{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:7CWRI3VFPSXQYIHJLJNLGYL6IQ","short_pith_number":"pith:7CWRI3VF","schema_version":"1.0","canonical_sha256":"f8ad146ea57caf0c20e95a5ab3617e4429ef3a1150fab21b367bf38d2504ae09","source":{"kind":"arxiv","id":"2306.12550","version":1},"attestation_state":"computed","paper":{"title":"On Evaluation of Document Classification using RVL-CDIP","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gordon Lim, Kevin Leach, Stefan Larson","submitted_at":"2023-06-21T20:32:22Z","abstract_excerpt":"The RVL-CDIP benchmark is widely used for measuring performance on the task of document classification. Despite its widespread use, we reveal several undesirable characteristics of the RVL-CDIP benchmark. These include (1) substantial amounts of label noise, which we estimate to be 8.1% (ranging between 1.6% to 16.9% per document category); (2) presence of many ambiguous or multi-label documents; (3) a large overlap between test and train splits, which can inflate model performance metrics; and (4) presence of sensitive personally-identifiable information like US Social Security numbers (SSNs)"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.12550","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-21T20:32:22Z","cross_cats_sorted":[],"title_canon_sha256":"8f5ba6df4e27a6af58662fc4ada368cdcdd2d2b858c0171a270abd501df2af0f","abstract_canon_sha256":"89f5ea4c544d65737c7362eb8b9af33e863cdeb2dfa9f4c12e3d1bbf01682502"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:23:39.767704Z","signature_b64":"aRpRfHde/Iu9MKjjFg3rxNBNDGgYVRtZAb7HTue55JzCsCr58zQBlwiMfW9TKHM7OiGoL1HwriY10c45fQg2Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8ad146ea57caf0c20e95a5ab3617e4429ef3a1150fab21b367bf38d2504ae09","last_reissued_at":"2026-07-05T06:23:39.767307Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:23:39.767307Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Evaluation of Document Classification using RVL-CDIP","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gordon Lim, Kevin Leach, Stefan Larson","submitted_at":"2023-06-21T20:32:22Z","abstract_excerpt":"The RVL-CDIP benchmark is widely used for measuring performance on the task of document classification. Despite its widespread use, we reveal several undesirable characteristics of the RVL-CDIP benchmark. These include (1) substantial amounts of label noise, which we estimate to be 8.1% (ranging between 1.6% to 16.9% per document category); (2) presence of many ambiguous or multi-label documents; (3) a large overlap between test and train splits, which can inflate model performance metrics; and (4) presence of sensitive personally-identifiable information like US Social Security numbers (SSNs)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.12550","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/2306.12550/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.12550","created_at":"2026-07-05T06:23:39.767373+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.12550v1","created_at":"2026-07-05T06:23:39.767373+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.12550","created_at":"2026-07-05T06:23:39.767373+00:00"},{"alias_kind":"pith_short_12","alias_value":"7CWRI3VFPSXQ","created_at":"2026-07-05T06:23:39.767373+00:00"},{"alias_kind":"pith_short_16","alias_value":"7CWRI3VFPSXQYIHJ","created_at":"2026-07-05T06:23:39.767373+00:00"},{"alias_kind":"pith_short_8","alias_value":"7CWRI3VF","created_at":"2026-07-05T06:23:39.767373+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ","json":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ.json","graph_json":"https://pith.science/api/pith-number/7CWRI3VFPSXQYIHJLJNLGYL6IQ/graph.json","events_json":"https://pith.science/api/pith-number/7CWRI3VFPSXQYIHJLJNLGYL6IQ/events.json","paper":"https://pith.science/paper/7CWRI3VF"},"agent_actions":{"view_html":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ","download_json":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ.json","view_paper":"https://pith.science/paper/7CWRI3VF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.12550&json=true","fetch_graph":"https://pith.science/api/pith-number/7CWRI3VFPSXQYIHJLJNLGYL6IQ/graph.json","fetch_events":"https://pith.science/api/pith-number/7CWRI3VFPSXQYIHJLJNLGYL6IQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ/action/storage_attestation","attest_author":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ/action/author_attestation","sign_citation":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ/action/citation_signature","submit_replication":"https://pith.science/pith/7CWRI3VFPSXQYIHJLJNLGYL6IQ/action/replication_record"}},"created_at":"2026-07-05T06:23:39.767373+00:00","updated_at":"2026-07-05T06:23:39.767373+00:00"}