{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:3NOXMB6OICW4FJVUFO3DEIRBGQ","short_pith_number":"pith:3NOXMB6O","canonical_record":{"source":{"id":"2003.04983","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-02-29T00:39:12Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"a64cc15c71650650107a79e5f6a42f4364ad605ae0cc710d1c85b4cd69b3385c","abstract_canon_sha256":"4d0311d2c618140cc63956e45a2b07b708e4bee400f32c77e7e8d6288849e267"},"schema_version":"1.0"},"canonical_sha256":"db5d7607ce40adc2a6b42bb632222134065c54826e2e141ee167d7c9acfddd9a","source":{"kind":"arxiv","id":"2003.04983","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2003.04983","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"arxiv_version","alias_value":"2003.04983v1","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.04983","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"pith_short_12","alias_value":"3NOXMB6OICW4","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"pith_short_16","alias_value":"3NOXMB6OICW4FJVU","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"pith_short_8","alias_value":"3NOXMB6O","created_at":"2026-07-05T00:47:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:3NOXMB6OICW4FJVUFO3DEIRBGQ","target":"record","payload":{"canonical_record":{"source":{"id":"2003.04983","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-02-29T00:39:12Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"a64cc15c71650650107a79e5f6a42f4364ad605ae0cc710d1c85b4cd69b3385c","abstract_canon_sha256":"4d0311d2c618140cc63956e45a2b07b708e4bee400f32c77e7e8d6288849e267"},"schema_version":"1.0"},"canonical_sha256":"db5d7607ce40adc2a6b42bb632222134065c54826e2e141ee167d7c9acfddd9a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:47:19.954234Z","signature_b64":"ySUqagKU8tB5Nn+FWYr7vFHAbn9dqFX4xQxnnQQ4alV7lIUwxzjIwMk1DnVRZ9oflCJ+47akmF3/01O3S30gDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db5d7607ce40adc2a6b42bb632222134065c54826e2e141ee167d7c9acfddd9a","last_reissued_at":"2026-07-05T00:47:19.953809Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:47:19.953809Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2003.04983","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":"2026-07-05T00:47:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gTVz+Mf40Ys8+zkfClFaFgWUQfTMR6sOcva+57N3kJKurGLKMztrUPZLH40ZdlW00j91uz3iMBp5+4rM5cJ4DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T17:50:18.221455Z"},"content_sha256":"13d9e045be355d9559adbaa042a0e8d43478b3838121141d73f68517cdf4f3d5","schema_version":"1.0","event_id":"sha256:13d9e045be355d9559adbaa042a0e8d43478b3838121141d73f68517cdf4f3d5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:3NOXMB6OICW4FJVUFO3DEIRBGQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Understanding the Downstream Instability of Word Embeddings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Avner May, Christopher R. Aberger, Christopher R\\'e, Jian Zhang, Megan Leszczynski, Sen Wu","submitted_at":"2020-02-29T00:39:12Z","abstract_excerpt":"Many industrial machine learning (ML) systems require frequent retraining to keep up-to-date with constantly changing data. This retraining exacerbates a large challenge facing ML systems today: model training is unstable, i.e., small changes in training data can cause significant changes in the model's predictions. In this paper, we work on developing a deeper understanding of this instability, with a focus on how a core building block of modern natural language processing (NLP) pipelines---pre-trained word embeddings---affects the instability of downstream NLP models. We first empirically re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.04983","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/2003.04983/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":"2026-07-05T00:47:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k8f6B/SmRsuiEiSltFv3hqQ5vjrVl54FZMaQPPwIvNHtwtgHd06do4ITF/wBk8xKP224g/j5t51NPEP1syH3Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T17:50:18.222496Z"},"content_sha256":"25cfcd4e954dd10011960e43410ddcbc1b46424fbe5593120402ccd8475cf0b2","schema_version":"1.0","event_id":"sha256:25cfcd4e954dd10011960e43410ddcbc1b46424fbe5593120402ccd8475cf0b2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ/bundle.json","state_url":"https://pith.science/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ/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":"2026-08-03T17:50:18Z","links":{"resolver":"https://pith.science/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ","bundle":"https://pith.science/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ/bundle.json","state":"https://pith.science/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3NOXMB6OICW4FJVUFO3DEIRBGQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:3NOXMB6OICW4FJVUFO3DEIRBGQ","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":"4d0311d2c618140cc63956e45a2b07b708e4bee400f32c77e7e8d6288849e267","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-02-29T00:39:12Z","title_canon_sha256":"a64cc15c71650650107a79e5f6a42f4364ad605ae0cc710d1c85b4cd69b3385c"},"schema_version":"1.0","source":{"id":"2003.04983","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2003.04983","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"arxiv_version","alias_value":"2003.04983v1","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.04983","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"pith_short_12","alias_value":"3NOXMB6OICW4","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"pith_short_16","alias_value":"3NOXMB6OICW4FJVU","created_at":"2026-07-05T00:47:19Z"},{"alias_kind":"pith_short_8","alias_value":"3NOXMB6O","created_at":"2026-07-05T00:47:19Z"}],"graph_snapshots":[{"event_id":"sha256:25cfcd4e954dd10011960e43410ddcbc1b46424fbe5593120402ccd8475cf0b2","target":"graph","created_at":"2026-07-05T00:47:19Z","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/2003.04983/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Many industrial machine learning (ML) systems require frequent retraining to keep up-to-date with constantly changing data. This retraining exacerbates a large challenge facing ML systems today: model training is unstable, i.e., small changes in training data can cause significant changes in the model's predictions. In this paper, we work on developing a deeper understanding of this instability, with a focus on how a core building block of modern natural language processing (NLP) pipelines---pre-trained word embeddings---affects the instability of downstream NLP models. We first empirically re","authors_text":"Avner May, Christopher R. Aberger, Christopher R\\'e, Jian Zhang, Megan Leszczynski, Sen Wu","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-02-29T00:39:12Z","title":"Understanding the Downstream Instability of Word Embeddings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.04983","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":"sha256:13d9e045be355d9559adbaa042a0e8d43478b3838121141d73f68517cdf4f3d5","target":"record","created_at":"2026-07-05T00:47:19Z","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":"4d0311d2c618140cc63956e45a2b07b708e4bee400f32c77e7e8d6288849e267","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-02-29T00:39:12Z","title_canon_sha256":"a64cc15c71650650107a79e5f6a42f4364ad605ae0cc710d1c85b4cd69b3385c"},"schema_version":"1.0","source":{"id":"2003.04983","kind":"arxiv","version":1}},"canonical_sha256":"db5d7607ce40adc2a6b42bb632222134065c54826e2e141ee167d7c9acfddd9a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"db5d7607ce40adc2a6b42bb632222134065c54826e2e141ee167d7c9acfddd9a","first_computed_at":"2026-07-05T00:47:19.953809Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:47:19.953809Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ySUqagKU8tB5Nn+FWYr7vFHAbn9dqFX4xQxnnQQ4alV7lIUwxzjIwMk1DnVRZ9oflCJ+47akmF3/01O3S30gDg==","signature_status":"signed_v1","signed_at":"2026-07-05T00:47:19.954234Z","signed_message":"canonical_sha256_bytes"},"source_id":"2003.04983","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:13d9e045be355d9559adbaa042a0e8d43478b3838121141d73f68517cdf4f3d5","sha256:25cfcd4e954dd10011960e43410ddcbc1b46424fbe5593120402ccd8475cf0b2"],"state_sha256":"788ce97f84fd8a41c94329b76bc4284c562a636c86c66819976097b2a8827077"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Mfll+KefgXPkLEFNspvOeVYcvbEl5qlj2myecFCq7Rvzc09MIDRSJhhkEodyTiXlfCm/Y7Ou0UpMM7XR8JZNCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T17:50:18.229095Z","bundle_sha256":"b7ad976374be906d122c5be4f1a27b710e7582931380ba41e5a60492c4ccb1c5"}}