{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:R66FZ4A7I2RS6QJD7CHVX44ND6","short_pith_number":"pith:R66FZ4A7","canonical_record":{"source":{"id":"2107.09211","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-20T01:05:31Z","cross_cats_sorted":[],"title_canon_sha256":"8d114e3b651c82b30787b13194e2ff1d7de9770948cd0ec17a7a926ecbeecae7","abstract_canon_sha256":"d3c23075f5947153a3db55f45950266c1e87d537a831bb6a5bea3547cbb61551"},"schema_version":"1.0"},"canonical_sha256":"8fbc5cf01f46a32f4123f88f5bf38d1fab1807b4f2b96790b3433dbda63e4580","source":{"kind":"arxiv","id":"2107.09211","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.09211","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"arxiv_version","alias_value":"2107.09211v1","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.09211","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"pith_short_12","alias_value":"R66FZ4A7I2RS","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"pith_short_16","alias_value":"R66FZ4A7I2RS6QJD","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"pith_short_8","alias_value":"R66FZ4A7","created_at":"2026-07-05T02:59:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:R66FZ4A7I2RS6QJD7CHVX44ND6","target":"record","payload":{"canonical_record":{"source":{"id":"2107.09211","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-20T01:05:31Z","cross_cats_sorted":[],"title_canon_sha256":"8d114e3b651c82b30787b13194e2ff1d7de9770948cd0ec17a7a926ecbeecae7","abstract_canon_sha256":"d3c23075f5947153a3db55f45950266c1e87d537a831bb6a5bea3547cbb61551"},"schema_version":"1.0"},"canonical_sha256":"8fbc5cf01f46a32f4123f88f5bf38d1fab1807b4f2b96790b3433dbda63e4580","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:59:27.198293Z","signature_b64":"xvN7oPT6Twt7EwbbNz51e5XCPDXZ4/ZuaeII4YdJbvK8UnzsAg2RK0XM5rEn9QJO31yLgN6nSL43paGDotmuAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8fbc5cf01f46a32f4123f88f5bf38d1fab1807b4f2b96790b3433dbda63e4580","last_reissued_at":"2026-07-05T02:59:27.197894Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:59:27.197894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2107.09211","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":"2026-07-05T02:59:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qkZooHaG471brjyPvz50uPxCQMXub0mHpD56LQJbjI3FlfrcoufArb8ChmBecXpkEtBfvsvMOslTl84GctbQAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T15:43:32.871936Z"},"content_sha256":"3847e26883716ce3fc560874d157977e2b86d45ba6065dd6c83765cc4b8fcc11","schema_version":"1.0","event_id":"sha256:3847e26883716ce3fc560874d157977e2b86d45ba6065dd6c83765cc4b8fcc11"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:R66FZ4A7I2RS6QJD7CHVX44ND6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Understanding Gender and Racial Disparities in Image Recognition Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anindya Chakravarti, Rohan Mahadev","submitted_at":"2021-07-20T01:05:31Z","abstract_excerpt":"Large scale image classification models trained on top of popular datasets such as Imagenet have shown to have a distributional skew which leads to disparities in prediction accuracies across different subsections of population demographics. A lot of approaches have been made to solve for this distributional skew using methods that alter the model pre, post and during training. We investigate one such approach - which uses a multi-label softmax loss with cross-entropy as the loss function instead of a binary cross-entropy on a multi-label classification problem on the Inclusive Images dataset "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.09211","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/2107.09211/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":"2026-07-05T02:59:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"746SVpk9lxgvsD+rMED2si5EplpHiJ+F/dbUd7j8ehnmqL6nuJ8yjyXgUJ4cDxCOqO+hk5HnHw8BH+OkIBeEAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T15:43:32.872545Z"},"content_sha256":"65245060806ae8ceeb648eeec24fa8f4123cc795e1c81635d8b505a1e3be9246","schema_version":"1.0","event_id":"sha256:65245060806ae8ceeb648eeec24fa8f4123cc795e1c81635d8b505a1e3be9246"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R66FZ4A7I2RS6QJD7CHVX44ND6/bundle.json","state_url":"https://pith.science/pith/R66FZ4A7I2RS6QJD7CHVX44ND6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R66FZ4A7I2RS6QJD7CHVX44ND6/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":"2026-08-18T15:43:32Z","links":{"resolver":"https://pith.science/pith/R66FZ4A7I2RS6QJD7CHVX44ND6","bundle":"https://pith.science/pith/R66FZ4A7I2RS6QJD7CHVX44ND6/bundle.json","state":"https://pith.science/pith/R66FZ4A7I2RS6QJD7CHVX44ND6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R66FZ4A7I2RS6QJD7CHVX44ND6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:R66FZ4A7I2RS6QJD7CHVX44ND6","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":"d3c23075f5947153a3db55f45950266c1e87d537a831bb6a5bea3547cbb61551","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-20T01:05:31Z","title_canon_sha256":"8d114e3b651c82b30787b13194e2ff1d7de9770948cd0ec17a7a926ecbeecae7"},"schema_version":"1.0","source":{"id":"2107.09211","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.09211","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"arxiv_version","alias_value":"2107.09211v1","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.09211","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"pith_short_12","alias_value":"R66FZ4A7I2RS","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"pith_short_16","alias_value":"R66FZ4A7I2RS6QJD","created_at":"2026-07-05T02:59:27Z"},{"alias_kind":"pith_short_8","alias_value":"R66FZ4A7","created_at":"2026-07-05T02:59:27Z"}],"graph_snapshots":[{"event_id":"sha256:65245060806ae8ceeb648eeec24fa8f4123cc795e1c81635d8b505a1e3be9246","target":"graph","created_at":"2026-07-05T02:59:27Z","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/2107.09211/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large scale image classification models trained on top of popular datasets such as Imagenet have shown to have a distributional skew which leads to disparities in prediction accuracies across different subsections of population demographics. A lot of approaches have been made to solve for this distributional skew using methods that alter the model pre, post and during training. We investigate one such approach - which uses a multi-label softmax loss with cross-entropy as the loss function instead of a binary cross-entropy on a multi-label classification problem on the Inclusive Images dataset ","authors_text":"Anindya Chakravarti, Rohan Mahadev","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-20T01:05:31Z","title":"Understanding Gender and Racial Disparities in Image Recognition Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.09211","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":"sha256:3847e26883716ce3fc560874d157977e2b86d45ba6065dd6c83765cc4b8fcc11","target":"record","created_at":"2026-07-05T02:59:27Z","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":"d3c23075f5947153a3db55f45950266c1e87d537a831bb6a5bea3547cbb61551","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-20T01:05:31Z","title_canon_sha256":"8d114e3b651c82b30787b13194e2ff1d7de9770948cd0ec17a7a926ecbeecae7"},"schema_version":"1.0","source":{"id":"2107.09211","kind":"arxiv","version":1}},"canonical_sha256":"8fbc5cf01f46a32f4123f88f5bf38d1fab1807b4f2b96790b3433dbda63e4580","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8fbc5cf01f46a32f4123f88f5bf38d1fab1807b4f2b96790b3433dbda63e4580","first_computed_at":"2026-07-05T02:59:27.197894Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:59:27.197894Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xvN7oPT6Twt7EwbbNz51e5XCPDXZ4/ZuaeII4YdJbvK8UnzsAg2RK0XM5rEn9QJO31yLgN6nSL43paGDotmuAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:59:27.198293Z","signed_message":"canonical_sha256_bytes"},"source_id":"2107.09211","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3847e26883716ce3fc560874d157977e2b86d45ba6065dd6c83765cc4b8fcc11","sha256:65245060806ae8ceeb648eeec24fa8f4123cc795e1c81635d8b505a1e3be9246"],"state_sha256":"17a4d9a5275a15bd1161162e91d7390b280d2146a77a76a2142672443e6cc17d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Wa4hGdwWFK+5QZEe1LvByV84ue5Om1RU8vQg74fRZ0mc1h8UywFy7p8fwuR3UJ/MwRB/cT2M8OrsiMhor2G+AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T15:43:32.877419Z","bundle_sha256":"36827f75141783eef51aee2103dbb65e333a670e5415e6be41206cd6ef25219c"}}