIDQ7NP6ZNNQ","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":"e4b03443747f62268bdb8935a8ae1eef9f765d98bdec542027853ed491f6ef72","cross_cats_sorted":["cs.AI","cs.CV","cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-03-01T15:28:26Z","title_canon_sha256":"1155b3045b123ed01ac923fbaa63df48772a106c45b5d657037285db68929aea"},"schema_version":"1.0","source":{"id":"2303.00586","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.00586","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"arxiv_version","alias_value":"2303.00586v2","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.00586","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"pith_short_12","alias_value":"VJSTI6M6PFXM","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"pith_short_16","alias_value":"VJSTI6M6PFXM3WNI","created_at":"2026-07-05T07:26:36Z"},{"alias_kind":"pith_short_8","alias_value":"VJSTI6M6","created_at":"2026-07-05T07:26:36Z"}],"graph_snapshots":[{"event_id":"sha256:fa8f1d767a901448a4dcce6f1e3927b1c2f55eca077e919eb1c6e1d3df226479","target":"graph","created_at":"2026-07-05T07:26:36Z","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/2303.00586/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Ensembling multiple Deep Neural Networks (DNNs) is a simple and effective way to improve top-line metrics and to outperform a larger single model. In this work, we go beyond top-line metrics and instead explore the impact of ensembling on subgroup performances. Surprisingly, we observe that even with a simple homogeneous ensemble -- all the individual DNNs share the same training set, architecture, and design choices -- the minority group performance disproportionately improves with the number of models compared to the majority group, i.e. fairness naturally emerges from ensembling. Even more ","authors_text":"Daniel D'souza, Karina Nguyen, Randall Balestriero, Sara Hooker, Wei-Yin Ko","cross_cats":["cs.AI","cs.CV","cs.CY","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-03-01T15:28:26Z","title":"FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensembling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.00586","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":"sha256:023c542a6146d4737542fc3a388f7bb19fdd3df62cb3342a31d073afdd06844b","target":"record","created_at":"2026-07-05T07:26:36Z","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":"e4b03443747f62268bdb8935a8ae1eef9f765d98bdec542027853ed491f6ef72","cross_cats_sorted":["cs.AI","cs.CV","cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-03-01T15:28:26Z","title_canon_sha256":"1155b3045b123ed01ac923fbaa63df48772a106c45b5d657037285db68929aea"},"schema_version":"1.0","source":{"id":"2303.00586","kind":"arxiv","version":2}},"canonical_sha256":"aa6534799e796ecdd9a81c3ed7fb2d6c1e69d372f49a5ee522e787eb46e02a59","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa6534799e796ecdd9a81c3ed7fb2d6c1e69d372f49a5ee522e787eb46e02a59","first_computed_at":"2026-07-05T07:26:36.404106Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:26:36.404106Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uk8TCezFvMNrSpNHOdqjyYALkyvspfBxxxpF/I9jvse26HkIcivnk65mlpjtd09HYSz5AfjPyOWIN+utV0N4AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:26:36.404612Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.00586","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:023c542a6146d4737542fc3a388f7bb19fdd3df62cb3342a31d073afdd06844b","sha256:fa8f1d767a901448a4dcce6f1e3927b1c2f55eca077e919eb1c6e1d3df226479"],"state_sha256":"7cad1c4579bd65ba396fe11c8c46f7349b76de87fd6da1f463bd446fbeb382a5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yVYpJiT/PM01/Sh3oPyl0wC27RVJc3FkHLlDo2s9s8T/3Q6yfq5GeZ4n0P9X/Tvfd17ucksG5WxnJwLo+IFLDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T19:38:37.885425Z","bundle_sha256":"6118a2a0356c8fe1485c97ada6f9c0fafb7f83ed2ccad6e5d7d514e866fd20ab"}}