{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ES5H65KEHE4MZ3N3523S6WXSNK","short_pith_number":"pith:ES5H65KE","schema_version":"1.0","canonical_sha256":"24ba7f75443938ccedbbeeb72f5af26a855d80e993b30e3d9edccadc53fc40d9","source":{"kind":"arxiv","id":"2401.03038","version":2},"attestation_state":"computed","paper":{"title":"SPADE: Synthesizing Data Quality Assertions for Large Language Model Pipelines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.DB","authors_text":"Aditya G. Parameswaran, Eugene Wu, Haotian Li, Harrison Chase, J.D. Zamfirescu-Pereira, Madelon Hulsebos, Parth Asawa, Shreya Shankar, Will Fu-Hinthorn, Yiming Lin","submitted_at":"2024-01-05T19:27:58Z","abstract_excerpt":"Large language models (LLMs) are being increasingly deployed as part of pipelines that repeatedly process or generate data of some sort. However, a common barrier to deployment are the frequent and often unpredictable errors that plague LLMs. Acknowledging the inevitability of these errors, we propose {\\em data quality assertions} to identify when LLMs may be making mistakes. We present SPADE, a method for automatically synthesizing data quality assertions that identify bad LLM outputs. We make the observation that developers often identify data quality issues during prototyping prior to deplo"},"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":"2401.03038","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DB","submitted_at":"2024-01-05T19:27:58Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"1bd2a2b45a7172895566a71d92bbff37d1aa647acdc7098c38f9eafa9c752b31","abstract_canon_sha256":"78c047b1840e4d8c69fb4f1beaf61ecfab357b6d3e403576ff37039a02329b79"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:50.085908Z","signature_b64":"MB1EVFAZVijmg3syhmpxeUL7EX3PF7lKNFYNuCGCyht3JDcHLtLy1uFWXKIfTxsqbOmREvm/88fKdpbdkDzIBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24ba7f75443938ccedbbeeb72f5af26a855d80e993b30e3d9edccadc53fc40d9","last_reissued_at":"2026-07-05T08:02:50.085482Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:50.085482Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPADE: Synthesizing Data Quality Assertions for Large Language Model Pipelines","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.DB","authors_text":"Aditya G. Parameswaran, Eugene Wu, Haotian Li, Harrison Chase, J.D. Zamfirescu-Pereira, Madelon Hulsebos, Parth Asawa, Shreya Shankar, Will Fu-Hinthorn, Yiming Lin","submitted_at":"2024-01-05T19:27:58Z","abstract_excerpt":"Large language models (LLMs) are being increasingly deployed as part of pipelines that repeatedly process or generate data of some sort. However, a common barrier to deployment are the frequent and often unpredictable errors that plague LLMs. Acknowledging the inevitability of these errors, we propose {\\em data quality assertions} to identify when LLMs may be making mistakes. We present SPADE, a method for automatically synthesizing data quality assertions that identify bad LLM outputs. We make the observation that developers often identify data quality issues during prototyping prior to deplo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.03038","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/2401.03038/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":"2401.03038","created_at":"2026-07-05T08:02:50.085536+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.03038v2","created_at":"2026-07-05T08:02:50.085536+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.03038","created_at":"2026-07-05T08:02:50.085536+00:00"},{"alias_kind":"pith_short_12","alias_value":"ES5H65KEHE4M","created_at":"2026-07-05T08:02:50.085536+00:00"},{"alias_kind":"pith_short_16","alias_value":"ES5H65KEHE4MZ3N3","created_at":"2026-07-05T08:02:50.085536+00:00"},{"alias_kind":"pith_short_8","alias_value":"ES5H65KE","created_at":"2026-07-05T08:02:50.085536+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/ES5H65KEHE4MZ3N3523S6WXSNK","json":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK.json","graph_json":"https://pith.science/api/pith-number/ES5H65KEHE4MZ3N3523S6WXSNK/graph.json","events_json":"https://pith.science/api/pith-number/ES5H65KEHE4MZ3N3523S6WXSNK/events.json","paper":"https://pith.science/paper/ES5H65KE"},"agent_actions":{"view_html":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK","download_json":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK.json","view_paper":"https://pith.science/paper/ES5H65KE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.03038&json=true","fetch_graph":"https://pith.science/api/pith-number/ES5H65KEHE4MZ3N3523S6WXSNK/graph.json","fetch_events":"https://pith.science/api/pith-number/ES5H65KEHE4MZ3N3523S6WXSNK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK/action/storage_attestation","attest_author":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK/action/author_attestation","sign_citation":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK/action/citation_signature","submit_replication":"https://pith.science/pith/ES5H65KEHE4MZ3N3523S6WXSNK/action/replication_record"}},"created_at":"2026-07-05T08:02:50.085536+00:00","updated_at":"2026-07-05T08:02:50.085536+00:00"}