{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ATDEY3YCGMFYPYOEOECMIYKX3F","short_pith_number":"pith:ATDEY3YC","canonical_record":{"source":{"id":"2605.15308","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T18:21:08Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"6046e3748de2f3baff5ad298678eaea239cb760b8353a337b2700668eed0b45b","abstract_canon_sha256":"d81474f8a9dfaa4756b9375ca1905419eda7d185ebfbe63b5910c9c0e0f6e075"},"schema_version":"1.0"},"canonical_sha256":"04c64c6f02330b87e1c47104c46157d94267afd6f1c37b9213108c328df84fc9","source":{"kind":"arxiv","id":"2605.15308","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15308","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15308v1","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15308","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"pith_short_12","alias_value":"ATDEY3YCGMFY","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"pith_short_16","alias_value":"ATDEY3YCGMFYPYOE","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"pith_short_8","alias_value":"ATDEY3YC","created_at":"2026-05-20T00:00:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ATDEY3YCGMFYPYOEOECMIYKX3F","target":"record","payload":{"canonical_record":{"source":{"id":"2605.15308","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T18:21:08Z","cross_cats_sorted":["cs.LG","cs.MA"],"title_canon_sha256":"6046e3748de2f3baff5ad298678eaea239cb760b8353a337b2700668eed0b45b","abstract_canon_sha256":"d81474f8a9dfaa4756b9375ca1905419eda7d185ebfbe63b5910c9c0e0f6e075"},"schema_version":"1.0"},"canonical_sha256":"04c64c6f02330b87e1c47104c46157d94267afd6f1c37b9213108c328df84fc9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:51.858248Z","signature_b64":"qlhFKToAPPtbjF6JosJJCBp/H6mGzmSI8+lUCJvXzDyXNY1Yfgr58aXSuKzUKv3V4ZMa6qdkyyAN0iRWRe2pDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"04c64c6f02330b87e1c47104c46157d94267afd6f1c37b9213108c328df84fc9","last_reissued_at":"2026-05-20T00:00:51.857467Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:51.857467Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.15308","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:00:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eUMU/UZRH/puL3fSCh9vDpVY4JjFjY1TLpnCj5gdFYyY1pt/Kv/aigLVBc66hdLS1CkVlDzplPn+BE4hTDyfDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T18:30:23.581240Z"},"content_sha256":"6feee9ad6ca549abea705104f5769a8c26c85b394342879052edbeabead8ca68","schema_version":"1.0","event_id":"sha256:6feee9ad6ca549abea705104f5769a8c26c85b394342879052edbeabead8ca68"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ATDEY3YCGMFYPYOEOECMIYKX3F","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SMCEvolve: Principled Scientific Discovery via Sequential Monte Carlo Evolution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo sampler.","cross_cats":["cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Huminhao Zhu, Jiachen Jiang, Zhihui Zhu","submitted_at":"2026-05-14T18:21:08Z","abstract_excerpt":"LLM-driven program evolution has emerged as a powerful tool for automated scientific discovery, yet existing frameworks offer no principled guide for designing their individual components and provide no guarantee that the search converges. We introduce SMCEvolve, which recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo (SMC) sampler. From this view, three core mechanisms emerge as principled components: adaptive parent resampling, mixture of mutation with acceptance, and automatic convergence control. We further provide"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo (SMC) sampler, supplying finite-sample complexity analysis that bounds the LLM-call budget required to reach a target approximation error.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the chosen mutation operators and acceptance probabilities, when combined inside the SMC framework, produce unbiased samples from the intended reward-tilted distribution without requiring additional corrections or domain-specific tuning.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SMCEvolve applies Sequential Monte Carlo sampling to LLM program search with adaptive resampling, mutation mixtures, and convergence control, delivering finite-sample complexity bounds and benchmark gains over prior systems.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo sampler.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"354de964fb276aa58486336512026368e31d5b74157deaba2a334effebbe9184"},"source":{"id":"2605.15308","kind":"arxiv","version":1},"verdict":{"id":"e297ad28-5018-45ef-abce-479a2f342801","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T16:17:05.087504Z","strongest_claim":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo (SMC) sampler, supplying finite-sample complexity analysis that bounds the LLM-call budget required to reach a target approximation error.","one_line_summary":"SMCEvolve applies Sequential Monte Carlo sampling to LLM program search with adaptive resampling, mutation mixtures, and convergence control, delivering finite-sample complexity bounds and benchmark gains over prior systems.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the chosen mutation operators and acceptance probabilities, when combined inside the SMC framework, produce unbiased samples from the intended reward-tilted distribution without requiring additional corrections or domain-specific tuning.","pith_extraction_headline":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo sampler."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15308/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.342482Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T16:27:02.237233Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T14:41:54.222503Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.776075Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"b813e36ff8d4266e590972ec4b2a7e956f1afc212ddf33661647bd66edcf6ae2"},"references":{"count":36,"sample":[{"doi":"","year":2024,"title":"Reevo: Large language models as hyper-heuristics with reflective evolution,","work_id":"472bbb13-4c89-444c-a9b6-d1de924b83a5","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Evolution of heuristics: Towards efficient automatic algorithm design using large language model","work_id":"7725f0cb-c6dd-4c18-b729-bcc99edf14a0","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Mathematical discoveries from program search with large language models,","work_id":"66841237-63ba-4670-9d90-97b75482e63e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","work_id":"76a0f850-d490-4e4f-ab98-8d25df82cd23","ref_index":4,"cited_arxiv_id":"2506.13131","is_internal_anchor":true},{"doi":"","year":2025,"title":"ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution","work_id":"41f30c7f-ee82-49b7-9ce9-463d3124269c","ref_index":5,"cited_arxiv_id":"2509.19349","is_internal_anchor":true}],"resolved_work":36,"snapshot_sha256":"35fc9019c837369bf7581ddf9c88a890b758c0a1ab6f046a688af7d4c997d905","internal_anchors":5},"formal_canon":{"evidence_count":2,"snapshot_sha256":"42b009227eb31edf5cf11b5157971f158c38fc1e26f1a774b38732f87686e41d"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"e297ad28-5018-45ef-abce-479a2f342801"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:00:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gnqjD3h+LtgKW+wlEN/JVUQfWlnGnih1QenOWU35NwqvRbN1KGo40zIKh6LPuQ1e50wgrKXujq62buMOBjJXDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T18:30:23.582294Z"},"content_sha256":"7409c1059355a39b9feb473581e2a97b12b3c555aea63b53485d2b5885c9a001","schema_version":"1.0","event_id":"sha256:7409c1059355a39b9feb473581e2a97b12b3c555aea63b53485d2b5885c9a001"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ATDEY3YCGMFYPYOEOECMIYKX3F/bundle.json","state_url":"https://pith.science/pith/ATDEY3YCGMFYPYOEOECMIYKX3F/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ATDEY3YCGMFYPYOEOECMIYKX3F/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":"2026-05-27T18:30:23Z","links":{"resolver":"https://pith.science/pith/ATDEY3YCGMFYPYOEOECMIYKX3F","bundle":"https://pith.science/pith/ATDEY3YCGMFYPYOEOECMIYKX3F/bundle.json","state":"https://pith.science/pith/ATDEY3YCGMFYPYOEOECMIYKX3F/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ATDEY3YCGMFYPYOEOECMIYKX3F/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ATDEY3YCGMFYPYOEOECMIYKX3F","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":"d81474f8a9dfaa4756b9375ca1905419eda7d185ebfbe63b5910c9c0e0f6e075","cross_cats_sorted":["cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T18:21:08Z","title_canon_sha256":"6046e3748de2f3baff5ad298678eaea239cb760b8353a337b2700668eed0b45b"},"schema_version":"1.0","source":{"id":"2605.15308","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15308","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15308v1","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15308","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"pith_short_12","alias_value":"ATDEY3YCGMFY","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"pith_short_16","alias_value":"ATDEY3YCGMFYPYOE","created_at":"2026-05-20T00:00:51Z"},{"alias_kind":"pith_short_8","alias_value":"ATDEY3YC","created_at":"2026-05-20T00:00:51Z"}],"graph_snapshots":[{"event_id":"sha256:7409c1059355a39b9feb473581e2a97b12b3c555aea63b53485d2b5885c9a001","target":"graph","created_at":"2026-05-20T00:00:51Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo (SMC) sampler, supplying finite-sample complexity analysis that bounds the LLM-call budget required to reach a target approximation error."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the chosen mutation operators and acceptance probabilities, when combined inside the SMC framework, produce unbiased samples from the intended reward-tilted distribution without requiring additional corrections or domain-specific tuning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SMCEvolve applies Sequential Monte Carlo sampling to LLM program search with adaptive resampling, mutation mixtures, and convergence control, delivering finite-sample complexity bounds and benchmark gains over prior systems."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo sampler."}],"snapshot_sha256":"354de964fb276aa58486336512026368e31d5b74157deaba2a334effebbe9184"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"42b009227eb31edf5cf11b5157971f158c38fc1e26f1a774b38732f87686e41d"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T16:31:18.342482Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:27:02.237233Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T14:41:54.222503Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.776075Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.15308/integrity.json","findings":[],"snapshot_sha256":"b813e36ff8d4266e590972ec4b2a7e956f1afc212ddf33661647bd66edcf6ae2","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM-driven program evolution has emerged as a powerful tool for automated scientific discovery, yet existing frameworks offer no principled guide for designing their individual components and provide no guarantee that the search converges. We introduce SMCEvolve, which recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo (SMC) sampler. From this view, three core mechanisms emerge as principled components: adaptive parent resampling, mixture of mutation with acceptance, and automatic convergence control. We further provide","authors_text":"Huminhao Zhu, Jiachen Jiang, Zhihui Zhu","cross_cats":["cs.LG","cs.MA"],"headline":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo sampler.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T18:21:08Z","title":"SMCEvolve: Principled Scientific Discovery via Sequential Monte Carlo Evolution"},"references":{"count":36,"internal_anchors":5,"resolved_work":36,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Reevo: Large language models as hyper-heuristics with reflective evolution,","work_id":"472bbb13-4c89-444c-a9b6-d1de924b83a5","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Evolution of heuristics: Towards efficient automatic algorithm design using large language model","work_id":"7725f0cb-c6dd-4c18-b729-bcc99edf14a0","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Mathematical discoveries from program search with large language models,","work_id":"66841237-63ba-4670-9d90-97b75482e63e","year":2024},{"cited_arxiv_id":"2506.13131","doi":"","is_internal_anchor":true,"ref_index":4,"title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","work_id":"76a0f850-d490-4e4f-ab98-8d25df82cd23","year":2025},{"cited_arxiv_id":"2509.19349","doi":"","is_internal_anchor":true,"ref_index":5,"title":"ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution","work_id":"41f30c7f-ee82-49b7-9ce9-463d3124269c","year":2025}],"snapshot_sha256":"35fc9019c837369bf7581ddf9c88a890b758c0a1ab6f046a688af7d4c997d905"},"source":{"id":"2605.15308","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T16:17:05.087504Z","id":"e297ad28-5018-45ef-abce-479a2f342801","model_set":{"reader":"grok-4.3"},"one_line_summary":"SMCEvolve applies Sequential Monte Carlo sampling to LLM program search with adaptive resampling, mutation mixtures, and convergence control, delivering finite-sample complexity bounds and benchmark gains over prior systems.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo sampler.","strongest_claim":"SMCEvolve recasts program search as sampling from a reward-tilted target distribution and approximates it with a Sequential Monte Carlo (SMC) sampler, supplying finite-sample complexity analysis that bounds the LLM-call budget required to reach a target approximation error.","weakest_assumption":"That the chosen mutation operators and acceptance probabilities, when combined inside the SMC framework, produce unbiased samples from the intended reward-tilted distribution without requiring additional corrections or domain-specific tuning."}},"verdict_id":"e297ad28-5018-45ef-abce-479a2f342801"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6feee9ad6ca549abea705104f5769a8c26c85b394342879052edbeabead8ca68","target":"record","created_at":"2026-05-20T00:00:51Z","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":"d81474f8a9dfaa4756b9375ca1905419eda7d185ebfbe63b5910c9c0e0f6e075","cross_cats_sorted":["cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-14T18:21:08Z","title_canon_sha256":"6046e3748de2f3baff5ad298678eaea239cb760b8353a337b2700668eed0b45b"},"schema_version":"1.0","source":{"id":"2605.15308","kind":"arxiv","version":1}},"canonical_sha256":"04c64c6f02330b87e1c47104c46157d94267afd6f1c37b9213108c328df84fc9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"04c64c6f02330b87e1c47104c46157d94267afd6f1c37b9213108c328df84fc9","first_computed_at":"2026-05-20T00:00:51.857467Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:00:51.857467Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qlhFKToAPPtbjF6JosJJCBp/H6mGzmSI8+lUCJvXzDyXNY1Yfgr58aXSuKzUKv3V4ZMa6qdkyyAN0iRWRe2pDw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:00:51.858248Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15308","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6feee9ad6ca549abea705104f5769a8c26c85b394342879052edbeabead8ca68","sha256:7409c1059355a39b9feb473581e2a97b12b3c555aea63b53485d2b5885c9a001"],"state_sha256":"a89d3cdc91be49b491c654442c4c402fc6458ac87ea6879ba19d7ab4b09fce68"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nk7/J1R9gFlWQwC8XAEkx1B7MNpmMmIKOYEDUh607cAvVYOfXwnQRaxNhK7RP8mTuihsku+O5v74Ypk9sMkzDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T18:30:23.586925Z","bundle_sha256":"b5024db210ac8e8d9acf1729ba66027c8e21b45f1d66fac13cdd2007232db2de"}}