{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5EGDK5I2ZIABKJZHAP7H3XNXYC","short_pith_number":"pith:5EGDK5I2","canonical_record":{"source":{"id":"2511.16084","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T06:19:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"af8e378f9091089da4c2bf8831cb0a2bbcd30e44b6fbb1fbecbfd613a5c376ba","abstract_canon_sha256":"384ee8eaa77e56b1163d5e5b4539d3b0991b4b16081787f5a0e0c8a36aabe43f"},"schema_version":"1.0"},"canonical_sha256":"e90c35751aca0015272703fe7dddb7c088809b246aa035e04f0439714202c9bc","source":{"kind":"arxiv","id":"2511.16084","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.16084","created_at":"2026-05-18T03:09:33Z"},{"alias_kind":"arxiv_version","alias_value":"2511.16084v2","created_at":"2026-05-18T03:09:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.16084","created_at":"2026-05-18T03:09:33Z"},{"alias_kind":"pith_short_12","alias_value":"5EGDK5I2ZIAB","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"5EGDK5I2ZIABKJZH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"5EGDK5I2","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5EGDK5I2ZIABKJZHAP7H3XNXYC","target":"record","payload":{"canonical_record":{"source":{"id":"2511.16084","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T06:19:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"af8e378f9091089da4c2bf8831cb0a2bbcd30e44b6fbb1fbecbfd613a5c376ba","abstract_canon_sha256":"384ee8eaa77e56b1163d5e5b4539d3b0991b4b16081787f5a0e0c8a36aabe43f"},"schema_version":"1.0"},"canonical_sha256":"e90c35751aca0015272703fe7dddb7c088809b246aa035e04f0439714202c9bc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:33.122095Z","signature_b64":"CLTP3tqKN7S3cuDt2qug0WhzdhDTlzXTDHPBU8fZ0uJndHEiBY+pRO15iuyxAjMqLi7mXWUBvqo9vZkZVkqcBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e90c35751aca0015272703fe7dddb7c088809b246aa035e04f0439714202c9bc","last_reissued_at":"2026-05-18T03:09:33.121420Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:33.121420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2511.16084","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":"2026-05-18T03:09:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kwBTNMM54kws6Za1eM3YruA5VMBFX00WDr/XAshFFI8hg7LeW5NtG8NJBzGZL94Zy1ffAE6NyLV/JAiPUUUDBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T04:35:08.091156Z"},"content_sha256":"9ab36db5589562c208111dc6ebbbbd95444179fe1cdddbf1c7c72cd4df25cd30","schema_version":"1.0","event_id":"sha256:9ab36db5589562c208111dc6ebbbbd95444179fe1cdddbf1c7c72cd4df25cd30"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5EGDK5I2ZIABKJZHAP7H3XNXYC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SpectralTrain: A Universal Framework for Hyperspectral Image Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"SpectralTrain speeds hyperspectral image model training by 2-7x using curriculum learning and PCA spectral reduction while keeping accuracy close to full training.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jiarui Zhao, Liping Yu, Meihua Zhou, Nan Wan, Ruiguo Hu, Wai Kin Fung, Wenzhuo Liu, Xinyu Tong","submitted_at":"2025-11-20T06:19:26Z","abstract_excerpt":"Hyperspectral image (HSI) classification typically involves large-scale data and computationally intensive training, which limits the practical deployment of deep learning models in real-world remote sensing tasks. This study introduces SpectralTrain, a universal, architecture-agnostic training framework that enhances learning efficiency by integrating curriculum learning (CL) with principal component analysis (PCA)-based spectral downsampling. By gradually introducing spectral complexity while preserving essential information, SpectralTrain enables efficient learning of spectral -- spatial pa"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SpectralTrain enables efficient learning of spectral-spatial patterns at significantly reduced computational costs, delivering consistent 2-7x training speedups with small-to-moderate accuracy deltas across backbones on three benchmark datasets.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That PCA-based spectral downsampling combined with a curriculum schedule preserves essential information for accurate classification while the gradual complexity increase reliably improves learning efficiency over standard training.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SpectralTrain is a universal training framework that combines curriculum learning and PCA spectral downsampling to deliver 2-7x faster training for hyperspectral image classification across multiple backbones and datasets with only small accuracy trade-offs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SpectralTrain speeds hyperspectral image model training by 2-7x using curriculum learning and PCA spectral reduction while keeping accuracy close to full training.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2395eaf2bd3882fb8d58b84ec623bb455df2f9a2097cfd23640867cea4a64602"},"source":{"id":"2511.16084","kind":"arxiv","version":2},"verdict":{"id":"296e128e-55d8-4ce0-83ca-f7695384dfa9","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T20:54:41.007292Z","strongest_claim":"SpectralTrain enables efficient learning of spectral-spatial patterns at significantly reduced computational costs, delivering consistent 2-7x training speedups with small-to-moderate accuracy deltas across backbones on three benchmark datasets.","one_line_summary":"SpectralTrain is a universal training framework that combines curriculum learning and PCA spectral downsampling to deliver 2-7x faster training for hyperspectral image classification across multiple backbones and datasets with only small accuracy trade-offs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That PCA-based spectral downsampling combined with a curriculum schedule preserves essential information for accurate classification while the gradual complexity increase reliably improves learning efficiency over standard training.","pith_extraction_headline":"SpectralTrain speeds hyperspectral image model training by 2-7x using curriculum learning and PCA spectral reduction while keeping accuracy close to full training."},"references":{"count":48,"sample":[{"doi":"","year":2025,"title":"IEEE Transactions on Geoscience and Remote Sensing (2025)","work_id":"dec1a8b6-ac40-4bb3-a461-02dd57566c29","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"El-Gabri, A.R., Aly, H.A., Ghoniemy, T.S.,et al.: DLRA-Net: Deep local residual attention network with contextual refinement for spectral super- resolution. Int. J. Comput. Vis.133, 1499–1531 (2025) h","work_id":"da4cc966-6280-4ae9-83c2-a1d19ac85a18","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/j.rse.2007.07.028","year":2009,"title":"Recent advances in techniques for hyperspectral image processing","work_id":"bd6ef533-3819-4a2b-aaaa-47d7acfaacea","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2002,"title":"IEEE Signal processing magazine19(1), 17–28 (2002)","work_id":"4986113e-7b93-4f6a-ac2c-6fd824cb41bc","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"ISPRS Journal of Photogrammetry and Remote Sensing158, 279–317 (2019) 29","work_id":"cc14eb11-19fd-4888-93bb-e9bf5568cc48","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":48,"snapshot_sha256":"66dba3eec6d770b906da771d080d8aa7cccbe5407c99703b4b0cf516301f1384","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":"296e128e-55d8-4ce0-83ca-f7695384dfa9"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kl3AnPlj1BWJ4e4Kx09h9xrgYZq2rOhpNX5szObado8d0qlkS0zH2N1KbwZ5kGdQd4OGa2xArHf5T1oPkDQiBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T04:35:08.092307Z"},"content_sha256":"da783786a742cff2219b7ab073d11a7630ce4d79b1ce78ced7cea120c5c92bd3","schema_version":"1.0","event_id":"sha256:da783786a742cff2219b7ab073d11a7630ce4d79b1ce78ced7cea120c5c92bd3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC/bundle.json","state_url":"https://pith.science/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC/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":"2026-05-27T04:35:08Z","links":{"resolver":"https://pith.science/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC","bundle":"https://pith.science/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC/bundle.json","state":"https://pith.science/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5EGDK5I2ZIABKJZHAP7H3XNXYC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5EGDK5I2ZIABKJZHAP7H3XNXYC","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":"384ee8eaa77e56b1163d5e5b4539d3b0991b4b16081787f5a0e0c8a36aabe43f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T06:19:26Z","title_canon_sha256":"af8e378f9091089da4c2bf8831cb0a2bbcd30e44b6fbb1fbecbfd613a5c376ba"},"schema_version":"1.0","source":{"id":"2511.16084","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.16084","created_at":"2026-05-18T03:09:33Z"},{"alias_kind":"arxiv_version","alias_value":"2511.16084v2","created_at":"2026-05-18T03:09:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.16084","created_at":"2026-05-18T03:09:33Z"},{"alias_kind":"pith_short_12","alias_value":"5EGDK5I2ZIAB","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"5EGDK5I2ZIABKJZH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"5EGDK5I2","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:da783786a742cff2219b7ab073d11a7630ce4d79b1ce78ced7cea120c5c92bd3","target":"graph","created_at":"2026-05-18T03:09:33Z","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":"SpectralTrain enables efficient learning of spectral-spatial patterns at significantly reduced computational costs, delivering consistent 2-7x training speedups with small-to-moderate accuracy deltas across backbones on three benchmark datasets."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That PCA-based spectral downsampling combined with a curriculum schedule preserves essential information for accurate classification while the gradual complexity increase reliably improves learning efficiency over standard training."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SpectralTrain is a universal training framework that combines curriculum learning and PCA spectral downsampling to deliver 2-7x faster training for hyperspectral image classification across multiple backbones and datasets with only small accuracy trade-offs."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SpectralTrain speeds hyperspectral image model training by 2-7x using curriculum learning and PCA spectral reduction while keeping accuracy close to full training."}],"snapshot_sha256":"2395eaf2bd3882fb8d58b84ec623bb455df2f9a2097cfd23640867cea4a64602"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Hyperspectral image (HSI) classification typically involves large-scale data and computationally intensive training, which limits the practical deployment of deep learning models in real-world remote sensing tasks. This study introduces SpectralTrain, a universal, architecture-agnostic training framework that enhances learning efficiency by integrating curriculum learning (CL) with principal component analysis (PCA)-based spectral downsampling. By gradually introducing spectral complexity while preserving essential information, SpectralTrain enables efficient learning of spectral -- spatial pa","authors_text":"Jiarui Zhao, Liping Yu, Meihua Zhou, Nan Wan, Ruiguo Hu, Wai Kin Fung, Wenzhuo Liu, Xinyu Tong","cross_cats":["cs.AI"],"headline":"SpectralTrain speeds hyperspectral image model training by 2-7x using curriculum learning and PCA spectral reduction while keeping accuracy close to full training.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T06:19:26Z","title":"SpectralTrain: A Universal Framework for Hyperspectral Image Classification"},"references":{"count":48,"internal_anchors":0,"resolved_work":48,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"IEEE Transactions on Geoscience and Remote Sensing (2025)","work_id":"dec1a8b6-ac40-4bb3-a461-02dd57566c29","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"El-Gabri, A.R., Aly, H.A., Ghoniemy, T.S.,et al.: DLRA-Net: Deep local residual attention network with contextual refinement for spectral super- resolution. Int. J. Comput. Vis.133, 1499–1531 (2025) h","work_id":"da4cc966-6280-4ae9-83c2-a1d19ac85a18","year":2025},{"cited_arxiv_id":"","doi":"10.1016/j.rse.2007.07.028","is_internal_anchor":false,"ref_index":3,"title":"Recent advances in techniques for hyperspectral image processing","work_id":"bd6ef533-3819-4a2b-aaaa-47d7acfaacea","year":2009},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"IEEE Signal processing magazine19(1), 17–28 (2002)","work_id":"4986113e-7b93-4f6a-ac2c-6fd824cb41bc","year":2002},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"ISPRS Journal of Photogrammetry and Remote Sensing158, 279–317 (2019) 29","work_id":"cc14eb11-19fd-4888-93bb-e9bf5568cc48","year":2019}],"snapshot_sha256":"66dba3eec6d770b906da771d080d8aa7cccbe5407c99703b4b0cf516301f1384"},"source":{"id":"2511.16084","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-17T20:54:41.007292Z","id":"296e128e-55d8-4ce0-83ca-f7695384dfa9","model_set":{"reader":"grok-4.3"},"one_line_summary":"SpectralTrain is a universal training framework that combines curriculum learning and PCA spectral downsampling to deliver 2-7x faster training for hyperspectral image classification across multiple backbones and datasets with only small accuracy trade-offs.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SpectralTrain speeds hyperspectral image model training by 2-7x using curriculum learning and PCA spectral reduction while keeping accuracy close to full training.","strongest_claim":"SpectralTrain enables efficient learning of spectral-spatial patterns at significantly reduced computational costs, delivering consistent 2-7x training speedups with small-to-moderate accuracy deltas across backbones on three benchmark datasets.","weakest_assumption":"That PCA-based spectral downsampling combined with a curriculum schedule preserves essential information for accurate classification while the gradual complexity increase reliably improves learning efficiency over standard training."}},"verdict_id":"296e128e-55d8-4ce0-83ca-f7695384dfa9"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9ab36db5589562c208111dc6ebbbbd95444179fe1cdddbf1c7c72cd4df25cd30","target":"record","created_at":"2026-05-18T03:09:33Z","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":"384ee8eaa77e56b1163d5e5b4539d3b0991b4b16081787f5a0e0c8a36aabe43f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T06:19:26Z","title_canon_sha256":"af8e378f9091089da4c2bf8831cb0a2bbcd30e44b6fbb1fbecbfd613a5c376ba"},"schema_version":"1.0","source":{"id":"2511.16084","kind":"arxiv","version":2}},"canonical_sha256":"e90c35751aca0015272703fe7dddb7c088809b246aa035e04f0439714202c9bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e90c35751aca0015272703fe7dddb7c088809b246aa035e04f0439714202c9bc","first_computed_at":"2026-05-18T03:09:33.121420Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:33.121420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CLTP3tqKN7S3cuDt2qug0WhzdhDTlzXTDHPBU8fZ0uJndHEiBY+pRO15iuyxAjMqLi7mXWUBvqo9vZkZVkqcBw==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:33.122095Z","signed_message":"canonical_sha256_bytes"},"source_id":"2511.16084","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9ab36db5589562c208111dc6ebbbbd95444179fe1cdddbf1c7c72cd4df25cd30","sha256:da783786a742cff2219b7ab073d11a7630ce4d79b1ce78ced7cea120c5c92bd3"],"state_sha256":"96ea99d08fb480a415fd101a4b2847fea921d2deee096f81e0e95f701e88f49f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h9RG1iRFI3mvEqK/54MsQOv07L47bTeyF+pxeGdE494/+V63uHbqBakTGoR9gTCNzlZ6xvi5uldyqH4bfuJoAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T04:35:08.097275Z","bundle_sha256":"f5b4e525c82b86bf87d660202abf9ea9b138074847f17a13423457cf18977246"}}