{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:SRRECEPPYATJNPUAPZT4NETTJQ","short_pith_number":"pith:SRRECEPP","canonical_record":{"source":{"id":"2306.03081","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-05T17:55:05Z","cross_cats_sorted":["cs.CL","cs.PL","stat.CO"],"title_canon_sha256":"5d89a7e24576605c4bceb8c68f377e77e54070d3625680c3cb2d636dace14c1c","abstract_canon_sha256":"42a576647d942890631da89637b2bf22b576d63634f690c7edac498a174acc50"},"schema_version":"1.0"},"canonical_sha256":"94624111efc02696be807e67c692734c2517c66604c851a9c69bd8c8c182238d","source":{"kind":"arxiv","id":"2306.03081","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.03081","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"arxiv_version","alias_value":"2306.03081v2","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.03081","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"pith_short_12","alias_value":"SRRECEPPYATJ","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"pith_short_16","alias_value":"SRRECEPPYATJNPUA","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"pith_short_8","alias_value":"SRRECEPP","created_at":"2026-07-05T07:16:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:SRRECEPPYATJNPUAPZT4NETTJQ","target":"record","payload":{"canonical_record":{"source":{"id":"2306.03081","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-05T17:55:05Z","cross_cats_sorted":["cs.CL","cs.PL","stat.CO"],"title_canon_sha256":"5d89a7e24576605c4bceb8c68f377e77e54070d3625680c3cb2d636dace14c1c","abstract_canon_sha256":"42a576647d942890631da89637b2bf22b576d63634f690c7edac498a174acc50"},"schema_version":"1.0"},"canonical_sha256":"94624111efc02696be807e67c692734c2517c66604c851a9c69bd8c8c182238d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:16:42.589457Z","signature_b64":"Y7+bRFzzm0qkxpwtkWfeG6ufqHTZnjZu83bkvFANm6e9Ma7RBHho4ANnivKkUsGcxPk4ipB6RtUnsNhcuOVbBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94624111efc02696be807e67c692734c2517c66604c851a9c69bd8c8c182238d","last_reissued_at":"2026-07-05T07:16:42.589018Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:16:42.589018Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.03081","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:16:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"C9vkOLTbBjkTX9EtP32RtnRa++cDEiJ8cmArmzV7mcIhuPRtXbML+X9WWzKQoHjWKV0775t5WBKasK9F/Vx2Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:27:53.906774Z"},"content_sha256":"ca15f028835c450e6b19d05b0d74cec850208760d2295930315f02435d0ddb1e","schema_version":"1.0","event_id":"sha256:ca15f028835c450e6b19d05b0d74cec850208760d2295930315f02435d0ddb1e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:SRRECEPPYATJNPUAPZT4NETTJQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.PL","stat.CO"],"primary_cat":"cs.AI","authors_text":"Alexander K. Lew, Gabriel Grand, Tan Zhi-Xuan, Vikash K. Mansinghka","submitted_at":"2023-06-05T17:55:05Z","abstract_excerpt":"Even after fine-tuning and reinforcement learning, large language models (LLMs) can be difficult, if not impossible, to control reliably with prompts alone. We propose a new inference-time approach to enforcing syntactic and semantic constraints on the outputs of LLMs, called sequential Monte Carlo (SMC) steering. The key idea is to specify language generation tasks as posterior inference problems in a class of discrete probabilistic sequence models, and replace standard decoding with sequential Monte Carlo inference. For a computational cost similar to that of beam search, SMC can steer LLMs "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.03081","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/2306.03081/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:16:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rmt8TNHPmHc/xV1wVaLEcCKo4PvE0Q7wHhUh9veHnCYKGtPIROHM2haWu7YSnwJrOUKlJQSsghmakrkKbSx0Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:27:53.907309Z"},"content_sha256":"1554410bbbdc6868a3d1d5cf6d792a511e8a47039ca47c9ecccb5a4f86a85055","schema_version":"1.0","event_id":"sha256:1554410bbbdc6868a3d1d5cf6d792a511e8a47039ca47c9ecccb5a4f86a85055"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SRRECEPPYATJNPUAPZT4NETTJQ/bundle.json","state_url":"https://pith.science/pith/SRRECEPPYATJNPUAPZT4NETTJQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SRRECEPPYATJNPUAPZT4NETTJQ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T17:27:53Z","links":{"resolver":"https://pith.science/pith/SRRECEPPYATJNPUAPZT4NETTJQ","bundle":"https://pith.science/pith/SRRECEPPYATJNPUAPZT4NETTJQ/bundle.json","state":"https://pith.science/pith/SRRECEPPYATJNPUAPZT4NETTJQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SRRECEPPYATJNPUAPZT4NETTJQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SRRECEPPYATJNPUAPZT4NETTJQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"42a576647d942890631da89637b2bf22b576d63634f690c7edac498a174acc50","cross_cats_sorted":["cs.CL","cs.PL","stat.CO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-05T17:55:05Z","title_canon_sha256":"5d89a7e24576605c4bceb8c68f377e77e54070d3625680c3cb2d636dace14c1c"},"schema_version":"1.0","source":{"id":"2306.03081","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.03081","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"arxiv_version","alias_value":"2306.03081v2","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.03081","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"pith_short_12","alias_value":"SRRECEPPYATJ","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"pith_short_16","alias_value":"SRRECEPPYATJNPUA","created_at":"2026-07-05T07:16:42Z"},{"alias_kind":"pith_short_8","alias_value":"SRRECEPP","created_at":"2026-07-05T07:16:42Z"}],"graph_snapshots":[{"event_id":"sha256:1554410bbbdc6868a3d1d5cf6d792a511e8a47039ca47c9ecccb5a4f86a85055","target":"graph","created_at":"2026-07-05T07:16:42Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2306.03081/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Even after fine-tuning and reinforcement learning, large language models (LLMs) can be difficult, if not impossible, to control reliably with prompts alone. We propose a new inference-time approach to enforcing syntactic and semantic constraints on the outputs of LLMs, called sequential Monte Carlo (SMC) steering. The key idea is to specify language generation tasks as posterior inference problems in a class of discrete probabilistic sequence models, and replace standard decoding with sequential Monte Carlo inference. For a computational cost similar to that of beam search, SMC can steer LLMs ","authors_text":"Alexander K. Lew, Gabriel Grand, Tan Zhi-Xuan, Vikash K. Mansinghka","cross_cats":["cs.CL","cs.PL","stat.CO"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-05T17:55:05Z","title":"Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.03081","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ca15f028835c450e6b19d05b0d74cec850208760d2295930315f02435d0ddb1e","target":"record","created_at":"2026-07-05T07:16:42Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"42a576647d942890631da89637b2bf22b576d63634f690c7edac498a174acc50","cross_cats_sorted":["cs.CL","cs.PL","stat.CO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-05T17:55:05Z","title_canon_sha256":"5d89a7e24576605c4bceb8c68f377e77e54070d3625680c3cb2d636dace14c1c"},"schema_version":"1.0","source":{"id":"2306.03081","kind":"arxiv","version":2}},"canonical_sha256":"94624111efc02696be807e67c692734c2517c66604c851a9c69bd8c8c182238d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"94624111efc02696be807e67c692734c2517c66604c851a9c69bd8c8c182238d","first_computed_at":"2026-07-05T07:16:42.589018Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:16:42.589018Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Y7+bRFzzm0qkxpwtkWfeG6ufqHTZnjZu83bkvFANm6e9Ma7RBHho4ANnivKkUsGcxPk4ipB6RtUnsNhcuOVbBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:16:42.589457Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.03081","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca15f028835c450e6b19d05b0d74cec850208760d2295930315f02435d0ddb1e","sha256:1554410bbbdc6868a3d1d5cf6d792a511e8a47039ca47c9ecccb5a4f86a85055"],"state_sha256":"39afbf7c5ca69fb9b1c592cfcf673d447a7fbb4f11bad7f2db28cededdf2215c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KtHY8L8DI2t0ZJnOqMwKzuolilN9tT2dQ21jyHxz6x1Pg8H3q8+zduOd93nN2nJ9Sv2zkd52AQnGZLMSZLemCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T17:27:53.911681Z","bundle_sha256":"eb44de87b8c89fc33074a5bba1b153d2f2bf7c8ad08abf81acd12828b2226ce8"}}