{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5WWEN2ZLRTJ6CHNBZLPSNEGZ46","short_pith_number":"pith:5WWEN2ZL","schema_version":"1.0","canonical_sha256":"edac46eb2b8cd3e11da1cadf2690d9e7999d0f2a81e50b93ca76c7c6bd7ae554","source":{"kind":"arxiv","id":"2504.13139","version":2},"attestation_state":"computed","paper":{"title":"Syntactic and Semantic Control of Large Language Models via Sequential Monte Carlo","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexander K. Lew, Benjamin Lebrun, Ben Lipkin, Clemente Pasti, Gabriel Grand, Jason Eisner, Jo\\~ao Loula, Li Du, Marjorie Freedman, Ryan Cotterell, Tianyu Liu, Timothy J. O'Donnell, Tim Vieira, Vikash Mansinghka, Yahya Emara","submitted_at":"2025-04-17T17:49:40Z","abstract_excerpt":"A wide range of LM applications require generating text that conforms to syntactic or semantic constraints. Imposing such constraints can be naturally framed as probabilistic conditioning, but exact generation from the resulting distribution -- which can differ substantially from the LM's base distribution -- is generally intractable. In this work, we develop an architecture for controlled LM generation based on sequential Monte Carlo (SMC). Our SMC framework allows us to flexibly incorporate domain- and problem-specific constraints at inference time, and efficiently reallocate computational r"},"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":"2504.13139","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-17T17:49:40Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"36d9528fcf1ec86bc82a57967d96ed7a59c0dd6103f45c461735001cc4934c1a","abstract_canon_sha256":"94eda5412a404adb871b7b76243367cc149676e62d254acc07d61b281956dd27"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:06.670936Z","signature_b64":"VVEOmHezu8uQb+pdMM0l/X3Lqki/iho/Qhc0bwUVZuzHnoJoW3Ve/9VSxsmJNeZ8nZNpPi5ruTCn3mMHTZtUBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"edac46eb2b8cd3e11da1cadf2690d9e7999d0f2a81e50b93ca76c7c6bd7ae554","last_reissued_at":"2026-07-05T10:51:06.670290Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:06.670290Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Syntactic and Semantic Control of Large Language Models via Sequential Monte Carlo","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexander K. Lew, Benjamin Lebrun, Ben Lipkin, Clemente Pasti, Gabriel Grand, Jason Eisner, Jo\\~ao Loula, Li Du, Marjorie Freedman, Ryan Cotterell, Tianyu Liu, Timothy J. O'Donnell, Tim Vieira, Vikash Mansinghka, Yahya Emara","submitted_at":"2025-04-17T17:49:40Z","abstract_excerpt":"A wide range of LM applications require generating text that conforms to syntactic or semantic constraints. Imposing such constraints can be naturally framed as probabilistic conditioning, but exact generation from the resulting distribution -- which can differ substantially from the LM's base distribution -- is generally intractable. In this work, we develop an architecture for controlled LM generation based on sequential Monte Carlo (SMC). Our SMC framework allows us to flexibly incorporate domain- and problem-specific constraints at inference time, and efficiently reallocate computational r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13139","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/2504.13139/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":"2504.13139","created_at":"2026-07-05T10:51:06.670382+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.13139v2","created_at":"2026-07-05T10:51:06.670382+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13139","created_at":"2026-07-05T10:51:06.670382+00:00"},{"alias_kind":"pith_short_12","alias_value":"5WWEN2ZLRTJ6","created_at":"2026-07-05T10:51:06.670382+00:00"},{"alias_kind":"pith_short_16","alias_value":"5WWEN2ZLRTJ6CHNB","created_at":"2026-07-05T10:51:06.670382+00:00"},{"alias_kind":"pith_short_8","alias_value":"5WWEN2ZL","created_at":"2026-07-05T10:51:06.670382+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14892","citing_title":"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems","ref_index":116,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15365","citing_title":"Greedy or not, here I come: Language production under vocabulary constraints in humans and resource-rational models","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2509.01082","citing_title":"RefineStat: Efficient Exploration for Probabilistic Program Synthesis","ref_index":30,"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":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14892","citing_title":"Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems","ref_index":115,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16453","citing_title":"Sampling for Quality: Training-Free Reward-Guided LLM Decoding via Sequential Monte Carlo","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01846","citing_title":"Do Large Language Models Plan Answer Positions? Position Bias in Multiple-Choice Question Generation","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46","json":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46.json","graph_json":"https://pith.science/api/pith-number/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/graph.json","events_json":"https://pith.science/api/pith-number/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/events.json","paper":"https://pith.science/paper/5WWEN2ZL"},"agent_actions":{"view_html":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46","download_json":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46.json","view_paper":"https://pith.science/paper/5WWEN2ZL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.13139&json=true","fetch_graph":"https://pith.science/api/pith-number/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/graph.json","fetch_events":"https://pith.science/api/pith-number/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/action/storage_attestation","attest_author":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/action/author_attestation","sign_citation":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/action/citation_signature","submit_replication":"https://pith.science/pith/5WWEN2ZLRTJ6CHNBZLPSNEGZ46/action/replication_record"}},"created_at":"2026-07-05T10:51:06.670382+00:00","updated_at":"2026-07-05T10:51:06.670382+00:00"}