{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:Y5TDPHCKKOSTKG2V3ANKZCXYZL","short_pith_number":"pith:Y5TDPHCK","canonical_record":{"source":{"id":"2605.14048","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T19:09:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9181553884db85a017d2a03bab5f5d3494ce41ce77e22554e62d36254c0aeed6","abstract_canon_sha256":"c8d4b48c49e064c56a11f538ac0c49f35475acbe66b048cf9ec43878edc577dd"},"schema_version":"1.0"},"canonical_sha256":"c766379c4a53a5351b55d81aac8af8cacd9f8791b41916dd9239cbc7c166e3b9","source":{"kind":"arxiv","id":"2605.14048","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14048","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14048v1","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14048","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"pith_short_12","alias_value":"Y5TDPHCKKOST","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"Y5TDPHCKKOSTKG2V","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"Y5TDPHCK","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:Y5TDPHCKKOSTKG2V3ANKZCXYZL","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14048","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T19:09:44Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9181553884db85a017d2a03bab5f5d3494ce41ce77e22554e62d36254c0aeed6","abstract_canon_sha256":"c8d4b48c49e064c56a11f538ac0c49f35475acbe66b048cf9ec43878edc577dd"},"schema_version":"1.0"},"canonical_sha256":"c766379c4a53a5351b55d81aac8af8cacd9f8791b41916dd9239cbc7c166e3b9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:12.673457Z","signature_b64":"6UEuLSjB6E9uN7IUj7dM+ZkbWVdEMWiK9nXh+stl/eRHe6s/PdMrtQP8g/O3q71l9tMYNz0iY5Ea3lgidMr0Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c766379c4a53a5351b55d81aac8af8cacd9f8791b41916dd9239cbc7c166e3b9","last_reissued_at":"2026-05-17T23:39:12.672806Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:12.672806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14048","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":"2026-05-17T23:39:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wonpUo249fNjV+lGyX5ozMM3rVv42O12hChCtqjU9j52yX84pWD/4+Mg+KBCBNdxEoquGz/uHD2bGk2UI4U/Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T23:36:08.228427Z"},"content_sha256":"188ba7441c885192d533d68349df671b8e885e833c2cc90fe400eb66cd97cfe5","schema_version":"1.0","event_id":"sha256:188ba7441c885192d533d68349df671b8e885e833c2cc90fe400eb66cd97cfe5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:Y5TDPHCKKOSTKG2V3ANKZCXYZL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Network-Aware Bilinear Tokenization for Brain Functional Connectivity Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Partitioning brain functional connectivity matrices into network-specific patches and embedding them via bilinear factorization produces more stable, transferable representations for predicting behavior and psychopathology.","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Bahram Jafrasteh, Leo Milecki, Mert R. Sabuncu, Qingyu Hu, Qingyu Zhao","submitted_at":"2026-05-13T19:09:44Z","abstract_excerpt":"Masked autoencoders (MAEs) have recently shown promise for self-supervised representation learning of resting-state brain functional connectivity (FC). However, a fundamental question remains unresolved: how should FC matrices be tokenized to align with the intrinsic modular organization of large-scale brain networks? Existing approaches typically adopt region-centric or graph-based schemes that treat FC as structurally homogeneous elements and overlook the large-scale network brain organization. We introduce NERVE (Network-Aware Representations of Brain Functional Connectivity via Bilinear To"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Compared to structurally agnostic MAE variants and graph-based self-supervised baselines, the proposed network-aware formulation yields more stable and transferable representations, particularly in cross-cohort evaluation.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That partitioning FC matrices into patches of intra- and inter-network connectivity blocks using an anatomically grounded parcellation aligns with the intrinsic modular organization and that this alignment improves representation learning for downstream prediction tasks.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NERVE introduces network-aware bilinear tokenization in masked autoencoders to learn transferable representations of brain functional connectivity by respecting large-scale network organization.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Partitioning brain functional connectivity matrices into network-specific patches and embedding them via bilinear factorization produces more stable, transferable representations for predicting behavior and psychopathology.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d14cb681bbec40a7ee8296487b1f1e24a29d37bfaaef2aa4e1c00dc3c0679934"},"source":{"id":"2605.14048","kind":"arxiv","version":1},"verdict":{"id":"8ab80604-51cd-4ee8-b7dc-e4754026d3b8","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T05:23:48.152636Z","strongest_claim":"Compared to structurally agnostic MAE variants and graph-based self-supervised baselines, the proposed network-aware formulation yields more stable and transferable representations, particularly in cross-cohort evaluation.","one_line_summary":"NERVE introduces network-aware bilinear tokenization in masked autoencoders to learn transferable representations of brain functional connectivity by respecting large-scale network organization.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That partitioning FC matrices into patches of intra- and inter-network connectivity blocks using an anatomically grounded parcellation aligns with the intrinsic modular organization and that this alignment improves representation learning for downstream prediction tasks.","pith_extraction_headline":"Partitioning brain functional connectivity matrices into network-specific patches and embedding them via bilinear factorization produces more stable, transferable representations for predicting behavior and psychopathology."},"references":{"count":33,"sample":[{"doi":"","year":1978,"title":"Psychological Bulletin85(6), 1275–1301 (1978)","work_id":"7bcfaa02-c61f-4f21-8c21-00faeb73d168","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2013,"title":"Button, K.S., Ioannidis, J.P., Mokrysz, C., Nosek, B.A., Flint, J., Robinson, E.S., et al.: Power failure: why small sample size undermines the reliability of neuroscience. Nat. Rev. Neurosci.14(5), 3","work_id":"5efb0db6-13b4-4522-bf4e-789ee00e863b","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Caro, J.O., Fonseca, A.H.d.O., Averill, C., Rizvi, S.A., Rosati, M., Cross, J.L., et al.: BrainLM: A foundation model for brain activity recordings. In: ICLR. Curran Associates, Inc. (2024)","work_id":"1387c9af-d63d-416e-b975-6c512d55f5e0","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Dong, Z., Li, R., Wu, Y., Nguyen, T., Su, J., Chong, et al.: Brain-JEPA: Brain Dy- namics Foundation Model with Gradient Positioning and Spatiotemporal Masking. In: NeurIPS. vol. 37, pp. 86048–86073. ","work_id":"ce87441c-c2f1-42c8-809e-d68ead42b5a7","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Frontiers in Neuroscience13(2019)","work_id":"2fccf567-2e6f-4429-9d60-98b169a43968","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":33,"snapshot_sha256":"23273e08a784f85e597208f02a25600712ceddbfcfc016efae6e3c0861e39cb9","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a457f742f3d6685b662588de2aa02b14ecc585d0a6cdc353c735403edef221ff"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"8ab80604-51cd-4ee8-b7dc-e4754026d3b8"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2HAfsYGbCTpPWji5H4zBj1oaiEuZccUvjuTW1nlMdcjBZiM6dlpaVo1obdJQ1L5NeBrblhltftBWdJSyezN3Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T23:36:08.229732Z"},"content_sha256":"383a41ede1d8f51966caf281bac07d99582a22f713073a2e3e75591f521843b1","schema_version":"1.0","event_id":"sha256:383a41ede1d8f51966caf281bac07d99582a22f713073a2e3e75591f521843b1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL/bundle.json","state_url":"https://pith.science/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL/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":"2026-05-30T23:36:08Z","links":{"resolver":"https://pith.science/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL","bundle":"https://pith.science/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL/bundle.json","state":"https://pith.science/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y5TDPHCKKOSTKG2V3ANKZCXYZL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Y5TDPHCKKOSTKG2V3ANKZCXYZL","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":"c8d4b48c49e064c56a11f538ac0c49f35475acbe66b048cf9ec43878edc577dd","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T19:09:44Z","title_canon_sha256":"9181553884db85a017d2a03bab5f5d3494ce41ce77e22554e62d36254c0aeed6"},"schema_version":"1.0","source":{"id":"2605.14048","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14048","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14048v1","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14048","created_at":"2026-05-17T23:39:12Z"},{"alias_kind":"pith_short_12","alias_value":"Y5TDPHCKKOST","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"Y5TDPHCKKOSTKG2V","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"Y5TDPHCK","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:383a41ede1d8f51966caf281bac07d99582a22f713073a2e3e75591f521843b1","target":"graph","created_at":"2026-05-17T23:39:12Z","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":"Compared to structurally agnostic MAE variants and graph-based self-supervised baselines, the proposed network-aware formulation yields more stable and transferable representations, particularly in cross-cohort evaluation."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That partitioning FC matrices into patches of intra- and inter-network connectivity blocks using an anatomically grounded parcellation aligns with the intrinsic modular organization and that this alignment improves representation learning for downstream prediction tasks."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"NERVE introduces network-aware bilinear tokenization in masked autoencoders to learn transferable representations of brain functional connectivity by respecting large-scale network organization."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Partitioning brain functional connectivity matrices into network-specific patches and embedding them via bilinear factorization produces more stable, transferable representations for predicting behavior and psychopathology."}],"snapshot_sha256":"d14cb681bbec40a7ee8296487b1f1e24a29d37bfaaef2aa4e1c00dc3c0679934"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a457f742f3d6685b662588de2aa02b14ecc585d0a6cdc353c735403edef221ff"},"paper":{"abstract_excerpt":"Masked autoencoders (MAEs) have recently shown promise for self-supervised representation learning of resting-state brain functional connectivity (FC). However, a fundamental question remains unresolved: how should FC matrices be tokenized to align with the intrinsic modular organization of large-scale brain networks? Existing approaches typically adopt region-centric or graph-based schemes that treat FC as structurally homogeneous elements and overlook the large-scale network brain organization. We introduce NERVE (Network-Aware Representations of Brain Functional Connectivity via Bilinear To","authors_text":"Bahram Jafrasteh, Leo Milecki, Mert R. Sabuncu, Qingyu Hu, Qingyu Zhao","cross_cats":["cs.LG"],"headline":"Partitioning brain functional connectivity matrices into network-specific patches and embedding them via bilinear factorization produces more stable, transferable representations for predicting behavior and psychopathology.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T19:09:44Z","title":"Network-Aware Bilinear Tokenization for Brain Functional Connectivity Representation Learning"},"references":{"count":33,"internal_anchors":0,"resolved_work":33,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Psychological Bulletin85(6), 1275–1301 (1978)","work_id":"7bcfaa02-c61f-4f21-8c21-00faeb73d168","year":1978},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Button, K.S., Ioannidis, J.P., Mokrysz, C., Nosek, B.A., Flint, J., Robinson, E.S., et al.: Power failure: why small sample size undermines the reliability of neuroscience. Nat. Rev. Neurosci.14(5), 3","work_id":"5efb0db6-13b4-4522-bf4e-789ee00e863b","year":2013},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Caro, J.O., Fonseca, A.H.d.O., Averill, C., Rizvi, S.A., Rosati, M., Cross, J.L., et al.: BrainLM: A foundation model for brain activity recordings. In: ICLR. Curran Associates, Inc. (2024)","work_id":"1387c9af-d63d-416e-b975-6c512d55f5e0","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Dong, Z., Li, R., Wu, Y., Nguyen, T., Su, J., Chong, et al.: Brain-JEPA: Brain Dy- namics Foundation Model with Gradient Positioning and Spatiotemporal Masking. In: NeurIPS. vol. 37, pp. 86048–86073. ","work_id":"ce87441c-c2f1-42c8-809e-d68ead42b5a7","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Frontiers in Neuroscience13(2019)","work_id":"2fccf567-2e6f-4429-9d60-98b169a43968","year":2019}],"snapshot_sha256":"23273e08a784f85e597208f02a25600712ceddbfcfc016efae6e3c0861e39cb9"},"source":{"id":"2605.14048","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T05:23:48.152636Z","id":"8ab80604-51cd-4ee8-b7dc-e4754026d3b8","model_set":{"reader":"grok-4.3"},"one_line_summary":"NERVE introduces network-aware bilinear tokenization in masked autoencoders to learn transferable representations of brain functional connectivity by respecting large-scale network organization.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Partitioning brain functional connectivity matrices into network-specific patches and embedding them via bilinear factorization produces more stable, transferable representations for predicting behavior and psychopathology.","strongest_claim":"Compared to structurally agnostic MAE variants and graph-based self-supervised baselines, the proposed network-aware formulation yields more stable and transferable representations, particularly in cross-cohort evaluation.","weakest_assumption":"That partitioning FC matrices into patches of intra- and inter-network connectivity blocks using an anatomically grounded parcellation aligns with the intrinsic modular organization and that this alignment improves representation learning for downstream prediction tasks."}},"verdict_id":"8ab80604-51cd-4ee8-b7dc-e4754026d3b8"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:188ba7441c885192d533d68349df671b8e885e833c2cc90fe400eb66cd97cfe5","target":"record","created_at":"2026-05-17T23:39:12Z","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":"c8d4b48c49e064c56a11f538ac0c49f35475acbe66b048cf9ec43878edc577dd","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T19:09:44Z","title_canon_sha256":"9181553884db85a017d2a03bab5f5d3494ce41ce77e22554e62d36254c0aeed6"},"schema_version":"1.0","source":{"id":"2605.14048","kind":"arxiv","version":1}},"canonical_sha256":"c766379c4a53a5351b55d81aac8af8cacd9f8791b41916dd9239cbc7c166e3b9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c766379c4a53a5351b55d81aac8af8cacd9f8791b41916dd9239cbc7c166e3b9","first_computed_at":"2026-05-17T23:39:12.672806Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:12.672806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6UEuLSjB6E9uN7IUj7dM+ZkbWVdEMWiK9nXh+stl/eRHe6s/PdMrtQP8g/O3q71l9tMYNz0iY5Ea3lgidMr0Ag==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:12.673457Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14048","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:188ba7441c885192d533d68349df671b8e885e833c2cc90fe400eb66cd97cfe5","sha256:383a41ede1d8f51966caf281bac07d99582a22f713073a2e3e75591f521843b1"],"state_sha256":"8f2c9047afc2be8bcd063c1ad8c18ec05252d138c07365e0d6f262cee497ae6c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OzTakue1nxcRo89WmqOuxpl76NOVAFh/cRsMF3r68pi2EXjtez1ajWw2yxq2LFXI/l3hqmakPmRFEs/pHIwFCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T23:36:08.235793Z","bundle_sha256":"a38b886d67ed20d7b8b7018530cde22b3e924c864ddbe5bfc78a255821552282"}}