{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2VC2HONQ65LKSW7MLJ2PP5FQM4","short_pith_number":"pith:2VC2HONQ","schema_version":"1.0","canonical_sha256":"d545a3b9b0f756a95bec5a74f7f4b06732848a17dbcc81036c99dca788a84f57","source":{"kind":"arxiv","id":"2303.00716","version":2},"attestation_state":"computed","paper":{"title":"Aligning benchmark datasets for table structure recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Brandon Smock, Robin Abraham, Rohith Pesala","submitted_at":"2023-03-01T18:20:24Z","abstract_excerpt":"Benchmark datasets for table structure recognition (TSR) must be carefully processed to ensure they are annotated consistently. However, even if a dataset's annotations are self-consistent, there may be significant inconsistency across datasets, which can harm the performance of models trained and evaluated on them. In this work, we show that aligning these benchmarks$\\unicode{x2014}$removing both errors and inconsistency between them$\\unicode{x2014}$improves model performance significantly. We demonstrate this through a data-centric approach where we adopt one model architecture, the Table Tr"},"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":"2303.00716","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-01T18:20:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1024ea139a8a7cda1bbd18254ea79055f16f39ce58d5d9c7bd6581a4aef2c596","abstract_canon_sha256":"310f18930051851c710a55f9bb3f53c3a8fa4d5f1060b26587f858f44e67ee9c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:19.722455Z","signature_b64":"i4TUyNp06vXqZXCtGbiPXyyH9P9uaStziQNRUP8qmhIc8lfYr4CovLAGJ5BJvgU/Dsfzm+uVM4o/uxrEpcbqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d545a3b9b0f756a95bec5a74f7f4b06732848a17dbcc81036c99dca788a84f57","last_reissued_at":"2026-07-05T06:13:19.722031Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:19.722031Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Aligning benchmark datasets for table structure recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Brandon Smock, Robin Abraham, Rohith Pesala","submitted_at":"2023-03-01T18:20:24Z","abstract_excerpt":"Benchmark datasets for table structure recognition (TSR) must be carefully processed to ensure they are annotated consistently. However, even if a dataset's annotations are self-consistent, there may be significant inconsistency across datasets, which can harm the performance of models trained and evaluated on them. In this work, we show that aligning these benchmarks$\\unicode{x2014}$removing both errors and inconsistency between them$\\unicode{x2014}$improves model performance significantly. We demonstrate this through a data-centric approach where we adopt one model architecture, the Table Tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.00716","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/2303.00716/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":"2303.00716","created_at":"2026-07-05T06:13:19.722085+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.00716v2","created_at":"2026-07-05T06:13:19.722085+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.00716","created_at":"2026-07-05T06:13:19.722085+00:00"},{"alias_kind":"pith_short_12","alias_value":"2VC2HONQ65LK","created_at":"2026-07-05T06:13:19.722085+00:00"},{"alias_kind":"pith_short_16","alias_value":"2VC2HONQ65LKSW7M","created_at":"2026-07-05T06:13:19.722085+00:00"},{"alias_kind":"pith_short_8","alias_value":"2VC2HONQ","created_at":"2026-07-05T06:13:19.722085+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/2VC2HONQ65LKSW7MLJ2PP5FQM4","json":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4.json","graph_json":"https://pith.science/api/pith-number/2VC2HONQ65LKSW7MLJ2PP5FQM4/graph.json","events_json":"https://pith.science/api/pith-number/2VC2HONQ65LKSW7MLJ2PP5FQM4/events.json","paper":"https://pith.science/paper/2VC2HONQ"},"agent_actions":{"view_html":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4","download_json":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4.json","view_paper":"https://pith.science/paper/2VC2HONQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.00716&json=true","fetch_graph":"https://pith.science/api/pith-number/2VC2HONQ65LKSW7MLJ2PP5FQM4/graph.json","fetch_events":"https://pith.science/api/pith-number/2VC2HONQ65LKSW7MLJ2PP5FQM4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4/action/storage_attestation","attest_author":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4/action/author_attestation","sign_citation":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4/action/citation_signature","submit_replication":"https://pith.science/pith/2VC2HONQ65LKSW7MLJ2PP5FQM4/action/replication_record"}},"created_at":"2026-07-05T06:13:19.722085+00:00","updated_at":"2026-07-05T06:13:19.722085+00:00"}