{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EQQJICZ7XOUDGV7TG63W4KXRD7","short_pith_number":"pith:EQQJICZ7","schema_version":"1.0","canonical_sha256":"2420940b3fbba83357f337b76e2af11fd5db5129a504afbac7997776387141ce","source":{"kind":"arxiv","id":"2504.02327","version":2},"attestation_state":"computed","paper":{"title":"LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junfeng Zhao, Rihong Qiu, Tianyu Jia, Weibin Liao, Xin Gao, Xinyu Ma, Yang Lin, Yasha Wang, Yifan Zhu","submitted_at":"2025-04-03T06:59:44Z","abstract_excerpt":"Natural Language to SQL (NL2SQL) aims to translate natural language queries into executable SQL statements, offering non-expert users intuitive access to databases. While recent approaches leveraging large-scale private LLMs such as GPT-4 have achieved state-of-the-art results, they face two critical challenges: the lack of openness and reproducibility, and the prohibitive computational cost of test-time scaling. To address these issues, we explore improving the model-level performance of small-scale public LLMs in NL2SQL under resource-constrained settings. Our exploratory experiments reveal "},"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":"2504.02327","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-03T06:59:44Z","cross_cats_sorted":[],"title_canon_sha256":"ca9c54cfec0d1cc74249ec1c544851946155b676d833f4e2186b6380327a5228","abstract_canon_sha256":"9ef8c5808262ed62fd4e33211af8eec04a1cd75d83a8d8eeadaedaa354a26c78"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:11.199987Z","signature_b64":"lxtibJRoHw3qmnqAiWKtjr3MLW2S7bSyHdHXe9NyOyZFcTNmCd/2e35wX9mi/Kc4RhgaL0YZkaJteK2sk+C6Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2420940b3fbba83357f337b76e2af11fd5db5129a504afbac7997776387141ce","last_reissued_at":"2026-07-03T01:17:11.199591Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:11.199591Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junfeng Zhao, Rihong Qiu, Tianyu Jia, Weibin Liao, Xin Gao, Xinyu Ma, Yang Lin, Yasha Wang, Yifan Zhu","submitted_at":"2025-04-03T06:59:44Z","abstract_excerpt":"Natural Language to SQL (NL2SQL) aims to translate natural language queries into executable SQL statements, offering non-expert users intuitive access to databases. While recent approaches leveraging large-scale private LLMs such as GPT-4 have achieved state-of-the-art results, they face two critical challenges: the lack of openness and reproducibility, and the prohibitive computational cost of test-time scaling. To address these issues, we explore improving the model-level performance of small-scale public LLMs in NL2SQL under resource-constrained settings. Our exploratory experiments reveal "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.02327","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/2504.02327/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":"2504.02327","created_at":"2026-07-03T01:17:11.199641+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.02327v2","created_at":"2026-07-03T01:17:11.199641+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.02327","created_at":"2026-07-03T01:17:11.199641+00:00"},{"alias_kind":"pith_short_12","alias_value":"EQQJICZ7XOUD","created_at":"2026-07-03T01:17:11.199641+00:00"},{"alias_kind":"pith_short_16","alias_value":"EQQJICZ7XOUDGV7T","created_at":"2026-07-03T01:17:11.199641+00:00"},{"alias_kind":"pith_short_8","alias_value":"EQQJICZ7","created_at":"2026-07-03T01:17:11.199641+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2606.28589","citing_title":"Search for Truth from Reasoning: A Dynamic Representation Editing Framework for Steering LLM Trajectories","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2606.06188","citing_title":"The Tell-Tale Norm: $\\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2606.28589","citing_title":"Search for Truth from Reasoning: A Dynamic Representation Editing Framework for Steering LLM Trajectories","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09295","citing_title":"LEAF-SQL: Level-wise Exploration with Adaptive Fine-graining for Text-to-SQL Skeleton Prediction","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7","json":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7.json","graph_json":"https://pith.science/api/pith-number/EQQJICZ7XOUDGV7TG63W4KXRD7/graph.json","events_json":"https://pith.science/api/pith-number/EQQJICZ7XOUDGV7TG63W4KXRD7/events.json","paper":"https://pith.science/paper/EQQJICZ7"},"agent_actions":{"view_html":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7","download_json":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7.json","view_paper":"https://pith.science/paper/EQQJICZ7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.02327&json=true","fetch_graph":"https://pith.science/api/pith-number/EQQJICZ7XOUDGV7TG63W4KXRD7/graph.json","fetch_events":"https://pith.science/api/pith-number/EQQJICZ7XOUDGV7TG63W4KXRD7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7/action/storage_attestation","attest_author":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7/action/author_attestation","sign_citation":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7/action/citation_signature","submit_replication":"https://pith.science/pith/EQQJICZ7XOUDGV7TG63W4KXRD7/action/replication_record"}},"created_at":"2026-07-03T01:17:11.199641+00:00","updated_at":"2026-07-03T01:17:11.199641+00:00"}