{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:B4QR64LJ3MCM7AG5SN74QAV57G","short_pith_number":"pith:B4QR64LJ","schema_version":"1.0","canonical_sha256":"0f211f7169db04cf80dd937fc802bdf983064c9b019ea8f74b584d233fc96de9","source":{"kind":"arxiv","id":"2107.07653","version":3},"attestation_state":"computed","paper":{"title":"TAPEX: Table Pre-training via Learning a Neural SQL Executor","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bei Chen, Jian-Guang Lou, Jiaqi Guo, Morteza Ziyadi, Qian Liu, Weizhu Chen, Zeqi Lin","submitted_at":"2021-07-16T00:40:11Z","abstract_excerpt":"Recent progress in language model pre-training has achieved a great success via leveraging large-scale unstructured textual data. However, it is still a challenge to apply pre-training on structured tabular data due to the absence of large-scale high-quality tabular data. In this paper, we propose TAPEX to show that table pre-training can be achieved by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries and their execution outputs. TAPEX addresses the data scarcity challenge via guiding the language model to mimic a SQ"},"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":"2107.07653","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-07-16T00:40:11Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"295144d8fe348fb327d5694522e31de2109d39ba6e1150b51d143dd0519c25fb","abstract_canon_sha256":"81da3205ab4b19416cfbaa9c7cd79f4e17d8974ddb4aadf4e7b82f3a4ae15d56"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:04:40.323088Z","signature_b64":"1Qh461exxdxZVIdtztmjY0yQ4hmxrfJjDmY/kjq/eTPWsR/iW0nRpVrg6ZAnkGXExIk6jF+I2GjMigll7cviCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f211f7169db04cf80dd937fc802bdf983064c9b019ea8f74b584d233fc96de9","last_reissued_at":"2026-07-05T04:04:40.322636Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:04:40.322636Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TAPEX: Table Pre-training via Learning a Neural SQL Executor","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bei Chen, Jian-Guang Lou, Jiaqi Guo, Morteza Ziyadi, Qian Liu, Weizhu Chen, Zeqi Lin","submitted_at":"2021-07-16T00:40:11Z","abstract_excerpt":"Recent progress in language model pre-training has achieved a great success via leveraging large-scale unstructured textual data. However, it is still a challenge to apply pre-training on structured tabular data due to the absence of large-scale high-quality tabular data. In this paper, we propose TAPEX to show that table pre-training can be achieved by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries and their execution outputs. TAPEX addresses the data scarcity challenge via guiding the language model to mimic a SQ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07653","kind":"arxiv","version":3},"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/2107.07653/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":"2107.07653","created_at":"2026-07-05T04:04:40.322693+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.07653v3","created_at":"2026-07-05T04:04:40.322693+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07653","created_at":"2026-07-05T04:04:40.322693+00:00"},{"alias_kind":"pith_short_12","alias_value":"B4QR64LJ3MCM","created_at":"2026-07-05T04:04:40.322693+00:00"},{"alias_kind":"pith_short_16","alias_value":"B4QR64LJ3MCM7AG5","created_at":"2026-07-05T04:04:40.322693+00:00"},{"alias_kind":"pith_short_8","alias_value":"B4QR64LJ","created_at":"2026-07-05T04:04:40.322693+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11537","citing_title":"MoCA-Agent: A Market-of-Claims Code Agent for Financial and Numerical Reasoning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10316","citing_title":"TabClaw: An Interactive and Self-Evolving Agent for Spreadsheet Manipulation and Table Reasoning","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29955","citing_title":"SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflows","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2509.17680","citing_title":"When TableQA Meets Noise: A Dual Denoising Framework for Complex Questions and Large-scale Tables","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2510.09671","citing_title":"Table Question Answering in the Era of Large Language Models: A Comprehensive Survey of Tasks, Methods, and Evaluation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00199","citing_title":"RSAT: Structured Attribution Makes Small Language Models Faithful Table Reasoners","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G","json":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G.json","graph_json":"https://pith.science/api/pith-number/B4QR64LJ3MCM7AG5SN74QAV57G/graph.json","events_json":"https://pith.science/api/pith-number/B4QR64LJ3MCM7AG5SN74QAV57G/events.json","paper":"https://pith.science/paper/B4QR64LJ"},"agent_actions":{"view_html":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G","download_json":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G.json","view_paper":"https://pith.science/paper/B4QR64LJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.07653&json=true","fetch_graph":"https://pith.science/api/pith-number/B4QR64LJ3MCM7AG5SN74QAV57G/graph.json","fetch_events":"https://pith.science/api/pith-number/B4QR64LJ3MCM7AG5SN74QAV57G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G/action/storage_attestation","attest_author":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G/action/author_attestation","sign_citation":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G/action/citation_signature","submit_replication":"https://pith.science/pith/B4QR64LJ3MCM7AG5SN74QAV57G/action/replication_record"}},"created_at":"2026-07-05T04:04:40.322693+00:00","updated_at":"2026-07-05T04:04:40.322693+00:00"}