{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:OV2ULC6HP64Q42EQCSYUROL7WC","short_pith_number":"pith:OV2ULC6H","schema_version":"1.0","canonical_sha256":"7575458bc77fb90e689014b148b97fb09e93b0c4f06a472aa86bce85eeafaa07","source":{"kind":"arxiv","id":"2010.12773","version":3},"attestation_state":"computed","paper":{"title":"Structure-Grounded Pretraining for Text-to-SQL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ahmed Hassan Awadallah, Christopher Meek, Huan Sun, Matthew Richardson, Oleksandr Polozov, Xiang Deng","submitted_at":"2020-10-24T04:35:35Z","abstract_excerpt":"Learning to capture text-table alignment is essential for tasks like text-to-SQL. A model needs to correctly recognize natural language references to columns and values and to ground them in the given database schema. In this paper, we present a novel weakly supervised Structure-Grounded pretraining framework (StruG) for text-to-SQL that can effectively learn to capture text-table alignment based on a parallel text-table corpus. We identify a set of novel prediction tasks: column grounding, value grounding and column-value mapping, and leverage them to pretrain a text-table encoder. Additional"},"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":"2010.12773","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-24T04:35:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"bb6bf5788ffc12c5038994c224355a32ce380143b5dd6a15c19ada46737db1ff","abstract_canon_sha256":"d58cd691f32555afb02d0822fdd31b0606b0bf76311ce814ce9c5325b70ddac9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:53:23.993778Z","signature_b64":"3qw1ul6gvnOHnYztvvkKpIycJotu8k3R58rKVESLLkQefGYSuot+TdtzP5PfBmE52wpF6thEy/jt7iMwu5jUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7575458bc77fb90e689014b148b97fb09e93b0c4f06a472aa86bce85eeafaa07","last_reissued_at":"2026-07-05T04:53:23.993394Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:53:23.993394Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Structure-Grounded Pretraining for Text-to-SQL","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ahmed Hassan Awadallah, Christopher Meek, Huan Sun, Matthew Richardson, Oleksandr Polozov, Xiang Deng","submitted_at":"2020-10-24T04:35:35Z","abstract_excerpt":"Learning to capture text-table alignment is essential for tasks like text-to-SQL. A model needs to correctly recognize natural language references to columns and values and to ground them in the given database schema. In this paper, we present a novel weakly supervised Structure-Grounded pretraining framework (StruG) for text-to-SQL that can effectively learn to capture text-table alignment based on a parallel text-table corpus. We identify a set of novel prediction tasks: column grounding, value grounding and column-value mapping, and leverage them to pretrain a text-table encoder. Additional"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.12773","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/2010.12773/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":"2010.12773","created_at":"2026-07-05T04:53:23.993450+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.12773v3","created_at":"2026-07-05T04:53:23.993450+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.12773","created_at":"2026-07-05T04:53:23.993450+00:00"},{"alias_kind":"pith_short_12","alias_value":"OV2ULC6HP64Q","created_at":"2026-07-05T04:53:23.993450+00:00"},{"alias_kind":"pith_short_16","alias_value":"OV2ULC6HP64Q42EQ","created_at":"2026-07-05T04:53:23.993450+00:00"},{"alias_kind":"pith_short_8","alias_value":"OV2ULC6H","created_at":"2026-07-05T04:53:23.993450+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.21575","citing_title":"StreamLink: Large-Language-Model Driven Distributed Data Engineering System","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC","json":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC.json","graph_json":"https://pith.science/api/pith-number/OV2ULC6HP64Q42EQCSYUROL7WC/graph.json","events_json":"https://pith.science/api/pith-number/OV2ULC6HP64Q42EQCSYUROL7WC/events.json","paper":"https://pith.science/paper/OV2ULC6H"},"agent_actions":{"view_html":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC","download_json":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC.json","view_paper":"https://pith.science/paper/OV2ULC6H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.12773&json=true","fetch_graph":"https://pith.science/api/pith-number/OV2ULC6HP64Q42EQCSYUROL7WC/graph.json","fetch_events":"https://pith.science/api/pith-number/OV2ULC6HP64Q42EQCSYUROL7WC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC/action/storage_attestation","attest_author":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC/action/author_attestation","sign_citation":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC/action/citation_signature","submit_replication":"https://pith.science/pith/OV2ULC6HP64Q42EQCSYUROL7WC/action/replication_record"}},"created_at":"2026-07-05T04:53:23.993450+00:00","updated_at":"2026-07-05T04:53:23.993450+00:00"}