{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:FOLWTXQHHFEZXREWS4B4JDI54W","short_pith_number":"pith:FOLWTXQH","canonical_record":{"source":{"id":"2303.06094","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2023-03-10T17:23:07Z","cross_cats_sorted":[],"title_canon_sha256":"1d28852c61f0a156b1ecec212e111e5f390a36190a08abc3366cb76a552c72fd","abstract_canon_sha256":"24a88a9a461f688ccc1b4eed5b4339d893b3eb0132d5977122987460fb2abd5e"},"schema_version":"1.0"},"canonical_sha256":"2b9769de0739499bc4969703c48d1de58831ef244b31b05711b015662f15f055","source":{"kind":"arxiv","id":"2303.06094","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.06094","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"arxiv_version","alias_value":"2303.06094v1","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.06094","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"pith_short_12","alias_value":"FOLWTXQHHFEZ","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"pith_short_16","alias_value":"FOLWTXQHHFEZXREW","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"pith_short_8","alias_value":"FOLWTXQH","created_at":"2026-07-05T05:50:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:FOLWTXQHHFEZXREWS4B4JDI54W","target":"record","payload":{"canonical_record":{"source":{"id":"2303.06094","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2023-03-10T17:23:07Z","cross_cats_sorted":[],"title_canon_sha256":"1d28852c61f0a156b1ecec212e111e5f390a36190a08abc3366cb76a552c72fd","abstract_canon_sha256":"24a88a9a461f688ccc1b4eed5b4339d893b3eb0132d5977122987460fb2abd5e"},"schema_version":"1.0"},"canonical_sha256":"2b9769de0739499bc4969703c48d1de58831ef244b31b05711b015662f15f055","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:50:00.835699Z","signature_b64":"IoglZDBfBL/TUQSr5+Cb3jd8JbXNmqO4S+ORoZfSimUPwonGde7PR9fmbezEQzQnsj3nQuVamq8I0eHblgN2Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b9769de0739499bc4969703c48d1de58831ef244b31b05711b015662f15f055","last_reissued_at":"2026-07-05T05:50:00.835186Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:50:00.835186Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.06094","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":"2026-07-05T05:50:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TWwK11vzHGVOMLA+EC3vIb91/Pj0HvyvvPCL/jsx1sluV8/l07Wl64a1JvlPsHZZ6Cscs+49K+lnj7880UjABg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T13:13:39.772703Z"},"content_sha256":"bed2e639bacb264999bfce5943afc587286d9944088ea9f5dc95ee091f5f2523","schema_version":"1.0","event_id":"sha256:bed2e639bacb264999bfce5943afc587286d9944088ea9f5dc95ee091f5f2523"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:FOLWTXQHHFEZXREWS4B4JDI54W","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Moving Fast With Broken Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DB","authors_text":"Aditya G. Parameswaran, Karl Gyllstrom, Labib Fawaz, Shreya Shankar","submitted_at":"2023-03-10T17:23:07Z","abstract_excerpt":"Machine learning (ML) models in production pipelines are frequently retrained on the latest partitions of large, continually-growing datasets. Due to engineering bugs, partitions in such datasets almost always have some corrupted features; thus, it's critical to detect data issues and block retraining before downstream ML model accuracy decreases. However, it's difficult to identify when a partition is corrupted enough to block retraining. Blocking too often yields stale model snapshots in production; blocking too little yields broken model snapshots in production.\n  In this paper, we present "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.06094","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/2303.06094/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":"2026-07-05T05:50:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Mhi36Y/xfx6QJ01tO+IeOvDA0BlAF4eUPDHbK+OQ2v63qV7S68b8CJxX+80jevv+1TDlRo9K/6BisZGoBfdBCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T13:13:39.773242Z"},"content_sha256":"3504a6d6337232c90f507763d62bec86f1c05b950feae61d292300b7b267ffb4","schema_version":"1.0","event_id":"sha256:3504a6d6337232c90f507763d62bec86f1c05b950feae61d292300b7b267ffb4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FOLWTXQHHFEZXREWS4B4JDI54W/bundle.json","state_url":"https://pith.science/pith/FOLWTXQHHFEZXREWS4B4JDI54W/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FOLWTXQHHFEZXREWS4B4JDI54W/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":"2026-08-23T13:13:39Z","links":{"resolver":"https://pith.science/pith/FOLWTXQHHFEZXREWS4B4JDI54W","bundle":"https://pith.science/pith/FOLWTXQHHFEZXREWS4B4JDI54W/bundle.json","state":"https://pith.science/pith/FOLWTXQHHFEZXREWS4B4JDI54W/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FOLWTXQHHFEZXREWS4B4JDI54W/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:FOLWTXQHHFEZXREWS4B4JDI54W","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":"24a88a9a461f688ccc1b4eed5b4339d893b3eb0132d5977122987460fb2abd5e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2023-03-10T17:23:07Z","title_canon_sha256":"1d28852c61f0a156b1ecec212e111e5f390a36190a08abc3366cb76a552c72fd"},"schema_version":"1.0","source":{"id":"2303.06094","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.06094","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"arxiv_version","alias_value":"2303.06094v1","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.06094","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"pith_short_12","alias_value":"FOLWTXQHHFEZ","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"pith_short_16","alias_value":"FOLWTXQHHFEZXREW","created_at":"2026-07-05T05:50:00Z"},{"alias_kind":"pith_short_8","alias_value":"FOLWTXQH","created_at":"2026-07-05T05:50:00Z"}],"graph_snapshots":[{"event_id":"sha256:3504a6d6337232c90f507763d62bec86f1c05b950feae61d292300b7b267ffb4","target":"graph","created_at":"2026-07-05T05:50:00Z","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/2303.06094/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Machine learning (ML) models in production pipelines are frequently retrained on the latest partitions of large, continually-growing datasets. Due to engineering bugs, partitions in such datasets almost always have some corrupted features; thus, it's critical to detect data issues and block retraining before downstream ML model accuracy decreases. However, it's difficult to identify when a partition is corrupted enough to block retraining. Blocking too often yields stale model snapshots in production; blocking too little yields broken model snapshots in production.\n  In this paper, we present ","authors_text":"Aditya G. Parameswaran, Karl Gyllstrom, Labib Fawaz, Shreya Shankar","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2023-03-10T17:23:07Z","title":"Moving Fast With Broken Data"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.06094","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:bed2e639bacb264999bfce5943afc587286d9944088ea9f5dc95ee091f5f2523","target":"record","created_at":"2026-07-05T05:50:00Z","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":"24a88a9a461f688ccc1b4eed5b4339d893b3eb0132d5977122987460fb2abd5e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2023-03-10T17:23:07Z","title_canon_sha256":"1d28852c61f0a156b1ecec212e111e5f390a36190a08abc3366cb76a552c72fd"},"schema_version":"1.0","source":{"id":"2303.06094","kind":"arxiv","version":1}},"canonical_sha256":"2b9769de0739499bc4969703c48d1de58831ef244b31b05711b015662f15f055","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2b9769de0739499bc4969703c48d1de58831ef244b31b05711b015662f15f055","first_computed_at":"2026-07-05T05:50:00.835186Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:50:00.835186Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IoglZDBfBL/TUQSr5+Cb3jd8JbXNmqO4S+ORoZfSimUPwonGde7PR9fmbezEQzQnsj3nQuVamq8I0eHblgN2Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T05:50:00.835699Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.06094","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bed2e639bacb264999bfce5943afc587286d9944088ea9f5dc95ee091f5f2523","sha256:3504a6d6337232c90f507763d62bec86f1c05b950feae61d292300b7b267ffb4"],"state_sha256":"61177fac0f6bbfaa01b8921a100e23e4d5cfb5bab3baea9a1b413505b0e6baa5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VTFkhQGBPXHkD8OoXb73rVgod9VG9YoVIZztGfWE63IT7P3GYF6fBioEAZ8dFMP3LkUcFU9Y8bce7GP903uHAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T13:13:39.778102Z","bundle_sha256":"d758c56806f670f1537e3d9c452f4a048bd4f08db30b276af5f077fc9b421894"}}