{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UEMSOXGLBETGM3AX56DLGIAZ5V","short_pith_number":"pith:UEMSOXGL","canonical_record":{"source":{"id":"2605.11130","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-11T18:38:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d52ad6c2e3076feefdc23a14b3097b6bd8a5b5756acb5992235707400b214783","abstract_canon_sha256":"50167656b4d48c5bb77df84e3654b05917354ccfa80f30e4300b7eb1a5740b50"},"schema_version":"1.0"},"canonical_sha256":"a119275ccb0926666c17ef86b32019ed506e10a20038b3f76050d50493ad3613","source":{"kind":"arxiv","id":"2605.11130","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11130","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11130v3","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11130","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"pith_short_12","alias_value":"UEMSOXGLBETG","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"pith_short_16","alias_value":"UEMSOXGLBETGM3AX","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"pith_short_8","alias_value":"UEMSOXGL","created_at":"2026-05-26T02:05:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UEMSOXGLBETGM3AX56DLGIAZ5V","target":"record","payload":{"canonical_record":{"source":{"id":"2605.11130","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-11T18:38:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d52ad6c2e3076feefdc23a14b3097b6bd8a5b5756acb5992235707400b214783","abstract_canon_sha256":"50167656b4d48c5bb77df84e3654b05917354ccfa80f30e4300b7eb1a5740b50"},"schema_version":"1.0"},"canonical_sha256":"a119275ccb0926666c17ef86b32019ed506e10a20038b3f76050d50493ad3613","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:05:10.421108Z","signature_b64":"v4tAWnslD085Wj167yDQbWXo4ouw1zQsMO06FpvXMFXajhtOjJVow2xjBiFFW8iQ5n/rTBDV7/OyOJc4qcmxAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a119275ccb0926666c17ef86b32019ed506e10a20038b3f76050d50493ad3613","last_reissued_at":"2026-05-26T02:05:10.420358Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:05:10.420358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.11130","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-26T02:05:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IB6GAyjRRfPHvyUe93qWd6t2io5GXHaQQp4rkW33gyhvxgTwdym5tqzs9P86KTUdSJUN4V6eemPleeUM9hMxAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T16:15:23.791621Z"},"content_sha256":"2c6e23c2d467cf6a2a93d0b8abb204de4db6f9914443ba5c3a5783ec2cb6ecef","schema_version":"1.0","event_id":"sha256:2c6e23c2d467cf6a2a93d0b8abb204de4db6f9914443ba5c3a5783ec2cb6ecef"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UEMSOXGLBETGM3AX56DLGIAZ5V","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HEPA: A Self-Supervised Horizon-Conditioned Event Predictive Architecture for Time Series","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"HEPA pretrains a causal Transformer on unlabeled time series by forecasting future representations at chosen horizons, then freezes the encoder to output accurate survival CDFs for rare events with far less labeled data.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"(2) Forgis, (3) University of Vienna), Camilla Mazzoleni, Federico Martelli, Gian-Alessandro Lombardi, Jonas Petersen, Philipp Petersen, Philipp Petersen (3) ((1) ETH Zurich, Riccardo Maggioni","submitted_at":"2026-05-11T18:38:03Z","abstract_excerpt":"Critical events in multivariate time series, from turbine failures to cardiac arrhythmias, demand accurate prediction, yet labeled data is scarce because such events are rare and costly to annotate. We introduce HEPA (Horizon-conditioned Event Predictive Architecture), built on two key principles. First, a causal Transformer encoder is pretrained via a Joint-Embedding Predictive Architecture (JEPA): a horizon-conditioned predictor learns to forecast future representations rather than future values, forcing the encoder to capture predictable temporal dynamics from unlabeled data alone. Second, "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"With fixed architecture and optimiser hyperparameters across all benchmarks, HEPA handles water contamination, cyberattack detection, volatility regimes, and eight further event types across 11 domains, exceeding leading time-series architectures including PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks, with an order of magnitude fewer tuned parameters and, on lifecycle datasets, an order of magnitude less labeled data.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the representations learned by horizon-conditioned JEPA pretraining on unlabeled data will transfer effectively to accurate event-specific survival CDF prediction after freezing the encoder, without requiring domain-specific architectural changes or extensive hyperparameter search.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"HEPA combines self-supervised JEPA pretraining on time series representations with horizon-conditioned finetuning to predict rare events via survival CDFs, outperforming PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks while using an order of magnitude fewer tuned params.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"HEPA pretrains a causal Transformer on unlabeled time series by forecasting future representations at chosen horizons, then freezes the encoder to output accurate survival CDFs for rare events with far less labeled data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3669115722eca7bfcc326df9c422a3f0993d9c4b1dd04c1014038ce84b896094"},"source":{"id":"2605.11130","kind":"arxiv","version":3},"verdict":{"id":"04d8a7a5-812d-44c9-85b4-990850a86f00","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:03:06.915737Z","strongest_claim":"With fixed architecture and optimiser hyperparameters across all benchmarks, HEPA handles water contamination, cyberattack detection, volatility regimes, and eight further event types across 11 domains, exceeding leading time-series architectures including PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks, with an order of magnitude fewer tuned parameters and, on lifecycle datasets, an order of magnitude less labeled data.","one_line_summary":"HEPA combines self-supervised JEPA pretraining on time series representations with horizon-conditioned finetuning to predict rare events via survival CDFs, outperforming PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks while using an order of magnitude fewer tuned params.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the representations learned by horizon-conditioned JEPA pretraining on unlabeled data will transfer effectively to accurate event-specific survival CDF prediction after freezing the encoder, without requiring domain-specific architectural changes or extensive hyperparameter search.","pith_extraction_headline":"HEPA pretrains a causal Transformer on unlabeled time series by forecasting future representations at chosen horizons, then freezes the encoder to output accurate survival CDFs for rare events with far less labeled data."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.11130/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T05:02:00.735626Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:34:52.886396Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T10:31:16.757392Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:45:26.894460Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"6069bace8efe3523cd69ca743ac11ee4aa0963eeec34cc5715d909597839b1f7"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7c74bd3e8ddd8de76eab779f946ae5e59ee43ba163aafdf52141f60ac049bc4f"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"04d8a7a5-812d-44c9-85b4-990850a86f00"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-26T02:05:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4NANVBu0l3S8L80W5puGSqnhjVcL8ECp0n4mDhfU1SIFSCgSA25Sfu+0JqjKh7edgjMvioPE6ggRi6ahVMyxCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T16:15:23.792476Z"},"content_sha256":"727da9ba1a3f7e2e390ba86c2d97a320282ed4954a6545e3ca18ff1becc4ce36","schema_version":"1.0","event_id":"sha256:727da9ba1a3f7e2e390ba86c2d97a320282ed4954a6545e3ca18ff1becc4ce36"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UEMSOXGLBETGM3AX56DLGIAZ5V/bundle.json","state_url":"https://pith.science/pith/UEMSOXGLBETGM3AX56DLGIAZ5V/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UEMSOXGLBETGM3AX56DLGIAZ5V/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":"2026-06-07T16:15:23Z","links":{"resolver":"https://pith.science/pith/UEMSOXGLBETGM3AX56DLGIAZ5V","bundle":"https://pith.science/pith/UEMSOXGLBETGM3AX56DLGIAZ5V/bundle.json","state":"https://pith.science/pith/UEMSOXGLBETGM3AX56DLGIAZ5V/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UEMSOXGLBETGM3AX56DLGIAZ5V/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UEMSOXGLBETGM3AX56DLGIAZ5V","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":"50167656b4d48c5bb77df84e3654b05917354ccfa80f30e4300b7eb1a5740b50","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-11T18:38:03Z","title_canon_sha256":"d52ad6c2e3076feefdc23a14b3097b6bd8a5b5756acb5992235707400b214783"},"schema_version":"1.0","source":{"id":"2605.11130","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11130","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11130v3","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11130","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"pith_short_12","alias_value":"UEMSOXGLBETG","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"pith_short_16","alias_value":"UEMSOXGLBETGM3AX","created_at":"2026-05-26T02:05:10Z"},{"alias_kind":"pith_short_8","alias_value":"UEMSOXGL","created_at":"2026-05-26T02:05:10Z"}],"graph_snapshots":[{"event_id":"sha256:727da9ba1a3f7e2e390ba86c2d97a320282ed4954a6545e3ca18ff1becc4ce36","target":"graph","created_at":"2026-05-26T02:05:10Z","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":"With fixed architecture and optimiser hyperparameters across all benchmarks, HEPA handles water contamination, cyberattack detection, volatility regimes, and eight further event types across 11 domains, exceeding leading time-series architectures including PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks, with an order of magnitude fewer tuned parameters and, on lifecycle datasets, an order of magnitude less labeled data."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the representations learned by horizon-conditioned JEPA pretraining on unlabeled data will transfer effectively to accurate event-specific survival CDF prediction after freezing the encoder, without requiring domain-specific architectural changes or extensive hyperparameter search."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"HEPA combines self-supervised JEPA pretraining on time series representations with horizon-conditioned finetuning to predict rare events via survival CDFs, outperforming PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks while using an order of magnitude fewer tuned params."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"HEPA pretrains a causal Transformer on unlabeled time series by forecasting future representations at chosen horizons, then freezes the encoder to output accurate survival CDFs for rare events with far less labeled data."}],"snapshot_sha256":"3669115722eca7bfcc326df9c422a3f0993d9c4b1dd04c1014038ce84b896094"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7c74bd3e8ddd8de76eab779f946ae5e59ee43ba163aafdf52141f60ac049bc4f"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T05:02:00.735626Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:34:52.886396Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T10:31:16.757392Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:45:26.894460Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.11130/integrity.json","findings":[],"snapshot_sha256":"6069bace8efe3523cd69ca743ac11ee4aa0963eeec34cc5715d909597839b1f7","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Critical events in multivariate time series, from turbine failures to cardiac arrhythmias, demand accurate prediction, yet labeled data is scarce because such events are rare and costly to annotate. We introduce HEPA (Horizon-conditioned Event Predictive Architecture), built on two key principles. First, a causal Transformer encoder is pretrained via a Joint-Embedding Predictive Architecture (JEPA): a horizon-conditioned predictor learns to forecast future representations rather than future values, forcing the encoder to capture predictable temporal dynamics from unlabeled data alone. Second, ","authors_text":"(2) Forgis, (3) University of Vienna), Camilla Mazzoleni, Federico Martelli, Gian-Alessandro Lombardi, Jonas Petersen, Philipp Petersen, Philipp Petersen (3) ((1) ETH Zurich, Riccardo Maggioni","cross_cats":["cs.AI"],"headline":"HEPA pretrains a causal Transformer on unlabeled time series by forecasting future representations at chosen horizons, then freezes the encoder to output accurate survival CDFs for rare events with far less labeled data.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-11T18:38:03Z","title":"HEPA: A Self-Supervised Horizon-Conditioned Event Predictive Architecture for Time Series"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.11130","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-14T21:03:06.915737Z","id":"04d8a7a5-812d-44c9-85b4-990850a86f00","model_set":{"reader":"grok-4.3"},"one_line_summary":"HEPA combines self-supervised JEPA pretraining on time series representations with horizon-conditioned finetuning to predict rare events via survival CDFs, outperforming PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks while using an order of magnitude fewer tuned params.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"HEPA pretrains a causal Transformer on unlabeled time series by forecasting future representations at chosen horizons, then freezes the encoder to output accurate survival CDFs for rare events with far less labeled data.","strongest_claim":"With fixed architecture and optimiser hyperparameters across all benchmarks, HEPA handles water contamination, cyberattack detection, volatility regimes, and eight further event types across 11 domains, exceeding leading time-series architectures including PatchTST, iTransformer, MAE, and Chronos-2 on at least 10 of 14 benchmarks, with an order of magnitude fewer tuned parameters and, on lifecycle datasets, an order of magnitude less labeled data.","weakest_assumption":"That the representations learned by horizon-conditioned JEPA pretraining on unlabeled data will transfer effectively to accurate event-specific survival CDF prediction after freezing the encoder, without requiring domain-specific architectural changes or extensive hyperparameter search."}},"verdict_id":"04d8a7a5-812d-44c9-85b4-990850a86f00"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2c6e23c2d467cf6a2a93d0b8abb204de4db6f9914443ba5c3a5783ec2cb6ecef","target":"record","created_at":"2026-05-26T02:05:10Z","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":"50167656b4d48c5bb77df84e3654b05917354ccfa80f30e4300b7eb1a5740b50","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-11T18:38:03Z","title_canon_sha256":"d52ad6c2e3076feefdc23a14b3097b6bd8a5b5756acb5992235707400b214783"},"schema_version":"1.0","source":{"id":"2605.11130","kind":"arxiv","version":3}},"canonical_sha256":"a119275ccb0926666c17ef86b32019ed506e10a20038b3f76050d50493ad3613","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a119275ccb0926666c17ef86b32019ed506e10a20038b3f76050d50493ad3613","first_computed_at":"2026-05-26T02:05:10.420358Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T02:05:10.420358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"v4tAWnslD085Wj167yDQbWXo4ouw1zQsMO06FpvXMFXajhtOjJVow2xjBiFFW8iQ5n/rTBDV7/OyOJc4qcmxAA==","signature_status":"signed_v1","signed_at":"2026-05-26T02:05:10.421108Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.11130","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2c6e23c2d467cf6a2a93d0b8abb204de4db6f9914443ba5c3a5783ec2cb6ecef","sha256:727da9ba1a3f7e2e390ba86c2d97a320282ed4954a6545e3ca18ff1becc4ce36"],"state_sha256":"a9a8ce87c2327d1065cae6d31b8c3696fd833ed1aac80be41ebc1a68b6b40d50"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VE810p3teK9ympWgRpK3yyXZP9j5yk1bVhaTFQkcvWxyaElml3AD1BFWfZf36/uTN5p/h98lMPOlye4z/reJCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T16:15:23.796510Z","bundle_sha256":"dec6bf67882a647a842daaa10e44cee759b31ad85975c4c06ff2577650e77b83"}}