{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:HHWZIQSQB6MOE6OZFS2454XJYT","short_pith_number":"pith:HHWZIQSQ","canonical_record":{"source":{"id":"2605.16535","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-15T18:31:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"017644cee8db5a533af7f202390dd33c98ce2a0b2f638357b509008e33405e45","abstract_canon_sha256":"032282df20ff9799a451519c37d190393962acec2c0132a83fb19f424d345a0a"},"schema_version":"1.0"},"canonical_sha256":"39ed9442500f98e279d92cb5cef2e9c4c032dba524ca5db986b296af79ca2dbc","source":{"kind":"arxiv","id":"2605.16535","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16535","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16535v1","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16535","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"pith_short_12","alias_value":"HHWZIQSQB6MO","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"pith_short_16","alias_value":"HHWZIQSQB6MOE6OZ","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"pith_short_8","alias_value":"HHWZIQSQ","created_at":"2026-05-20T00:02:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:HHWZIQSQB6MOE6OZFS2454XJYT","target":"record","payload":{"canonical_record":{"source":{"id":"2605.16535","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-15T18:31:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"017644cee8db5a533af7f202390dd33c98ce2a0b2f638357b509008e33405e45","abstract_canon_sha256":"032282df20ff9799a451519c37d190393962acec2c0132a83fb19f424d345a0a"},"schema_version":"1.0"},"canonical_sha256":"39ed9442500f98e279d92cb5cef2e9c4c032dba524ca5db986b296af79ca2dbc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:27.914166Z","signature_b64":"0VefIorlIEm0TTM2ET8DQHQHdejVozCCx/kdAfhyUP0dCb1Vsp2BZ+QHFBF5NfJjyxlkJokP14bDVq3RGL+lAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39ed9442500f98e279d92cb5cef2e9c4c032dba524ca5db986b296af79ca2dbc","last_reissued_at":"2026-05-20T00:02:27.913538Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:27.913538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.16535","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":"2026-05-20T00:02:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yK2V+CDNIazuUFPbWDDXhQtnEhLtLVSkymK4PBj0aW0xD3w7vcW0igKiRMjQysimhkhJuyBBHN3AglkUwbcmCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T02:12:26.996669Z"},"content_sha256":"d8545645fe36f4369a922758fff066190bfb65c71bb34145a8d6d90b3af8686f","schema_version":"1.0","event_id":"sha256:d8545645fe36f4369a922758fff066190bfb65c71bb34145a8d6d90b3af8686f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:HHWZIQSQB6MOE6OZFS2454XJYT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RAPT: Retrieval-Augmented Post-hoc Thresholding for Multi-Label Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"RAPT adapts label selection thresholds by retrieving similar past documents and aggregating their outcomes.","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Darren Nicol, Ikechukwu Nkisi-Orji, Lasal Jayawardena, Nirmalie Wiratunga","submitted_at":"2026-05-15T18:31:04Z","abstract_excerpt":"Industrial multi-label document understanding pipelines score candidate labels and threshold or rank them to form a label set per document. This early selection step directly affects the accuracy of downstream information extraction from the document, as well as the associated verification effort. In practice, OCR noise, label imbalance, instance-dependent label cardinality, and asymmetric error costs make global score thresholds brittle and hard to maintain as document formats evolve. We present RAPT, a deployment-oriented retrieval-augmented score thresholding wrapper, applied post-hoc to im"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"RAPT consistently outperformed global and label-wise static thresholding baselines. In the industrial setting, RAPT achieved its best predictive performance with metric learners, reaching 0.87 Macro-F1.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That documents retrieved by similarity on the classifier's representation space will have thresholding situations whose outcomes are relevant and transferable to the query document's optimal label set.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"RAPT improves multi-label label-set selection by retrieving similar documents and locally aggregating their thresholding outcomes to adapt per-instance cutoffs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"RAPT adapts label selection thresholds by retrieving similar past documents and aggregating their outcomes.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0d2ad2ee0559f378bf6a3ef42e7284911c93e87baf60a4bbb8311b40bfc0d58e"},"source":{"id":"2605.16535","kind":"arxiv","version":1},"verdict":{"id":"606c9b7c-ce8b-47b8-8327-adddbc3a364f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T21:28:08.621601Z","strongest_claim":"RAPT consistently outperformed global and label-wise static thresholding baselines. In the industrial setting, RAPT achieved its best predictive performance with metric learners, reaching 0.87 Macro-F1.","one_line_summary":"RAPT improves multi-label label-set selection by retrieving similar documents and locally aggregating their thresholding outcomes to adapt per-instance cutoffs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That documents retrieved by similarity on the classifier's representation space will have thresholding situations whose outcomes are relevant and transferable to the query document's optimal label set.","pith_extraction_headline":"RAPT adapts label selection thresholds by retrieving similar past documents and aggregating their outcomes."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16535/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:23.125152Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T21:41:16.886303Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:33:23.068859Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T19:21:56.925507Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"78e1c7d6fa907e59cc905c982723022c59d1638553542a3f7d7075e2d96bce96"},"references":{"count":37,"sample":[{"doi":"","year":1994,"title":"AI Communications7(1), 39–59 (1994)","work_id":"feb27468-8423-4dc6-955b-03c1366202d4","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Alsentzer, E., Murphy, J., Boag, W., et al.: Publicly Available Clinical BERT Em- beddings. In: Proc. of the 2nd Clinical Natural Language Processing Workshop. pp. 72–78. ACL (Jun 2019)","work_id":"002b4fd6-20f0-4cbe-9689-b7ac9bd33194","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"for Computational Linguistics","work_id":"6bda787a-093d-4bbc-a8e7-0e526d007f5b","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"In: Findings of the Association for Computational Linguistics: EMNLP 2020","work_id":"b8484d48-a0d3-4054-86f0-3f581f11418c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2006,"title":"Artificial Intelligence170(16), 1175–1192 (Nov 2006)","work_id":"edb455f8-9465-439d-8eb5-76c9ea79eb62","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":37,"snapshot_sha256":"d85641bb0dd1a8ed7c68498104497bde4b27497b059c11da3cfe97184d21d469","internal_anchors":5},"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":"606c9b7c-ce8b-47b8-8327-adddbc3a364f"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:02:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n3ZXpxLVqF9WEsRMexf+jQ8UDhTFWnOguZ6dAI445SjldWnGw8cV6ebOvnrqZg8a5D9AENSx0KTf+7xzXvzKBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T02:12:26.997461Z"},"content_sha256":"f97c83dfc67cb9a484a71ed5a213afedaafe75018709661df98f88d2fb9ff305","schema_version":"1.0","event_id":"sha256:f97c83dfc67cb9a484a71ed5a213afedaafe75018709661df98f88d2fb9ff305"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HHWZIQSQB6MOE6OZFS2454XJYT/bundle.json","state_url":"https://pith.science/pith/HHWZIQSQB6MOE6OZFS2454XJYT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HHWZIQSQB6MOE6OZFS2454XJYT/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":"2026-06-01T02:12:27Z","links":{"resolver":"https://pith.science/pith/HHWZIQSQB6MOE6OZFS2454XJYT","bundle":"https://pith.science/pith/HHWZIQSQB6MOE6OZFS2454XJYT/bundle.json","state":"https://pith.science/pith/HHWZIQSQB6MOE6OZFS2454XJYT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HHWZIQSQB6MOE6OZFS2454XJYT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HHWZIQSQB6MOE6OZFS2454XJYT","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":"032282df20ff9799a451519c37d190393962acec2c0132a83fb19f424d345a0a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-15T18:31:04Z","title_canon_sha256":"017644cee8db5a533af7f202390dd33c98ce2a0b2f638357b509008e33405e45"},"schema_version":"1.0","source":{"id":"2605.16535","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16535","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16535v1","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16535","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"pith_short_12","alias_value":"HHWZIQSQB6MO","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"pith_short_16","alias_value":"HHWZIQSQB6MOE6OZ","created_at":"2026-05-20T00:02:27Z"},{"alias_kind":"pith_short_8","alias_value":"HHWZIQSQ","created_at":"2026-05-20T00:02:27Z"}],"graph_snapshots":[{"event_id":"sha256:f97c83dfc67cb9a484a71ed5a213afedaafe75018709661df98f88d2fb9ff305","target":"graph","created_at":"2026-05-20T00:02:27Z","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":"RAPT consistently outperformed global and label-wise static thresholding baselines. In the industrial setting, RAPT achieved its best predictive performance with metric learners, reaching 0.87 Macro-F1."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That documents retrieved by similarity on the classifier's representation space will have thresholding situations whose outcomes are relevant and transferable to the query document's optimal label set."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"RAPT improves multi-label label-set selection by retrieving similar documents and locally aggregating their thresholding outcomes to adapt per-instance cutoffs."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"RAPT adapts label selection thresholds by retrieving similar past documents and aggregating their outcomes."}],"snapshot_sha256":"0d2ad2ee0559f378bf6a3ef42e7284911c93e87baf60a4bbb8311b40bfc0d58e"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:23.125152Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T21:41:16.886303Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T19:33:23.068859Z","status":"skipped","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T19:21:56.925507Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.16535/integrity.json","findings":[],"snapshot_sha256":"78e1c7d6fa907e59cc905c982723022c59d1638553542a3f7d7075e2d96bce96","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Industrial multi-label document understanding pipelines score candidate labels and threshold or rank them to form a label set per document. This early selection step directly affects the accuracy of downstream information extraction from the document, as well as the associated verification effort. In practice, OCR noise, label imbalance, instance-dependent label cardinality, and asymmetric error costs make global score thresholds brittle and hard to maintain as document formats evolve. We present RAPT, a deployment-oriented retrieval-augmented score thresholding wrapper, applied post-hoc to im","authors_text":"Darren Nicol, Ikechukwu Nkisi-Orji, Lasal Jayawardena, Nirmalie Wiratunga","cross_cats":["cs.AI"],"headline":"RAPT adapts label selection thresholds by retrieving similar past documents and aggregating their outcomes.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-15T18:31:04Z","title":"RAPT: Retrieval-Augmented Post-hoc Thresholding for Multi-Label Classification"},"references":{"count":37,"internal_anchors":5,"resolved_work":37,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"AI Communications7(1), 39–59 (1994)","work_id":"feb27468-8423-4dc6-955b-03c1366202d4","year":1994},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Alsentzer, E., Murphy, J., Boag, W., et al.: Publicly Available Clinical BERT Em- beddings. In: Proc. of the 2nd Clinical Natural Language Processing Workshop. pp. 72–78. ACL (Jun 2019)","work_id":"002b4fd6-20f0-4cbe-9689-b7ac9bd33194","year":2019},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"for Computational Linguistics","work_id":"6bda787a-093d-4bbc-a8e7-0e526d007f5b","year":2019},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"In: Findings of the Association for Computational Linguistics: EMNLP 2020","work_id":"b8484d48-a0d3-4054-86f0-3f581f11418c","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Artificial Intelligence170(16), 1175–1192 (Nov 2006)","work_id":"edb455f8-9465-439d-8eb5-76c9ea79eb62","year":2006}],"snapshot_sha256":"d85641bb0dd1a8ed7c68498104497bde4b27497b059c11da3cfe97184d21d469"},"source":{"id":"2605.16535","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T21:28:08.621601Z","id":"606c9b7c-ce8b-47b8-8327-adddbc3a364f","model_set":{"reader":"grok-4.3"},"one_line_summary":"RAPT improves multi-label label-set selection by retrieving similar documents and locally aggregating their thresholding outcomes to adapt per-instance cutoffs.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"RAPT adapts label selection thresholds by retrieving similar past documents and aggregating their outcomes.","strongest_claim":"RAPT consistently outperformed global and label-wise static thresholding baselines. In the industrial setting, RAPT achieved its best predictive performance with metric learners, reaching 0.87 Macro-F1.","weakest_assumption":"That documents retrieved by similarity on the classifier's representation space will have thresholding situations whose outcomes are relevant and transferable to the query document's optimal label set."}},"verdict_id":"606c9b7c-ce8b-47b8-8327-adddbc3a364f"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d8545645fe36f4369a922758fff066190bfb65c71bb34145a8d6d90b3af8686f","target":"record","created_at":"2026-05-20T00:02:27Z","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":"032282df20ff9799a451519c37d190393962acec2c0132a83fb19f424d345a0a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-15T18:31:04Z","title_canon_sha256":"017644cee8db5a533af7f202390dd33c98ce2a0b2f638357b509008e33405e45"},"schema_version":"1.0","source":{"id":"2605.16535","kind":"arxiv","version":1}},"canonical_sha256":"39ed9442500f98e279d92cb5cef2e9c4c032dba524ca5db986b296af79ca2dbc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"39ed9442500f98e279d92cb5cef2e9c4c032dba524ca5db986b296af79ca2dbc","first_computed_at":"2026-05-20T00:02:27.913538Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:02:27.913538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0VefIorlIEm0TTM2ET8DQHQHdejVozCCx/kdAfhyUP0dCb1Vsp2BZ+QHFBF5NfJjyxlkJokP14bDVq3RGL+lAg==","signature_status":"signed_v1","signed_at":"2026-05-20T00:02:27.914166Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16535","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d8545645fe36f4369a922758fff066190bfb65c71bb34145a8d6d90b3af8686f","sha256:f97c83dfc67cb9a484a71ed5a213afedaafe75018709661df98f88d2fb9ff305"],"state_sha256":"0f9893d6deb3414d1eb5291e642430c0545dab0d04fb23663791418758276544"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qUWUvFnUpxHmlVoh7MLnx3rQCVxZeRwMMbXRZhIoAeqsKB98DLLHQe2q/Or1UNy+ym4tuBJq8YA8iV4r4yskDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T02:12:27.006187Z","bundle_sha256":"f74a29c64a1b8347a022d4e44d9d1c679ca3abde1a7dfa1475c56d0107f7b639"}}