{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4NVG7ENXRCPLFGP35KT2MWVR75","short_pith_number":"pith:4NVG7ENX","schema_version":"1.0","canonical_sha256":"e36a6f91b7889eb299fbeaa7a65ab1ff70c59c110fd6f8c1b331cdeb365d44ae","source":{"kind":"arxiv","id":"2109.05093","version":1},"attestation_state":"computed","paper":{"title":"PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.PL"],"primary_cat":"cs.CL","authors_text":"Dzmitry Bahdanau, Nathan Schucher, Torsten Scholak","submitted_at":"2021-09-10T20:14:08Z","abstract_excerpt":"Large pre-trained language models for textual data have an unconstrained output space; at each decoding step, they can produce any of 10,000s of sub-word tokens. When fine-tuned to target constrained formal languages like SQL, these models often generate invalid code, rendering it unusable. We propose PICARD (code and trained models available at https://github.com/ElementAI/picard), a method for constraining auto-regressive decoders of language models through incremental parsing. PICARD helps to find valid output sequences by rejecting inadmissible tokens at each decoding step. On the challeng"},"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":"2109.05093","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-10T20:14:08Z","cross_cats_sorted":["cs.PL"],"title_canon_sha256":"43f7eab864bb79a4b6cdb2447778bfc1b2e2fb60a1f69b5c66920957dbdb046e","abstract_canon_sha256":"016cf22ed420560c5d3c6a69106503af2f485aa06b167cda6ffcdee56e294811"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:34.277895Z","signature_b64":"cVxSfgQ41wELIdGISJ2kTDkrMWwylh8atWvFE6oyqOksFM1UrS16razau7/UFrPKjyrg0rwdt69aolxJ5ulXDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e36a6f91b7889eb299fbeaa7a65ab1ff70c59c110fd6f8c1b331cdeb365d44ae","last_reissued_at":"2026-07-05T03:13:34.277391Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:34.277391Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.PL"],"primary_cat":"cs.CL","authors_text":"Dzmitry Bahdanau, Nathan Schucher, Torsten Scholak","submitted_at":"2021-09-10T20:14:08Z","abstract_excerpt":"Large pre-trained language models for textual data have an unconstrained output space; at each decoding step, they can produce any of 10,000s of sub-word tokens. When fine-tuned to target constrained formal languages like SQL, these models often generate invalid code, rendering it unusable. We propose PICARD (code and trained models available at https://github.com/ElementAI/picard), a method for constraining auto-regressive decoders of language models through incremental parsing. PICARD helps to find valid output sequences by rejecting inadmissible tokens at each decoding step. On the challeng"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05093","kind":"arxiv","version":1},"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/2109.05093/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":"2109.05093","created_at":"2026-07-05T03:13:34.277441+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.05093v1","created_at":"2026-07-05T03:13:34.277441+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05093","created_at":"2026-07-05T03:13:34.277441+00:00"},{"alias_kind":"pith_short_12","alias_value":"4NVG7ENXRCPL","created_at":"2026-07-05T03:13:34.277441+00:00"},{"alias_kind":"pith_short_16","alias_value":"4NVG7ENXRCPLFGP3","created_at":"2026-07-05T03:13:34.277441+00:00"},{"alias_kind":"pith_short_8","alias_value":"4NVG7ENX","created_at":"2026-07-05T03:13:34.277441+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26613","citing_title":"EcoTable: Cost-effective Table Integration in Data Lakes for Natural Language Queries","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12387","citing_title":"TAHOE: Text-to-SQL with Automated Hint Optimization from Experience","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2511.22277","citing_title":"TreeCoder: Systematic Exploration and Optimisation of Decoding and Constraints for LLM Code Generation","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2603.23375","citing_title":"Natural Language Interfaces for Spatial and Temporal Databases: A Comprehensive Overview of Methods, Taxonomy, and Future Directions","ref_index":102,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25928","citing_title":"CogRAG: Tackling Heterogeneous Cognitive Demands in RAG via Stratified Retrieval and Reasoning","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20117","citing_title":"To Know is to Construct: Schema-Constrained Generation for Agent Memory","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75","json":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75.json","graph_json":"https://pith.science/api/pith-number/4NVG7ENXRCPLFGP35KT2MWVR75/graph.json","events_json":"https://pith.science/api/pith-number/4NVG7ENXRCPLFGP35KT2MWVR75/events.json","paper":"https://pith.science/paper/4NVG7ENX"},"agent_actions":{"view_html":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75","download_json":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75.json","view_paper":"https://pith.science/paper/4NVG7ENX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.05093&json=true","fetch_graph":"https://pith.science/api/pith-number/4NVG7ENXRCPLFGP35KT2MWVR75/graph.json","fetch_events":"https://pith.science/api/pith-number/4NVG7ENXRCPLFGP35KT2MWVR75/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75/action/storage_attestation","attest_author":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75/action/author_attestation","sign_citation":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75/action/citation_signature","submit_replication":"https://pith.science/pith/4NVG7ENXRCPLFGP35KT2MWVR75/action/replication_record"}},"created_at":"2026-07-05T03:13:34.277441+00:00","updated_at":"2026-07-05T03:13:34.277441+00:00"}